l". V ' ”.v' 1

I t‘ 3-; .Vf‘ ’. " t i i "’ f

‘3‘4 \tu‘g.~:‘_ .3. . . ,L, ., 4;. :1 t

l ”I It: 3” V.” '. ‘i

K - a - ' ‘ .

5 ‘ ~ .5 IF‘ 4’ '. . l ‘

E? 29-. i

a ., ‘; a a

. _ V'rJ‘ é

,“‘ :‘
.. ’7'"

This is to certify that the

thesis entitled
Lgomem. mum ragga-WI Mia Alfmmmw

Mi Tile Hem : Contain/aged m: ‘TH/LLE FEED‘UC
TECH/«it (poeg

presented by

Rubi J [RA/LEE» TOD Hoist?

has been accepted towards fulfillment
of the requirements for

warm; degree in Smooce

Large Animal Clinical
Sciences

Wu A. ma My
0

Date 5'4‘8

0—7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

MSU

LIBRARIES
m

\r

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

ESOPHAGEAL MUCOSAL RESECTION AND ANASTOMOSIS IN THE HORSE:
COMPARISON OF THREE FEEDING TECHNIQUES

BY

Rory James Todhunter

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements for
the degree of

MASTER OF SCIENCE

Department of Large Animal Clinical Sciences

1983

ABSTRACT

ESOPHAGEAL MUCOSAL RESECTION AND ANASTOMOSIS IN THE HORSE:
COMPARISON OF THREE FEEDING TECHNIQUES

BY

Rory James Todhunter

Successful surgical management of esophageal
stricture in horses is difficult and the surgical techniques
have been poorly studied. I investigated the effects of
feeding techniques on the healing process following a method
of resecting a strictured area of esophageal mucosa. Twenty
horses had a three centimeter circumferential segment of
cervical esophageal mucosa resected and anastomosed. Five
horses were fed a moistened pelleted diet postoperatively:
five were fed this diet in slurry form via nasogastric tube
and ten were fed the same diet via an esophagostomy tube
until the fourteenth postoperative day. The pelleted diet
was then fed until euthanasia on the sixtieth postoperative
day. Endoscopic, radiologic, histopathologic and
morphometric techniques and compliance measurements were
used to compare healing between the groups of horses. Data
were compared using both parametric and nonparametric tests.
Strictures developed in all horses. The best feeding

technique was nasogastric tube—feeding.

DEDICATION

To my family -

Pam. Jim, Brett, Lisa, Shauna, Paddy, Steve, Nick.

ii

ACKNOWLEDGEMENTS

Without the encouragement and advice of Drs. John
Stick, Chris Brown and Ed Robinson and the particular
expertise and excellence of Dr. Ron Slocombe, this work

would never have matured. Thank you.

iii

TABLE OF CONTENTS

Page
INTRODUCTION AND LITERATURE REVIEW ...................... 1

"Successful" Equine Esophageal Resections
and Arlastomoses 0.0000000000000000000000000000000000 1

Previous Research .................................... 2
Historical Review (Human Esophageal Surgery) ......... 4
Stricture Classification ............................. 7
Surgical Considerations .............................. 8
Extraoral Alimentation ............................... 13
Diagnostic Techniques ................................ 15
MATERIALS AND METHODS ................................... 18
Experimental Design .................................. 18
Surgery .............................................. l9
Placement of Feeding Tubes ........................... 22
Physical Examination ................................. 23
Endoscopic Examination ............................... 24
Radiographic Procedure ............................... 25
Necropsy ............................................. 28
1. Pressure-Volume Measurements ................... 29
2. Qualitative PathOIOgy .......................... 32
3. Quantitative Pathology ......................... 33

Statistical Analysis ................................. 36

iv

RESULTS

1. Clinical Data

Miscellaneous Complications ..

2. Radiology

3. Pressure-Volume Measurements
4. Qualitative Pathology .......

Esophageal Wall Cross-Sectional Dimensions

Image Analysis

DISCUSSION AND CONCLUSIONS
Clinical Observations
Surgical Technique and Related Complications

Intraluminal Tubes

Esophagostomy Complications ..

Pressure-Volume Measurements .

Qualitative Histopathology ...

Morphometrics
Conclusion
SUMMARY

LIST OF REFERENCES

Page
37
37
47
52
64
66
78
81
83
84
84
88
89
91
92
94
96
97

99

LIST OF TABLES

TABLE 1. Number of days between surgery and mucosal
dehiscence (MD), muscularis externa dehiscence (MED),
opening of approach incision (OA), drain removal (DR)
and death (D) in all groups of horses .................. 38

TABLE 2. Clinical data results of Group I horses

(n = 5) ................................................ 39

TABLE 3. Clinical data results of Group II and
and Group III horses 0.00.00.........OOIOOOOOOOOOOOOO... 40

TABLE 4. Horse, day of death postoperatively and
cause of death in Group III (esophagostomy tube-fed
horseS) ......OOOOOIOOOOOOOO...OI.........IOOOOOOOOOOOOO 46

TABLE 5. Absolute stricture diameter (§+SEM in cm)

for each group of horses at 15, 30, 45 aHd 60 days
postoperatively measured from barium positive

pressure esophagrams ................................... 53

TABLE 6. Percentage stricture diameter (absolute

stricture diameter expressed as a percentage of the

maximum cranial esophageal diameter) (§+SEM) at

15, 30, 45 and 60 days postoperatively,-measured from
barium positive pressure esophagrams ................... 54

TABLE 7. Upger and lower esophageal wall defect areas
(itSEM in cm ) for each group of horses at 15, 30,
45 and 60 days Postoperatively I O O O I O O O O O O O O I I O O O O O I O O O O 55

TABLE 8. Upper and lower esophageal wall de ect areas
for horses in Groups II and III (§:SEM in cm ) ......... 64

TABLE 9. Pressure (§+SEM in torr) and volume (ml)

during inflation and deflation of the esophagus.

These data were used to estimate the compliance

of the esophageal wall for all Groups I and

II horses. The pressures are a composite of the

cranial, surgical and caudal sites ...... .............. . 67

vi

Page

TABLE 10. Computer-calculated valuefixof A & B

(§+SEM) for the equation Y = A[1-10 3

which describes the single rising exponential
pressure-volume curve for Group I and II horses

combined. Three regions (cranial, surgical site

and caudal) were measured, under both inflation

(Infl.) and deflation (Defl.) .......................... 68

TABLE 11. Cross-sectional width of esophageal
tissue layers (itSEM in mm) for horses in
Groups I. II, and III O.......C.......OOOOCOOCOOOO...... 80

TABLE 12. The relative area (§+SEM) of selectively

stained elastin and reticulum fibers as a ratio

of background area, for submucosal and muscularis

externa in Groups II and III horses .................... 82

vii

LIST OF FIGURES

Page
FIGURE 1. A radiographic representation of esophagus
at the surgery site distended under barium positive
pressure 00.0.0000.........OOOOOO........OOOOOOOOOOOOOOO 27

FIGURE 2. Diagrammatic representation of esophageal
wall pressure-volume measurement ....................... 31

FIGURE 3. A cross-section of the esophageal wall
illustrating the various wall compartments which were
measured ...-......CCIOOOOOOOO..........OOOOOOOOOOOOOOOO 35

FIGURE 4. Endoscopic appearance of the esophageal

mucosa 3 days postoperatively (Horse 10, Group II)
demonstrating considerable mucosal separation along

the suture line, exposing the submucosa (arrows).

The photograph was taken when the horse dislodged

its‘ nasogastric tube .................................. 41

FIGURE 5. Endoscopic appearance at anastomotic

site showing strictured area, traction

diverticulum and a fistula (arrow) at the base

of the diverticulum on the 50th postoperative

day (Horse 6, Group II) ................................ 42

FIGURE 6. Barium positive pressure esophagram of

Horse 7 in Group II 30 days postoperatively,

illustrating a stricture site (3), traction

diverticulum (arrows) and sinus tract (t) .............. 43

FIGURE 7. Endoscopic appearance of the esophageal

mucosa 1 day postoperatively, from Horse 18 in

Group III, demonstrating mucosal swelling and

early appearance of submucosa at the suture line

(arrow), indicating tension on the anastomosis ......... 45

FIGURE 8. Barium positive pressure esophagram

illustrating a stricture (s), fistula (f) and
esophagostomy stoma (es) (Horse 16, Group III) ......... 48

viii

Page
FIGURE 9. Endoscopic appearance of the larynx of
one of the horses (Horse 15, Group III) showing left
laryngeal hemiplegia in which the left arytenoid
cartilage fails to abduct on inspiration ............... 50

FIGURE 10. Endoscopic appearance of the larynx and

caudal dorsal pharyngeal area showing left laryngeal
hemiplegia and occlusion of the pharyngeal lumen due

to pressure from a retropharyngeal swelling

(Horse 16, Group III) .................................. 51

FIGURE 11. Histogram illustrating average percentage
stricture diameter (vertical axis) of all Group II and
III horses with days postoperatively on the horizontal

aXis ......OOOOIOOOOOOO............OOOOOOOOOOOOOOOOIO... 57

FIGURE 12. Endoscopic appearance of the stricture
30 days postoperatively in Horse 12, Group III ......... 58

FIGURE 13. Endoscopic appearance of the stricture
in the same horse as Figure 12, 60 days post-
operatively ......OOOOOOOOOOOOOOOCCO......OOOOOOOOOOOOOO 59

FIGURE 14. Barium double contrast esophagram made
30 days postoperatively from the same horse as in
Figures 12 and 13 O...........OOOOOOOOOOOOOOOOOOOO0.0... 60

FIGURE 15. Barium double contrast esophagram made
60 days postoperatively from the same horse as in
Figures 12' 13' and 14 OOOOOOOOOOOOOOOOOOO00.0.0000...O. 61

FIGURE 16. Barium double contrast esophagram of
Horse 7, Group II, taken 60 days postoperatively.
Note the traction diverticulum (t) and fistula (f) ..... 62

FIGURE 17. Barium positive pressure esophagram of
Horse 12, Group III, taken 60 days postoperatively ..... 63

FIGURE 18. Histogram showing upper and lower wall
defect areas for all horses in Groups II and III ....... 65

FIGURE 19. Pressure-volume curve of the esophagus
representing the average inflation and deflation

limbs of all horses in Group II. The intraluminal

pressure is shown on the horizontal axis, and the

volume of saline administered is shown on the

vertical axis. Standard error bars are marked.

There was a significant difference between the

inflation and deflation limbs at the 0.05 level ........ 69

ix

Page

FIGURE 20. Histopathologic appearance of a

longitudinal cross-section of the esophageal wall

from Horse 2, Group I, 3 days postoperatively.

Note the underrunning of the submucosa and the

abscess formation in the same area (Gomori's

Trichrome stain, original magnification X2).

m = mucosa; em = submucosa; a = abscess;

me = muscularis externa ................................ 70

FIGURE 21. Submucosal inflammation 3 days post-
operatively in the same horse as in Figure 20.

Submucosal tissues contain extensive amounts of
proteinaceous edema fluid (ef) in which numerous

bacteria are present (arrows). The inflammatory

cells present are predominantly neutrophils (H&E

stain; original magnification X1000) ................... 71

FIGURE 22. Purulent myositis in the muscularis

externa of the same horse as in Figures 20 and 21.

3 days postoperatively. The muscle fibers are

\ degenerating, and fibers are widely separated by

edema fluid and neutrophils (Gomori's Trichrome

stain; original magnification X100) .................... 73

FIGURE 23. Histopathologic appearance of the

esophageal wall in Horse 16, Group III, in transverse
section, illustrating the dense fibrous tissue in

the area through the muscularis externa incision.

Note that the mucosa is completely healed (Gomori's
Trichrome stain; original magnification X2) ........... 74

FIGURE 24. Histopathologic appearance of the

esophageal wall in longitudinal section, from Horse 6,

Group II, showing thickening in the submucosa and
hypertrophy and hyperplasia of the mucosa at the

surgical site in the center. Normal esophageal

wall layers are present on each side (Gomori's

Trichrome stain; original magnification X1.5) .......... 75

FIGURE 25. Histopathologic appearance of the
esophageal wall in longitudinal section on
demonstrating the extreme left an area of ulceration
and inflammatory cell lining from Horse 1, Group I.
Adjacent to this area, the mucosa has regenerated
(arrow) but is abnormal. There is vacuolation of

the mucosa and incomplete keratinization. 0n the

right of the photomicrograph, the mucosa is keratinized
(magenta color), but the mucosa remains hyperplastic.
The submucosa is hypercellular (Gomori's Trichrome
stain; original magnification X20) ..................... 76

X

Page

FIGURE 26. A longitudinal cross-section across

the mucosal surgical site in Horse 10, Group II,

showing suture granulomata (arrows) and a sinus

tract (t). Epithelial regeneration around the

sinus tract is incomplete and adjacent epithelium

is hypertrophied (Gomori‘s Trichrome stain;

original magnification X2) ............................. 77

FIGURE 27. Histopathologic appearance of the

adventitia in Horse 16, Group III, 60 days post-

operatively (viewed under polarized light),

illustrating granulomatous inflammation. Numerous

giant cells containing fragments of refractile

material are present. The fibrillar material that

is partially refractile is characteristic of

collagen. (H & E stain; original magnification

X100) ................................................. 79

xi

INTRODUCTION AND LITERATURE REVIEW

"Successful" Equine Esophageal Resections and Anastomoses

 

In the "Auburn Veterinarian" in the winter of 1963.
Vaughn and Hoffer described a surgical procedure for the
removal of 3 cm of strictured esophagus in a weanling.1
Unfortunately, the filly died 16 days postoperatively due to
an acute hemorrhagic enteritis of undetermined etiology.
There was a small sinus communicating with the esophageal
lumen near the anastomotic site. In their closing paragraph
in that paper, the authors suggest that the method of
esophageal closure should be more thoroughly investigated in
its application to large animal esophageal problems. Prior
to that time only two reports on the treatment of esophageal

2,3

disease in horses could be found, and there have been

4-17

but a few since then. With particular reference to

the postoperative care of an esophageal resection and

anastomosis, as a treatment for esophageal stricture in

horses, I could find only 4 reports.1'10'16'l7

1

Vaughn

and Hoffer as mentioned, removed the nasogastric tube

on the 14th postoperative day and allowed the weanling to
drink water and milk. She was not allowed access to solid

food before she died. In 1964, Lowelo

resected 2.5 cm
of cervical esophagus in a mare, and maintained her on

1

2

pellets. She ate hay once and showed signs of "choke". No
radiographic assessment was made of the degree of post-
operative stricture. However, nasogastric tubes of 1.5 cm
and 2.0 cm outside diameters were passed without difficulty.
Suannl6 resected 2 cm of cervical esophagus in a 2-year-old
colt, which was maintained on a mixture of oats, oaten chaff
bran and fresh green feed. However, there was evidence of
coughing and regurgitation when fed lucerne hay. Derksen and

Stick17

resected a 2 cm section of abnormal esophageal
mucosa and submucosa in a 3-month-old filly. She was fed
through an esophagostomy tube for 30 days postoperatively,
and was eating hay by the time she was discharged 45 days
post—operatively. Endoscopically, the esophagus was normal
2 years later. Postoperative esophagrams were not
performed. Thus it appears that in these 4 case reports
there is no uniform method of evaluating the ”success“ of
the surgical procedure, and the best that could be hoped
for, is that the horses could be maintained on a pelleted
ration post- operatively, except in the latest report by

Derksen and Stick, in which the horse ate hay.17

Previous Research

 

There are no published controlled experiments
dealing with the surgical technique for, and postoperative
healing process, following esophageal resection and

anastomosis in the horse. The healing of longitudinal

3
incisions in the normal esophagus has been studied in
ponies.18'19'20 Using endoscopic, radiographic and
morphometric techniques, Stick showed the progression of
healing in sutured longitudinal esophageal wounds to be a
function of diet rather than surgical technique. All the
esophagotomies in the hay-fed ponies dehisced, while first

intention healing occurred in all esophagotomies when the

ponies were fed a soft diet.20

The most common indication for esophageal resection
is stricture removal. The causes of stricture formation
vary among species. In horses, prolonged esophageal mucosal
damage as a result of esophageal obstruction (”choke") most

often results in stricture formation. Choke in horses can

13

occur as a result of food impaction, orally administered

3,21

caustic or corrosive medicinal agents, trauma to the

10

neck, and as a sequela to eSOphageal resection and

11

anastomosis. External trauma can lead to stricture

also, due to inflammation and secondary fibrosis in the
muscularis externa or periesophageal tissues, but this

occurs less commonly than primary mucosal erosion. In

23

people, esophageal tumors and severe esophagitis due

24

to ingestion of caustic agents, are the common causes

of stricture formation. Many advances have been made in the
treatment of human esophageal disease since World War II.

Esophageal surgery in people is based on experimental

24-32 '33 34-49

animals, and since the dog, cat

pig

4

SO 51

monkey, and rabbit have been used as experimental
models for human esophageal reconstruction following
resection and anastomosis, a review of the human literature

is in order.

Historical Review (Human Esophageal Surgery)

 

In 1877, Czerny52 first successfully resected
cancer of the cervical esophagus in a person. Through his

work and that of Bilroth.53

new sophistication developed

in gastrointestinal surgical techniques. In the early
1900's, staged reconstructive procedures for corrosive
stricture and malignant lesions were performed. In 1909.
the pharynx, larynx and cervical esophagus was resected in a
patient.54 Four years later, Torek55 successfully

resected the thoracic esophagus, performing cervical
esophagostomy and gastrostomy. A successful one—stage,
transpleural esophageal resection and esophagogastrostomy
for carcinoma remained an unattainable goal until Oshawa's

56

report in 1933. In the United States, the first

successful esophagogastrectomy for cancer was performed in

57

the Lahey Clinic in 1937. In 1941, the first

successful closure of a fistula and primary esophageal
anastomosis in congenital esophageal atresia and tracheo-

esophageal fistula was performed.58

5
It was not until after World War II, that esophageal
surgeons became aware of the many complexities of esophageal
function, which are so often disturbed by disease and by

59 In 1945, the first esophageal

surgical intervention.
resections with esophagogastrostomy and the first high
thoracic gastric replacements were performed in
California.60 Extensive literature is available on the
subjects of primary esophageal repair, viable and nonuviable

61 and complete historical

62

esophageal replacements,
descriptions of esophageal surgery in people.
ESOphageal disease in people is often extensive and
the major postoperative complication is leakage of luminal
contents at the surgical site. Since both leakage and
anastomotic tension predispose to stricture formation
following esophageal resection, a number of viable tissue
grafts and prostheses have been used for esophageal
substitution, in order to reduce these complications.
Viable substitutes have included:61
A. Pedicled gastrointestinal segments of

63 64 65

ileocecum, colon and jejunum;

B. Stomach via esophagogastrostomy,66 reverse

gastric tube formation based on the left gastroepiploic

artery and vein67 and isoperistaltic gastric tubes;68
C. Gastric antrum,69 jejunum,70 and
colon63 using microvascular anastomosis techniques;

6
D. Free skin grafts using split thickness

71 63

grafts, and full thickness grafts; and

E. Musculocutaneous flaps.72

The advantage of using the stomach as a replacement
organ in people is that it has excellent blood supply,
mobility and length. The distance it has to be stretched in
a human neck and thorax is proportionately much less than in
a horse for example. In addition, due to a person's upright
posture, together with vagotomy and pyloroplasty, gastric
emptying is greatly facilitated. The use of the colon has
increased dramatically since the 1950's, with the advent of
antibiotics and improved preoperative bowel preparation.

Its vascularity is good and it is relatively resistant to
peptic ulceration. It can also be protected by vagotomy and
gastric drainage procedures. Gastric and colonic esophageal
replacements constitute the majority of substitutes today.

Non-viable biological tissue and tissue substitutes
have also been used to replace esophageal segments or
reinforce esophageal anastomoses in animals. These have

73

included: diaphragm grafts, autogenous pericardial

grafts,74’75 fascia latae and aortic homografts,7S

76 77

autogenous fibrous tissue tubes, and deviced rings.

Even synthetic adhesives have been used to strengthen

78

esophageal anastomoses. Although some success has

been attained with non-viable substitutes, graft necrosis,

.7

infection, scarring or leakage suggest the use of viable

tissue substitutes is preferable.79

Stricture Classification

 

Esophageal strictures in horses are commonly
localized. “Choke" is most commonly caused by foreign body
or food impaction, resulting in localized areas of mucosal
and in more severe cases, submucosal loss. External neck
trauma and esophageal rupture also predispose to stricture
development. Full thickness mucosal or submucosal
esophageal defects heal predominently by wound contraction
and fibrous protein synthesis.80 The resulting annular
lesion can be classified into three types depending on the
anatomic location of induration and fibrosis: 1) mural
lesions that involve only the adventitia and muscularis, 2)
esophageal rings or webs that involve only the mucosa-
submucosa, and 3) annular stenosis that involves all layers

81 The size of the horse does

of the esophageal wall.
not lend itself particularly to esophageal replacements by
colon and stomach, due to the prolonged anesthesia period
required, the increased incidence of infection in prolonged
surgical procedures, and the more localized nature of the
lesion, compared to humans. Strictures also occur more
commonly in the cervical area, where a one-stage

reconstructive procedure is preferable. Depending on the

nature of the lesion, strictures of the equine esophagus,

8
which are unresponsive to medical management may be
corrected by esophagomyotomy, complete resection and
anastomosis, partial resection, esophagoplasty, or patch

grafting.1o,12.ez

Surgical Considerations

 

Special consideration must be given to surgery of
the es0phagus.79 The esophagus lacks an outer serosal
surface which would help limit leakage by exuding fibrin.
On the mucosal surface, however, a fibrin seal does form at

the margin of the healing mucosal incision following

longitudinal esophagotomy.18 Due to the segmental
nature of the blood supply to the esophagus83 special
80

care should be taken to avoid ischemic necrosis. It
has been demonstrated, at least in dogs, however, that
considerable portions of both the cervical and thoracic
esophagus can be mobilized, without subsequent loss of

24.26.31 24 mobilized

viability. Parker and Brockington
the entire canine thoracic esophagus from its mediastinal
bed and resected about 33%, with a 33% mortality. No
intensive care or antibiotics were used. Shek et al26
in a two-part experiment using dogs, mobilized the thoracic
eSOphagus from the aortic arch to the diaphragm, and
resected 3-5 cm portions, followed by anastomosis. Fourteen
of 15 dogs in the first part of the experiment survived for

two months prior to euthanasia. No strictures were noted.

9
In the second part, the dogs' entire thoracic esophagi were
freed and all their thoracic blood supply were severed.
Five of the six animals had a viable esophagus 14 days post-
operatively, judged by arteriography and necropsy. Macmanus
et al, after performing five different operations on 46 dogs
concluded that necrosis of the dog’s esophagus does not
occur following complete devascularization (division and
anastomosis) of the thoracic portion. However, it did occur
at the thoracic inlet after cervical and thoracic
mobilization.

Constant motion of the esophagus, swallowing and
diaphragmatic and adjacent organ movement in the thorax,
contradict one of the basic principles of wound healing -
tissue rest. Additionally, the esophageal wall poorly
tolerates longitudinal stretching and tension.27'28'49
Consequently, much attention has been given to the reduction
of tension at the anastomotic site, with different types of
esophageal myotomy. A particular application of myotomy has
been to reduce tension on the esophageal anastomosis in
children undergoing primary one-stage repair for esophageal
atresia or tracheo-esophageal fistula.84’86 Approximately
1 cm in length can be gained with each myotomy.

Experimental work in both (309327'28 35'39'49

and pigs
has shown that circular or spiral esophagomyotomies can
significantly reduce the degree of anastomotic tension on

the eSOphagus. One circular myotomy in piglets can decrease

10
the approximating force of a 5-6 cm resection by 50% and a

49

double myotomy by 75%. Circular myotomy in the presence

of gross shortening of the esophagus is valuable in the

repair of hiatal hernia.87

Single and double myectomies
in piglets, in which 2-4 cm of esophageal muscularis externa
was removed, did not affect esophageal motility according
to contrast radiographic procedures.36 Thus, transecting
the muscularis externa does not appear to alter esophageal
function clinically or radiographically. No manometric
studies to investigate motility have been performed
following such surgical intervention in these pigs.

The esophagus must be sutured with care to maximize
the suture holding power of the mucosa and submucosa and
minimize the danger of suture interference with blood supply

and wound healing.80'88

Omentum is not available to help
seal or localize a leak in the esophagus. Single interrupted
sutures in the esophageal anastomosis cause less vascular
compromise and are generally recommended in the mucosal-
submucosal closure rather than a continuous pattern.30'89
However, continuous patterns in esophageal mucosal closure
have been shown to heal well.17'30 Theoretically a
continuous line should have less tendency to leak relative
to a single interrupted closure. However, continuous suture
patterns can create a purse-string effect and pull-through
13

the mucosa when intraluminal pressure increases.

Effective closure can also be obtained with an everting

11
single layer interrupted horizontal mattress pattern. The
sutures are placed through the full thickness of the
esophageal wall, and tied with firm, but not strangulating

13’90'91 Variable results have been obtained in

tension.
experiments to assess the difference in healing character—
istics following esophageal resection and anastomosis using

43’45'48'89 Several conclusions

different suture materials.
may be drawn. Less reactive suture materials, e.g.
monofilament polypropylenea and polyglycolic acidb
appear to result in less stricture formation than silk.
Additionally, the researchers say chromic catgut produces a
far more severe reaction than silk.92
Single and double layer closures will alter lumen
size with either silk or polyglycolic acid.b However,
in an experiment comparing one layer, single-interrupted
chromic catgut and non-absorbable Teflon-coated polyester
fiber in esophageal anastomoses without resection, there
were 50% fewer survivors When non-absorbable suture material
was used, rather than an absorbable one. With a 5 cm
esophageal resection, there were 66% fewer survivors,
following anastomosis with non-absorbable material than with
absorbable material. Deaths were attributed to anastomotic

leaks, which seemed related to loosening and cutting through

of the suture, rather than a defect in the anastomotic

 

aProlene, Ethicon, Inc., Somerville, NJ.

Dexon, Davis & Geck, American Cyanamid Co., Pearl River, NY.

12
technique. The explanation proposed was that perhaps catgut
swells, increases in diameter and therefore may move and

93 There does seem to be a correlation between

loosen less.
the technical difficulty of the procedure (i.e. working in
the thorax compared with the neck) and the degree of post-
operative stricture. More experienced surgeons have less
postoperative strictures than less experienced ones. There
is a very strong correlation between the degree of
postoperative stricture formation and the degree of mucosal
dehiscence.34'35'39'42 Consequently, strictures may
develop even without luminal leakage because of excessive
tension alone, which then leads to mucosal separation. Most
surgeons tie the suture knots in the lumen to reduce scar
production, and promote extrusion of the suture into the

44.91

lumen. Most surgeons recommend closure of the

esophageal muscle with an interrupted mattress
pattern.65'88'9o'9l

Almost every author on esophageal surgery quotes the
leaking suture line as the major immediate complication, and
both leakage and surgical field contamination can predispose
to infection. Experiments have been done comparing post-
operative results following esophageal resection and
anastomosis, with and without the use of antibiotics.
Pearlstein showed that leakage increased without antibiotics

and systemic antibiotics (procaine penicillin G and

dihydrostreptomycin) significantly decreased operative

13

mortality.33

There is general agreement that antibiotics
are beneficial. No trials have been done comparing the
postoperative mortality with different antibiotics. Nor is
there any information on the normal equine esophageal
bacterial flora. Under scanning electron microscopy large
numbers of bacterial colonies can be seen, predominantly
composed of coccobacilli.94 The population of bacteria

and the population of desquamated cells increases in the

more caudal esophagus. However, the specific bacterial

species were not identified in this study.

Extraoral Alimentation

 

There is some contradiction in the literature with
regard to the use of intraluminal tubes following esophageal
surgery. Some authors suggest that they are contraindicated
and slow mucosal regeneration.80 They may also affect
certain segments of the esophageal wall, but not the whole

92

circumference. Conversely, in piglets, an intraluminal

transanastomotic tube is beneficial in long gap correction
to permit mucosal regeneration in a situation where some

degree of dehiscence is likely.37'40

Strictures, though
less severe than with no tube at all in the lumen, still
form. Others claim an indwelling pharyngeal feeding tube is
indicated to minimize movement of the organ and encourage

healing.91

14

Since the nature of the diet does affect the healing
process in the esophageal wound,20 researchers have used
a variety of postoperative feeding techniques following
esophageal resection and anastomosis. Dogs and piglets have
been placed on a soft diet within 2-4 days of surgery and
soon resumed a normal diet.24'25'29’30’32’34'36'37'43'45'47'48
Projects using piglets, in which the use of an intraluminal

tube was included,37'4o’46

merely stated that when there

was a good chance that mucosal separation would occur, an
indwelling tube across the anastomosis would allow mucosal
regeneration to take place, and decrease the degree of
stricture formation. Nasogastric and esophagostomy tube-
feeding have been used extensively in people to permit
extraoral alimentation during treatment of oral and
pharyngeal disease. Human surgical patients are usually fed
intravenously after esophageal resection and then placed on
a liquid diet or several days. The cost involved in the use
of total parenteral nutrition in horses and the difficulty
of finding a palatable liquid diet with its own attendant
problems, e.g. diarrhea, make these means of feeding
impractical in horses. Esophagitis has been induced in the
dog by the use of pharyngostomy tubes left in place for 14
days. There was mild gastric reflux and erosive and

ulcerative lesions along the esophagus.95

The potential
complications of esophagostomy tube-feeding in horses

include: traction diverticuli, reflux of food around the

15
feeding tube, obstruction of the feeding tubes,
mediastinitis, dissecting tracts in the neck, fistulae,
jugular thrombophlebitis, and dislodgement of the

4'13'19 However, its use in association with

tubes.
esophageal resection and anastomosis in horses has not been
fully evaluated. Nasogastric tube-feeding has been used
extensively in horses, and descriptions of tubes left in

96

place for up to 14 days have been made. Horses have

also been fed using a nasogastric tube passed 3 times daily

for 18 days.97

Chronic nasopharyngeal tubes have also been
associated with a number of complications in people --
esophageal stricture, traumatic pharyngeal
pseudodiverticulum, peritonitis, laryngeal necrosis,
abscessation of the nasal septum and inability to adequately

clear lower respiratory tact secretions.98"102

Diagnostic Techniques

 

The method most commonly used to assess the degree
of stricture postoperatively is contrast radiography, both
in vivo or at postmortem. In the large number of projects

34'36 and Kornfalt,7-41

performed by both Livaditis
on esophageal resection and anastomosis in piglets,
radiographic postmortem stricture diameter measurements were

41 It is

slightly larger than those measured in vivo.
disappointing that the postoperative appearance of the

cervical esophagus in the 4 successful equine esophageal

16

resections and anastomoses discussed at the beginning of
this review, was not documented by contrast radiography.
Methods for employing contrast radiography, and the
radiographic appearance of both the normal and diseased
equine cervical esophagus have been well
described.l3'18’81’92'93

The advent of the flexible fiberoptic endoscope in
equine practice, has greatly facilitated examination of the
esophagus in horses. The endoscopic appearance of the
normal esophagus, as well as the healing process after
sutured and non—sutured esophagotomies in ponies, has also
been documented.18

It is clear from the foregoing discussion that most
of the experimental data gathered on esophageal resection
and anastomosis has come from research on small animals and
piglets. The remainder has accumulated from clinical
results in people. There are many avenues for exploration
concerning esophageal resection and anastomosis in horses.
We know that feeding method has a significant effect on the
way a longitudinal incision in the esophagus heals. The
surgical techniques used did not change the final outcome,
however.20 This project therefore was designed to
assess the effect of three different, clinically applicable
feeding methods on the healing process following an

esophageal mucosal resection and anastomosis. A mucosal

resection was chosen since lesions resulting in stricture

17
formation in horses are often confined only to the mucosa

and submucosa, and are often of a localized nature.

MATERIALS AND METHODS

Experimental Design

 

Twenty clinically normal horses (15 geldings and 5
mares) were used. Preoperative endoscopy and contrast
radiography of each horse's esophagus showed no anatomic
abnormalities.

A ventral cervical midline approach was used to
perform an esophageal mucosal resection and anastomosis
through a longitudinal incision in the esophageal muscle on
all horses. Five horses (Group I) were given access to
water only for 2 days postoperatively, and then were fed a
mash dieta (14 g/kg of body weight/day), until the 14th
postoperative day. Five horses (Group II) were fed the same
diet in slurry form, via a nasogastric tube (placed at the
time of surgery), until the 14th postoperative day. Ten
horses (Group III) were fed the same diet through a distal
cervical esophagostomy tube, until the 14th postoperative
day. The slurry was made by soaking the daily pelleted
requirement in water, and dividing this into three daily
feedings. The quantity of water required to administer this

via a tube, causes marked diuresis and consequently

 

aComplete Horse Feed No. 705, The Andersons, Maumee, 0H.

18

l9
adequately fulfills the daily water requirement of the
horse. After the 14th postoperative day, all horses were

given the dry pelleted ration.

Surgery

Immediately prior to anesthetic induction, a
nasogastric tube was passed as far as the distal esophagus.
The horses were premedicated with xylazineb (0.05 mg/kg
of body weight), and were anesthetized with thiamylal
sodiumC (2%) and guaifenesind given to effect. The
horses were placed in dorsal recumbency. Anesthesia was
maintained with halothane administered in a semiclosed
system by endotracheal tube. The skin on the ventral
surface of the neck was prepared for aseptic surgery.

A 15-cm incision was made on the ventral midline,
centered over the junction of proximal and middle thirds of
the neck. The cutaneous colli and paired muscles of the
sternohyoid, sternothyroid, and omohyoid were separated
along the midline to expose the trachea. The trachea was
retracted to the right, and the esophagus was identified
with the nasogastric tube in its lumen. Electrocautery was
used to control bleeding. A balfour retractor was used to

retract structures within the carotid sheath abaxially and

 

bRompun, Chemagro, Division of Baychem Corp., Kansas City, MO.
CSurital, Parke, Davis and Co., Detroit, MI.

dGlycodex (5%), Burns-Biotec Laboratories, Oakland, CA.

20

the trachea to the right of midline. The neighboring
tissues were protected with 2% providone-iodinee soaked
sponges to reduce contamination from the luminal contents.

An 8-cm longitudinal incision was made through the
muscularis externa and submucosa of the esophagus, using a
scalpel. Using blunt dissection, the mucosa was then
separated from the submucosa and muscularis externa around
its entire circumference, for a length of about 4 cm. The
nasogastric tube was removed just proximal to the surgery
site, and a 3-cm length of esophageal mucosa resected.
Suction was used to evacuate the contents from the
esophageal lumen adjacent to the surgery site.
Electrocautery was used on the esophageal wall only if
hemorrhage obscured the surgical site and could not be
controlled with pressure. The mucosa was apposed beginning
on the dorsal surface with 4 continuous sutures of 2-0
monofilament polyprOpylene.f Each was tied when 900
of the mucosal circumference was apposed. The second suture
was begun on the dorsal surface and brought 900 around
the mucosal circumference in the other direction and tied.
The two remaining sutures were used to complete the
anastomosis on the ventral surface. All knots were tied
within the lumen. The suture was pulled only tight enough

to prevent leakage. By distending the mucosa with air blown

 

eBetadine, Purdue Frederick Co., Norwalk, CT.

Prolene, Ethicon, Inc., Somerville, NJ.

21
gently through the nasogastric tube, the anastomosis could
be checked for leakage. If any leaks were detected, a
simple interrupted suture was placed over the defect. For
horses in Group II, the feeding tube was passed through the
anastomosis into the stomach. In Group III horses, the tube
was passed into the thorax, to facilitate location of the
esophagus for the esophagostomy. The esophageal muscle and
submucosa were closed using simple interrupted sutures of
2-0 monofilament polypropylene.f A polyethylene
draing (6.5 mm OD) was placed beside the esophagus, and
brought out distal to the skin incision through a small stab
wound. Muscle layers and subcutaneous tissues were apposed
with 2-0 monofilament polypropylene in a simple continuous
pattern, except distally, where simple interrupted sutures
were used to close the distal 4 cm of each incision in case
further drainage was required. Lavage with 2%
povidone-iodine continued during surgery and closure. A
continuous horizontal mattress suture pattern was used in
the skin except the distal 4 cm which was closed with simple
interrupted sutures. Suction was maintained on the drain
using a 60 ml syringe. A 14 gauge needle was placed through
the plunger and over the end of the syringe to maintain the
plunger at the 50 m1 graduation. A 3-way stop-cock was
connected between the syringe and the drain and enabled

evacuation of fluid and re—establishment of suction every 2

 

gHemovac, Snyder Laboratories, New Philadelphia, OH.

22
hours initially, and then at longer intervals as less fluid

accumulated.

Placement of Feeding Tubes

 

In Group II horses, the distal end of the
nasogastric tube was positioned in the stomach and sutured
to the false nostril prior to recovery from anesthesia.
Group III horses were placed in right lateral recumbency
while still under general anesthesia and the left distal
jugular furrow was prepared for surgery. A S-cm incision
was made ventral to the left external jugular vein, as close
to the thoracic inlet as possible, followed by blunt
dissection dorsal to the sternocephalicus muscle toward the
trachea. The nasogastric tube was identified within the
esophageal lumen. The esophagus was elevated to permit a
4-cm longitudinal incision to be made in the muscularis
externa. The mucosa was grasped with Allis tissue forceps
and the nasogastric tube was removed. A longitudinal
incision was made in the mucosa just large enouqh to permit
passage of another nasogastric tube through this incision
into the stomach. The esophagostomy tube was fastened in
place using two butterfly tapes sutured proximally along the
neck. The butterfly tapes were fashioned using one inch
wide adhesive tape placed around the tube, leaving a flange
on each side of the tube which was then sutured to the skin.

The skin was left open. Upon recovery from general

23
anesthesia, the neck was bandaged using a nonadhesive
sterile cotton dressingh held in place with an adhesive
tapei placed around the neck. Phenylbutazone (4 mg/kg)
was given BID until the 5th postoperative day--intravenously

in Group I horses and orally to horses in Groups II and III.

Physical Examination

 

Skin wounds and suction drains were examined daily.
Swelling, discharge and/or dehiscence were used as
indicators of infection. The presence of saliva (or food
material in Group I) in the suction drain was used as
evidence of esophageal mucosal and muscle dehiscence.
Purulent material draining into the suction device indicated
an infection. When either occurred, the distal sutures in
the skin and cervical muscles were removed to allow greater
drainage, at which time the drains were removed. In those
horses in which dehiscence was severe, all sutures were
removed, and the wound left open to granulate. Lavage was
performed with 10% povidone-iodine. Remaining sutures were
removed when the skin was dry and drainage had stopped.
Skin incisions and esophagostomy stomas were considered
healed when: (1) they were completely covered with
epithelium and food or saliva did not leak from the wound;

and (2) barium did not leak from the esophagus or

 

hTriple pad, Parke Davis, Detroit, MI.

lElasticon, Johnson and Johnson, New Brunswick, NJ.

24
esophagostomy stoma when the esophagus was placed under
pressure during the positive pressure esophagram. This
latter criteria was found a more reliable indicator than the
leakage of food or saliva after the approach incision had
healed. A fistula was defined as a narrow communication
between the esophageal lumen and the skin, and occurred
subsequent to leakage or dehiscence of the esophageal mucosa
and muscle. If the same communication existed but did not
reach the skin, it was called a sinus tract. These were

only demonstrable radiographically.

Endoscopic Examination

 

The mucosal anastomoses were examined endoscopically:J
every other day starting on day l postoperatively, until the
14th postoperative day in the lst and 3rd groups of horses.
All horses were examined weekly after that time.

Examination was facilitated by distending the esophagus with
air through the endoscope. An assessment of the degree of
dehiscence and stricture formation, and the degree of
mucosal healing was made. A stricture was defined as any
compromise in the lumen diameter demonstrated either
endoscopically or radiographically. The mucosa was
considered healed when (i) a defect was not observed in the

mucosa, (ii) the edges of the mucosal incision were in

 

JGastroscope GIF, type D3, Olympus Corp. of America,
Elk Grove Village, IL.

25
apposition, (iii) saliva did not leak from the external

18

wound and (iv) a fistula or sinus tract could not be

demonstrated radiographically.

Radiographic Procedure

 

The cervical esophagus of all horses was
radiographed 15, 30, 45 and 60 days after surgery. Each
study included a lateral survey cervical radiograph, barium
paste esophagram, and esophagrams with barium liquid
administered under pressure through a 14 ounce dose syringe
connected to a nasogastric tube (10 mm OD) fitted with an
inflatable cuffk to prevent barium reflux, followed by a
double contrast film using air under pressure. The paste

and liquid contrast media were made by mixing 98.6% barium

l m

sulfate powder and 30% barium gastric suspension
(paste = 3.1 g of powder/ml of suspension, liquid = 2.1 g of
powder/ml of suspension). The following measurements were
made from the radiographs, as shown in Figure 1: absolute
stricture diameter (a), maximum esophageal diameter cranial
to the stricture measured under barium positive pressure
(b), and upper and lower esophageal wall defect areas

measured at the surgical site. The wall defect areas were

determined by extending the line created by the dorsal and

 

kSilastic tube with inflatable cuff (12 mm inner diameter),
Bivona Surgical Inc., Gary, IN.
1 Sol-o-Pake, E-Z—EM Co., Inc., Westbury, NY.

mLiquid Sol-o-Pake, E-Z-EM Co., Inc., Westbury, NY.

26

.musmmmum m>Hufimom ESAHMD Hops:
popcoumflp muflm hummusm exp um mommanmm m0 coflumucmmmumou ownmmumofipmu 4

H mmDOHm

27

mmovz>om>r (<2... ummmoam

 

 

 

 

 

 

 

ommmoa 5 .36on o.<mm.:oc_.cz_
fl ommmoa >._. mamaémm

a 652.39 3 3202mm

a 2.52:2. 65:39

 

28
ventral esophageal wall margins on the barium positive
pressure esophagrams cranial to the stricture, across to
their respective wall margins distal to the stricture. The
area encompassed by this line and the outline of the
stricture was then measured. In order to differentiate
quantitatively between those ventral areas at the stricture
site where a traction diverticulum was present (3 horses in
Group II) instead of an invagination of the wall, the former
areas were designated as negative (Figure 1). The term
traction diverticulum was used to describe a cone-shaped
outpouching of the esophageal wall at the surgical site,
demonstrated radiographically, and sometimes endoscopically.
In order to facilitate statistical calculation, the
measurements were transformed by adding 10 square cms. The

areas were measured using a planimeter.n

A percentage
stricture diameter was calculated by expressing a as a

percentage of b.

Necropsy
All animals were euthanatized 60 days postoperatively,

using pentobarbital sodium0

at a dose rate of 7.0 mg/kg
injected intravenously. Compliance studies were performed
in situ immediately upon euthanasia and then the esophagus

was removed for qualitative and quantitative histopathology.

 

nCompensating Polar Planimeter, No. 62 0002, Kueffel and
Esser Co., NY.

OFatal Plus, Vortech Pharmaceuticals Ltd., Dearborn, MI.

29
Difficulty was encountered in removing the esophagi due to
extensive adhesion formation at the surgical site.

1. Pressure-Volume Measurements

 

The lateral aspect of the esophageal wall was
approached at the surgery site. Sufficient dissection was
carried out to enable clampsp to be placed around the
esophagus, to occlude 3 adjacent 7 cm sections, the middle
one incorporating the anastomotic site (Figure 2). Any
fistulae were closed with a hemostat, close to the
muscularis externa. Beginning with the distal segment, a
glass Y-piece was introduced through the distal clamp on the
segment. One arm was attached to a 3-way stopcock and a 60
cc syringe, and the other arm of the Y-piece to a
transducer.q We measured the pressure changes of each
segment independently by injecting saline into the lumen in
5 cc increments and measuring the resultant changes in
pressure with the transducer. Saline was introduced until
the segment appeared maximally distended (approximately 50
cc). Following inflation measurements, the measurements
were repeated as the esophagus was deflated by removing 5 cc
of saline each time, back to base line pressure measurement.
The inflation and deflation measurements were repeated. The

same procedure was repeated on the other 2 segments. The

 

pBar-Lok, Cable Tie Tool, Dennison Manufacturing Co.,
Framingham, MA.

qVR-6, Physiological Recorder, Electronics for Medicine,
White Plains, NY.

3O

.ucoEmuzmmmE mEzao>|wusmmmum Hao3 Hmmmm£m0mo m0 coflumucmmmummu UfiucEEmumMAo

N mMDUHh

mmonzgmpr ummmmcmméorczm .zmpmcmmzmza

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

032.6
.r
n njih JR ‘ 1%» \I
0829 mcSmQ mzm 08%:
h I 1
quoal I luoalnl. lens. I \.

pmmmwcmm Ambzmocomm

32
data were averaged for the inflation and deflation limbs
separately. Using a data digitizer,r a computer based
curve-fitting routine was performed on each section. The
computer estimates the parameters (A and B) of the equation
Y = A[1—10(BX)], Where Y = volume and X = pressure.
Both inflation and deflation limbs of the pressure-volume
curve were single rising exponentials.

2. Qualitative Pathology

 

The esophagus was incised longitudinally along its
dorsal wall, and the gross appearance of the mucosal
incision was examined for the presence of the
circumferential scar and any fistulae. The segment which
included the surgery site was removed, pinned to a cardboard
base under tension, to prevent separation and fragmentation
of the layers, and fixed in 10% phosphate-buffered formalin,
which is the most suitable fixative for esophageal tissue.94
The sections for histologic examination were embedded in

S and sectionedt with sharpened steel

106

paraffin in batches,
knives at 4 um. Hematoxylin and eosin (H&E) staining

was applied to sections automaticallyu in batches,

107

but the remaining stains (Gomori's trichrome, Laidlow's

108

reticulum, and orcein elastinlog) were done manually.

 

rGP-6, Sonic Data Digitizer, Science Accessories Corp.,
Southport, CT.

SHistomatic Tissue Processor, Model 166, Fisher Scientific,
Pittsburg, PA.

tAmerican Optical Microtome, American Optical Scientific
Instruments Division, Buffalo, NY.

uModel GLS-360, Honeywell Inc., Denver, CO.

33

3. Quantitative Pathology

 

Sections stained with H&E and Gomori's trichrome
were used for assessing the width of each layer in the
esophageal wall.94 The following layers were identified
for the morphometric analysis: keratinized and
nonkeratinized epithelium, lamina propria, muscularis
mucosa, submucosa and muscularis externa. In those sections
where no muscularis mucosa was identified, the lamina
propria and submucosa were collectively called the submucosa
(Figure 3). The width of each esophageal wall layer was
measured directly from a light microscope with
planachromatic lenses,v using an ocular micrometer.

Using the x4 objective, 50 gradations was equal to 1 mm.

Sections stained with orcein-elastin and Laidlow's
reticulum stains were suitable for density-dependent image
analysis. The relative area of selectively stained fibers
(elastin and reticulum), compared to background area, was
assessed by an area count of a 60,000 bit video camera,w
which viewed an area of tissue 0.81 x 0.56 mm for each
measurement. The relative proportion of stained to
unstained tissue, expessed as a fraction of the total area,
was determined for submucosa and muscularis externa on each

slide.

 

vOptomax 2 Inc., Hollis, NH.
wNikon Optiphot, Mager Scientific Inc., Dexter, MI.

35

44 KERATINIZED EPITHELIUM
(4 NON-KERATINIZED EPITHELIUM

N lAMlNA PROPRIA
« MUSCULARIS MUCOSA

« SUBMUCOSA

« MUSCULARIS EXTERNA

 

 

« ADVENTITIA

36

Statistical Analysis

 

Percent live/dead ratios for each group of horses
were compared using a row X column contingency table. Skin
healing (SH) and mucosal healing (MH) times were compared
for each group using non-parametric test for replicated
measurements: the Rank Sum Test (Mann Whitney u).109
Quantitative data were analyzed using a split plot factorial

analysis of variance. Significant differences between means

were compared using Tukey‘s procedure at p<0.05.

RESULTS

1. Clinical Data

 

(a) Group I: Only 2 of the 5 horses survived to
the 60th postoperative day. The results are summarized in
Tables 1 and 2. Statistical analysis of such a small number
was not feasible, so only the data collected from Group II
(nasogastric tube-fed) and Group III (esophagostomy
tube-fed) horses were analyzed. All mucosal anastomoses
dehisced by day 3 postoperatively (Table l). The retraction
of the mucosa at the distal edge of the anastomosis was so
severe in 3 of the horses, that it prevented passage of both
ingesta and saliva. For this reason, 2 horses were
euthanatized 4 days postoperatively (horses 4 and 5), and 1
horse 5 days postoperatively (horse 2). Two horses had
their skin, subcutaneous and muscular approach incisions
opened completely to facilitate drainage. One of these
horses survived (horse 1). The other survivor had the
distal portion of the approach incision opened 6 days
postoperatively, at which time the drain was removed (horse
3). The approach incisions of the other 2 horses were not
opened. The suction drains in the 2 survivors were removed 3

days postoperatively. The remaining 2 horses were

37

38

TABLE 1

Number of days between surgery and mucosal dehiscence (MD),
muscularis externa dehiscence (MED), opening of
approach incision (0A), drain removal (DR)
and death (D) in all groups of horses

 

 

No. of horses No. of days between surgery and
Group I (oral) MD MED 0A DR D

l 3 3 3 3 N

2 3 3 3 3 Sb

3 2 6 6 6 NAb

4 3 NA NA NA 4b

5 3 3 NA NA 4
MeanSISEM 2.8:0.2 3.810.8 4.0:l.0 4.0:1.0 4.710.3

Group II (nasogastric)a

6 l4 5 4 2 NA
7 l4 5 2 2 NA
8 l4 5 5 5 NA
9 l4 3 5 5 NA
10 14 4 ll 5 NA

Means:SEM 14.0 4.4:0.4 5.4:l.5 3.8:0.7

Group III (esophagostomy)
11 NA NA NA NA 0b
12 7 NA 7 7 NA
13 3 NA 5 5 8
l4 3 NA 5 5 4
15 3 NA 4 7 NA
16 3 4 4 4 NA
17 5 NA 5 5 28
18 3 NA NA 5 NA
19 3 NA 5 5 8b
20 3 NA 5 5 14

Mean:SEM 3.7:0.5 4.0 5.010.3 5.3:0.3 10.3:4.0

 

NA = not applicable

aNo. endoscopic examinations performed till day 14
postoperatively

bThese horses were euthanatized

39
euthanatized before the drains were removed. Dehiscence of
the esophageal muscle incision occurred in 4 of the horses.
The mean skin healing time was 45.0 i 15.0 days and the mean
mucosal healing time was 53.5 i 9.2 days (Table 2). Both

these animals developed esophageal strictures.

 

TABLE 2
Clinical Data Results of Group 1 Horses (n = 5)
Fistula 4/5 horses
Sinus tract N/A
Skin healing time (n = 2) 45.0 i 15.0 days
Mucosal healing time (n = 2) 53.5 i 9.2 days

 

(b) Group II: All 5 horses survived to the 60th
postoperative day. The clinical data from horses in Groups
II and III are summarized in Tables 1 and 3. All mucosal.
anastomoses had dehisced to varying degrees by the 14th
postoperative day (Figure 4). All horses had their approach
incisions opened distally to provide drainage, subsequent to
drain removal. Two horses developed fistulae and 2
developed sinus tracts (Figures 5 and 6). Mean skin healing
time was 50.0 i 6.0 days and mean mucosal healing time 52.0
i 7.4 days. Strictures developed in all horses and traction

diverticuli developed in 3 of these horses (Figure 5).

4o

mhcp
nmmHOS
momuos
mwcp
mmmc
mmmHo;
omega;

momu0£

omooo " Am ..Hm UgmHOWWflmV

N.m + o.mm

Ln

um.md + m.m~

m.mm

h
h
H
+l

H

moH\¢

Acouswnsu >Eoumomm£m0mov

AHHH mucuwv

wauc60mmflcmflm ouw3 mummuomummsm HmaflEwm ha pmxucE when

mHDmOHo >Eouwomm£m0mm

m

Edasowuum>flp cofluomue

uuoEmOHm>op ouswofluum

In

0v.b H o.Nm mafia mcflamms Hemousz

0.0m mEflu mcwammz cflxm

o
m
+l

N

uomuu mscflm
N ucmEQOHm>op wasumflm

mm\m Hm>w>usm amploo

Apmwlonsu UfluawmmOmmcv

HH msouo

nomuog HHH msouc can HH QDOHO m0 muazmou bump HMOfizflHU

m flam<fi

41

 

FIGURE 4

Endoscopic appearance of the esophageal mucosa 3 days
postoperatively (Horse 10, Group II) demonstrating
considerable mucosal separation along the suture line,
exposing the submucosa (arrows). The photograph was taken
when the horse dislodged its' nasogastric tube.

42

 

FIGURE 5

Endoscopic appearance at anastomotic site showing strictured
area, traction diverticulum and a fistula (arrow) at the

base of the diverticulum on the 50th postoperative day
(Horse 6, Group II).

43

 

FIGURE 6

Barium positive pressure esophagram of Horse 7 in Group II
30 days postoperatively, illustrating a stricture site (3),
traction diverticulum (arrows) and sinus tract (t).

44

(c) Group III: The results are summarized in
Tables 1 and 3. Only 4 of the 10 horses in this group
survived to the 60th postoperative day. This number was
significantly less than the Group II horses. Mucosal
dehiscence occurred to varying degrees in this group also.
There was edema and signs of mucosal separation as early as
day l postoperatively (Figure 7). The time and causes of
death in Group III horses are listed in Table 4. Three
horses had their incisions opened completely to provide
drainage, and the incisions of another 5 horses were opened
distally only. The drains were removed when these
procedures were carried out. The remaining horse's drain
was removed 3 days postoperatively, but no further drainage
procedures were necessary.

Horse 11 was euthanatized on the recovery room floor
immediately postoperatively due to a forelimb myositis/
neuritis. Horse 6 was also euthanatized because he choked
on wood shavings the day following esoPhagostomy tube
removal, and in an effort to remove the shavings with
lavage, the nasogastric tube created a perimucosal tract,
distal to the surgery site. In an effort to replace the
esophagostomy tube, to facilitate further feeding and allow
the surgical site to rest, the same problem occurred.
Peri-esophageal tracts were made into the thoracic inlet, as
the lumen could not be found through the esophagostomy

stoma. The horse was humanely destroyed.

45

 

FIGURE 7

Endoscopic appearance of the esophageal mucosa 1 day
postoperatively, from Horse 18 in Group III, demonstrating
mucosal swelling and early appearance of submucosa at the
suture line (arrow), indicating tension on the anastomosis.

46

TABLE 4

Horse, day of death postoperatively and cause of death
in Group III (esophagostomy tube-fed horses)

Horse Time Cause
(days postoperatively)

 

 

 

11* 0 Forelimb
myositis/neuritis

13 8 Asphyxiation

l4 4 Asphyxiation

17 28 Unknown

19 8 **Phenoxybenzamine
overdose

20* 15 Peri-esophageal tracts

*euthanasia

**Dibenzaline, Smith, Kline and French, Division of
Smithkleine, Beckman Corp., Philadelphia, PA 19101.

47
Horse 17 died of unknown causes 28 days postoperatively.
Horse 19 developed acute laminitis, and died when it was
treated with the alpha-blocker, phenoxybenzamine, for the
second time. Horses l3 and 14 died from asphyxiation due to
bilateral laryngeal paralysis.

One horse developed a fistula (Figure 8), and one a
sinus tract. Mean skin healing time of 35.8 i 17.7 days was
significantly faster than the Group II horses. Similarly,
mean mucosal healing time (25.5 i 13.9 days) was
significantly faster than that of Group II. Five horses
developed strictures i.e., all those that survived long
enough to demonstrate them radiographically. The mean

esophagostomy stoma closure time was 33.0 i 8.2 days.

Miscellaneous Complications

 

Apart from the surgical.complications already
described, there were a number of other problems, some of
which were not related directly to the surgery. The
nasogastric tube, which was left in place for 14 days in the
horses in Group II caused a rhinitis characterized by
mucopurulent nasal discharge and nasal mucosal ulceration.
Wretching movements were not uncommon in these horses and
they occasionally managed to displace their tubes also.
Reflux of mash around and through the nasogastric tubes
occurred when the stomach was full. Tubal obstruction could

usually be relieved by pressure applied by mouth or a 14

48

13?.I331.
,
A

“-..-...",

. . g“ ,u.

 

-- a 5mm . v-wq-u». w...»

:11). 44.4910...

V.

 

   

FIGURE 8

ing a stricture

hagram illustrat

tive pressure esop

ium posi

Bar

(5), fistula (f) and esophagostomy stoma (es) (Horse 16,

Group III).

49
ounce dose syringe. Variable quantities of saliva leaked
from the esophagostomy stomas in the Group III horses
depending on the size of the incision in the esophageal
mucosa. Two of the surviving horses in this group developed
jugular thrombosis. In addition to the 2 horses in Group
III that asphyxiated due to bilateral laryngeal paralysis, 2
other horses in this group developed left laryngeal
hemiplegia (Figure 9), one of which did not resolve. One of
the horses in Group III that asphyxiated (horse 14) also had
radial nerve paralysis in the immediate postoperative
period. Two horses in Group II developed colic. Horse 7
had ileus immediately postoperatively and horse 9 had two
episodes of spasmodic colic 28 and 55 days postoperatively.
Both horses recovered. Two horses in this group, one of
whom died 28 days postoperatively (horse 17, Table 4) of
unknown etiology, developed retro-pharyngeal swelling in the
immediate postoperative period (Figure 10). The other in
this duo needed a tracheostomy (horse 16).

Choke was a continuing problem particularly in some
of the esophagostomy tube-fed and orally fed horses, due to
stricture formation (Figure 5). There was a tendency for
the horses to ingest the wood shavings from the stall floor,
necessitating the use of a bare stall for these animals.

The esophagostomy tube-fed horses also had more depressed
appetites when the tubes were removed 14 days post-

operatively, and some needed supplementary tube-feeding.

50

 

FIGURE 9

Endoscopic appearance of the larynx of one of the horses
(Horse 15, Group III) showing left laryngeal hemiplegia in
which the left arytenoid cartilage fails to abduct on
inspiration.

51

 

FIGURE 10

Endoscopic appearance of the larynx and caudal dorsal
pharyngeal area showing left laryngeal hemiplegia and
occlusion of the pharyngeal lumen due to pressure from a
retropharyngeal swelling (Horse 16, Group III).

52

2. Radiology

 

(a) Group I: The mean absolute stricture diameter
(Figure l) for the two horses that survived in this group,
is shown in Table 5. The calculated percentage stricture
diameters for the same time periods are shown in Table 6.
Although statistical analyses could not be performed as only
2 horses survived, the trends in the changes in the absolute
lumen diameter and percentage stricture diameter did not
follow the same pattern as those for the other two groups of
horses. There was a decrease in absolute lumen diameter and
percentage stricture diameter between 15 and 30 days
postoperatively, followed by an increase at the 45 and 60
postoperative day time periods. The decrease in diameter
was not seen in the other two groups of horses between 15
and 30 days postoperatively.

The upper and lower esophageal wall defect areas are
shown in Table 7. The general trend was for the upper wall
area to decrease, as the lumen enlarged over time, while the
lower wall area remained about the same.

(b) Group II and Group III: There was no
significant difference between these two groups in absolute
stricture diameter at the surgery site, percentage stricture
diameter, and wall defect area measurements above and below
the stricture site. Absolute stricture diameter and

percentage stricture diameter increased with time in both

53

.mmmuon m haco oumB when» dozen H msouw so mequuwm
Doc wm3 mammamcm HMUMumfiumum .am>oa mo.o way an ucoummmwp haunmowmacmfln
won one used 05mm on» SD cmuoonumpcs HHH one HH wmsouo mo mace:

 

 

 

 

.. I I I .4 n :v
m.m + m.m~ m.m + m.ma ¢.m + m.mH m.H + m.- HHH macaw
I I I I Am n so
m.m + m.e~ m.m + o.mm m.~ + o.ma e.~ + e.ea HH nacho
II l. .l. I. ANNE“
o.m + o.mH o.~ + 0.0 m.o + m.m o.~ + o.oH H macho

 

00 mg om ma

 

>Hm>flumumm0pmom when

 

mEmummszmo ousnmmum m>wyflmom Ezflumn Eoum
pouzmmoe >Hm>flumummoumom when 0M cum mu .Om .ma um
mwmnoz mo mucum some uOm AEU a“ Emm+mv wouofimwp ousuofluum musHOmnd

m mqmdfi

54

.H mzouw so poEquumm
no: mm3 mammamcm Hmowumflumum .Ho>oa no.0 on» an acmummmwp haucmowmwcmwm
you one used been onu ma concomumpc: HHH can HH mmsouw Eon“ memo:

 

 

 

I I I I Aw u as

c.oH + m.mm ~.~ + m.mv n.m + m.>m ¢.m + m.nm HHH scene
I .I II .I Am u as

~.m + ¢.mh o.m + o.ms ~.m + m.mm m.e + e.em HH nacho
II II I l. aNrIICv
4.0 + 0.0m m.o + m.mm 0.4 + o.o~ m.m + m.em H macho

 

00 m? Om ma

 

>Hm>flumuom0pmom when

 

mEcummSQOmm whammmum o>fluwmom Ezfiumn Scum ConsummE I
.maw>wumuom0umom exec on can mu .om .ma #0 Azmm+mv
Aumwwfimep HmommsmOmm Hawcmuo Eseflxms ecu mo mmmucoouom n no
pummwumxm HouoEmwp manpowuum oUDHOmnmv umumfimap musuofluum mmmucmoumm

m mdmde

.uommuo mEHu HGMUHMHcmHm

on mm3 ounce

 

 

4.0 H m.0 m.0 H 0.0 H.0 H 0.0 H.0 H 0.0

«.0 H 0.0 H.0 H 4.0 H.0 H n.0 m.0 H ~.H

m.H H n.0u N.H H ¢.HI m.H H 0.HI 0.0 H 0.H

0.0 H 0.0 m.0 H 5.0 m.0 H >.H m.0 H e.H

N.0 H v.~ H.0 H A.H 0.H H e.m ~.0 H m.m

H.0 H m.H «.0 H 4.0 H.0 H 0.0 0.0 H m.H
00 me on ma

HoBOA
momma
HHH macho
uo3oq
“0mm:
HH mzouw
uo3oa
momma

H msouw

 

>Hm>Humuomoumom mama

 

>~o>Humummoumom
whee om cum mv .Om .mH um nmmuox m0 mzowm Sumo How
ANEU :H Emm+mv nmmum uoommp HHm3 HummMSQOmm H030H cum Momma

h mamdfi

56
groups (Tables 5 and 6). In Group II, the nasogastric
tube-fed horses, there was a significant increase in
absolute lumen diameter between 30, 45 and 60 days
postoperatively (Figure 11). The greatest increase in
absolute lumen diameter occurred between 30 and 45 days
postoperatively. The same changes occurred in the Group III
(esophagostomy tube-fed) horses, the significant increase in
absolute lumen diameter occurring between 30 and 45 days
postoperatively. This change in lumen diameter in the same
horse is demonstrated by the endoscopic appearance of the
esophagus in horse 12, Group III, 30 days postoperatively
(Figure 12) and then at 60 days postoperatively (Figure 13).
The corresponding radiographic appearance of the esophagus
at the stricture site at the same time is shown in Figures
14 and 15. At each of the time periods measured, the
percentage stricture diameter also increased significantly
except between the 45 and 60 day time period in Group II and
between the 15 and 30 day time period in Group III horses.
The major changes occurred however, in both Groups between
30 and 45 days postoperatively. At 60 days postoperatively
the mean percentage stricture diameter was 78% of the
maximum cranial diameter in the Group II horses and 52% in
the Group III horses. These comparative differences in
lumen diameter are demonstrated raiographically in Figures

16 and 17.

PER CENT STRICTURE DIAMETER

.....
000000
666666

 

 

58

 

FIGURE 12

Endoscopic appearance of the stricture 30 days
postoperatively in Horse 12, Group III.

59

 

FIGURE 13

Endoscopic appearance of the stricture in the same horse as
Figure 12, 60 days postoperatively.

60

 

FIGURE 14

Barium double contrast esophagram made 30 days
postoperatively from the same horse as in Figures 12 and 13.

61

 

FIGURE 15

Barium double contrast esophagram made 60 days
postoperatively from the same horse as in Figures 12, 13,
and 14.

62

 

 

FIGURE 16

Barium double contrast esophagram of Horse 7, Group II,
taken 60 days postoperatively. Note the traction
diverticulum (t) and fistula (f).

63

 

 

FIGURE 17

Barium positive pressure esophagram of Horse 12, Group III,
taken 60 days postoperatively.

64

No significant difference could be demonstrated
between groups in the esophageal wall defect areas (Table
7). Nor was there a significant alteration over time.
However, the mean upper wall defect area was significantly
larger than the lower wall defect area in both Groups II and
III (Table 8) (Figure 18). This is a reflection of the
traction diverticuli that predominated in the healing

process in the Group II horses.

TABLE 8

Upper and lower esophageal wall defect areas for horses
in Groups II and III (§:SEM in cm )

 

 

Upper* Lower
Group II 1.0 i 0.4 -0.8 i 0.6
Group III 0.7 i 0.2 0.5 i 0.1

 

*Significant difference at 0.05 level between mean upper
and lower areas.

3. Pressure—Volume Measurements

 

The procedure for measuring the pressure-volume
curve of the esophageal wall was only performed on Group I
and II. Because there was no difference in compliance
between the area in which the surgery had been performed and

the relatively normal esophagus cranial and caudal to it in

65

AREA (CMZ)

 

I: 7/ WA“- DEFECT AREA
': Z fl:
'3 R

*Sig. Difference at 0.05 Level

FIGURE 18

Histogram showing upper and lower wall defect areas for all
horses in Groups II and III.

66
these groups, I did not perform the measurements for Group
III. The mean pressure changes occurring in all segments in
Group I and II horses, for each 5 ml change in volume are
shown in Table 9. The average pressure-volume curve derived
from all these values in Group II is illustrated in Figure
19. The esophagus was more compliant on deflation than
inflation. A computer-based curve—fitting routine was used
to derive the values of A and B, as already described and
these values were used to compare each curve and each

section (Table 10).

4. Qualitative Pathology

 

The sections stained with hematoxylin and eosin
(H&E) and Gomori's trichrome were used for qualitative light
microscopic evaluation and esophageal wall cross-sectional
measurements.

(a) Group I: Because 3 horses in this group had to
be euthanatized less than five days postoperatively, the
surgical sites could be examined immediately following
dehiscence. Lesions consisted of a suppurative esophagitis
which extended on either side of the underrun submucosa in
all layers of the esophageal wall (Figure 20). The
fibroblasts were hyperplastic and hypertrophic with large
nucleoli and irregular cytoplasms. Other changes included
abscess formation with bacterial colonization, vasculitis,

lymphatic distention and edema (Figure 21). There was also

67

 

 

 

 

 

mm.mm vam.eos mm.m~ H.mm.m- om
Nv.mH H mm.H~H Hm.mm H oe.emH om.mH H mH.eeH he.am_H mo.mmH me
mm.HH H om.sHH oo.mo H em.moa mo.ms H_Hm.~eH mo.me.H mm.¢~fl oe
so.o H ne.oe hv.m~ H mm.~v ve.m H ee.eo~ m.me_H mv.mm mm
Na.e H eo.me ma.m H em.mm o~.o H sm.eo mo.mm H_mm.He om
mm.H H ~m.m~ me.m H em.mH mo.m H_mm.~m em.hH H mo.em mm
He.o.H mm.m oo.¢ H ms.- mm.H H_ep.om me.HH H mm.m~ om
am.o H we.m ~m.m H mm.n m>.o H mo.HH mo.h_H m».¢H ms
om.o H oh.m oo.~ H oo.m mo.o H Ho.m mm.e H.hm.m oH
mm.@ H nm.~ eo.m H mv.v mm.o H es.~ o~.m H oH.w m
o o o o 0
HH macaw H macho HH msouw H macaw
:oHumHmwQ :oHumHmcH AHEV mEsHo>
whammwhm
mmuHm Hmonmo can HmonHDm .HQHGMHU

ofiv mo ouHmomEoo m who mousmmmum mSB
Ham How HHM3 HomOM£m0mm mnu mo mocmHHQEOU msu mmeHumw Cu

.msmmfim0mm mfie mo COHWMHmmp can

:OHuoHMGH mcHuso AHEV mESHo> cam AHHOH CH Emm+wv musmmmum

poms mumB dump mmmxs

a mdmfis

.mmmuos HH one H mmsouu

68

 

 

 

 

 

 

 

 

 

deco. + memo.a m.H + m.He omoo. + memo.u o.m + o.me mmoo. + memo.u n.¢ + ”.me
m a m a m a
.Hmmo .Hmwn .ammo
mmoo. H ammo.s m.H H m.Hv omoo. H_mnmo.u ~.m H m.ee mmoo. H ofleo.n m.v H o.me
m a m a m a
.chH .33; .HmcH
Hopsmo mHHw HMUHmHsm HchmuU

 

 

A.meov coHumHmmp paw A.Hm:Hv coHumHMCH Suon Hops:
.Hmwsmhov mconmH mmhnB

muHm HMUHmusm

.pmHSmmmE meB Aampsmo paw
.meHDEoo mmwuon HH can H msouw

MOM w>uso GEDHO>Imu5wmmum HmwucwsomMm mchHu mamch mfiu wmnHuowmp SUH£3
maxmvoHIHud n % :oHumsvw mnu How Azmm+mv m w d mo mmsHm> cmumHDUHmolumusmEOU

OH mqmdfi

I PRESSURE-VOLUME

 

 

 

 

 

 

25 50 7'5 IGO 1'25 I60
PRESSURE (torr)

FIGURE 19

Pressure-volume curve of the esophagus representing the
average inflation and deflation limbs of all horses in Group
II. The intraluminal pressure is shown on the horizontal
axis, and the volume of saline administered is shown on the
vertical axis. Standard error bars are marked. There was a
significant difference between the inflation and deflation
limbs at the 0.05 level.

70

A _., >444." .

' ’vr wo-vw—q-u twig. 11‘9”}.1 —.._‘..~,1.-. ..1-[p‘3-nr'qvlqfimfin ;”_-,

 

FIGURE 20

Histopathologic appearance of a longitudinal cross-section
of the esophageal wall from Horse 2, Group I, 3 days
postoperatively. Note the underrunning.of the submucosa and
the abscess formation in the same area (Gomori's Trichrome
stain, original magnification X2). m = mucosa: sm =
submucosa; a = abscess; me = muscularis externa.

71

 

 

FIGURE 21

Submucosal inflammation 3 days postoperatively in the same
horse as in Figure 20. Submucosal tissues contain extensive
amounts of proteinaceous edema fluid (ef) in which numerous
bacteria are present (arrows).~ The inflammatory cells
present are predominantly neutrophils (H&E stain: original
magnification X1000).

72
a purulent myositis (Figure 22) and considerable
periesophageal suppurative cellulitis.

The histopathologic changes in the two horses in
this group that followed the protocol to the 60th
postoperative day were those of a chronic inflammation. In
longitudinal sections taken across the mucosal scar and
extending through the muscularis externa incision, there was
dense fibrous tissue obliterating all layers underneath the
mucosa (Figure 23). Horse 16, Group III is used as an
example. These changes were also characteristic of the
other horses in Groups II and III that survived to the 60th
postoperative day. In sections away from the muscularis
externa incision, there was thickening of the submucosa
below the mucosal scar (Figure 24). Horse 6 in Group II is
used as an example. Regeneration and keratinization of the
hypertrophic hyperplastic mucosa was not always complete
(Figure 25). The predominant inflammatory cell types
present in the submucosa were macrophages, lymphocytes and
neutrophils. There was some atrophy of the muscularis
externa, and suture tracts associated with marked
granulomata formation and very dense collagen deposition
(Figure 26). Horse 10. Group II is used as an example.

(b) Group II: The chronic inflammatory reactions
in the esophageal wall in these horses were similar to Group
I horses in the 60 day postoperative period. There was

variability in the degree of mucosal healing in these

73

 

FIGURE 22

Purulent myositis in the muscularis externa of the same
horse as in Figures 20 and 21, 3 days postoperatively. The
muscle fibers are degenerating. and fibers are widely
separated by edema fluid and neutrophils (Gomori's Trichrome
stain; original magnification X100).

74

 

 

 

FIGURE 23

Histopathologic appearance of the esophageal wall in Horse
16. Group III, in transverse section, illustrating the dense
fibrous tissue in the area through the muscularis externa
incision. Note that the mucosa is completely healed
(Gomori's Trichrome stain; original magnification x2).

75

. :V' ,' 1
. h #-
.~a£fi¢%hfi”

.4 .-“ " 7!“;
3'“ figs-y ’35:??? q I"
' 3¢a%§géfigfifiggghf

ah

r,«xag_
: 9

3 v fa.
.- r.., ‘ _ ~ ~ ,q ., _
-....w up; , -- ‘3" .v.
“us-Jil- 3..., a‘th;vflmwéflsh m6?

 

 

FIGURE 24

Histopathologic appearance of the esophageal wall in
longitudinal section, from Horse 6, Group II, showing
thickening in the submucosa and hypertrophy and hyperplasia
of the mucosa at the surgical site in the center. Normal
esophageal wall layers are present on each side (Gomori's
Trichrome stain; original magnification X1.5).

76

 

 

FIGURE 25

Histopathologic appearance of the esophageal wall
demonstrating on the extreme left an area of ulceration and
inflammatory cell lining from Horse 1, Group I. Adjacent to
this area, the mucosa has regenerated (arrow) but is
abnormal. There is vacuolation of the mucosa and incomplete
keratinization. On the right of the photomicrograph, the
mucosa is keratinized (magenta color), but the mucosa
remains hyperplastic. The submucosa is hypercellular
(Gomori's Trichrome stain; original magnification X20).

 

 

FIGURE 26

A longitudinal cross-section across the mucosal surgical
site in Horse 10, Group II, showing suture granulomata
(arrows) and a sinus tract (t). Epithelial regeneration
around the sinus tract is incomplete and adjacent epithelium
is hypertrophied (Gomori's Trichrome stain; original
magnification X2).

78

horses. An ulcer remained in one horse (Figure 25) While in
another, the mucosa was nearly normal. A sinus tract was
present in one section (Figure 26). Particles of a
rhomboidal shape observed in one granuloma were birefringent
under polarized light and were suggestive of talc particles.

(c) Group III: With 2 exceptions, these horses
were similar to the above groups: Firstly, in all horses
mucosal repair was complete. In most horses the site of
surgical incision would still be identified because the
mucosa was hypertrophied but in 2 horses no zone of
hypertrophy was identified; secondly, this group
demonstrated more periesophageal cellulitis than horses in
other groups. One horse from Group III had an extensive
granulomatous reaction with giant cell formation in both the
submucosa and muscularis externa. In some of these giant
cells there were refractile fragments whose shape was
suggestive of suture material (Figure 27). Other changes in
the esophageal adventitia of this horse included nerve fiber
degeneration, severe fibroelastosis and thrombosis of

arteries and patchy muscular degeneration.

Esophageal Wall Cross-Sectional Dimensions

No significant differences in width could be
demonstrated between the horses in Groups II and III in any
layer of the esophageal wall that was measured (Table 11).

In both groups, however, the submucosa was significantly

79

 

FIGURE 27

Histopathologic appearance of the adventitia in Horse 16,
Group III. 60 days postoperatively (viewed under polarized
light), illustrating granulomatous inflammation. Numerous
giant cells containing fragments of refractile material are
present. The fibrillar material that is partially
refractile is characteristic of collagen (H&E stain;
original magnification X100).

80

 

 

 

0k~.0 H 0H0.e 000.0 H 000.0 000.0 H 000.0 HH03 H000000000 H0009
e0~.0 H 000.0 000.0 H 005.0 000.0 H emm.m maneuxm magmasomzz
000.0 H e0~.~ «00.0 H ~a0.H e0H.0 H 00H.H mammflu 0>Huomccoo H0009
000.0 H e00.H 0em.0 H 00H.H 00H.0 H 0eH.H 0000:2000
000090 000.0 H 000.0 000.0 H 0H0.0 mmoose wanessomsz
000090 000.0 H 000.0 000.0 H 000.0 3.30.20 001.54
«00.0 H 00m.0 ~0H.0 H «00.0 000.0 H 000.0 Essamsnemm H0009
0H0.0 H ¢0H.0 00H.0 H 0me.0 000.0 H 000.0 Esasmsunmm eunucsumumxucoz
000.0 H 000.0 000.0 H 000.0 0H0.0 H.0a0.0 sessmrnnme omnacuumumx
HHH QDOHU HH QDOHU H msouo um>mH mammHB
HHH paw .HH .H mmsouo :H mmmuon How

ASE :H Emmfimv mnmme mammHu HmmmMSQOmm mo SHOH3 HmsoHuommlmmouo

HH wand?

81
thicker at the mucosal surgical site than away from it.
These measurements were taken in an area well away from the
muscularis externa incision (Figure 24). The mean
submucosal width for both Group II and III horses at the
surgical site was 1.3 i 0.1 mm compared 0 1.0 i 0.1 mm away

from the surgical site.

Image Analysis

 

The density-dependent image analysis using the
reticulum and elastin stains was performed on the submucosa
and muscularis externa for horses in Groups II and III
(Table 12). There was a significantly greater amount of
both elastin and reticulum fiber staining in the Group III
horses. There was significantly more elastin and reticulum
fiber staining in the submucosa than the muscularis externa
in both groups. However, the reticulum staining was greater
in the muscularis externa in the Group II horses, but in the

submucosa in the Group III horses.

82

 

 

 

.Hm>mH mo.o um wmpHm cam mmsoum cmm3umn museummme HCQOHMHcmHm
0000. H 0000. 0000. H 0000. 0000. H 0000. 0000. H 0000. 0000000 0000000002
0000. H 0000. 0000. H 0000. 0000. H 0000. 0000. H.0000. 0000:0000
000000000 0000000 000000000 0000000
HHH @5000 HH @9000

 

00000: HHH paw HH mmsouw CH

msuouxm mHHMHsomsE ps0 mmoosEQsm 00m .0000 pcsoumXWMQ mo 00000 m 00
mquHm 550500000 pom sHumem pmchum >H0>HuumHmm mo 02mm+m9 0000 m>HumHmH 055

NH mqmfifi

DISCUSSION AND CONCLUSIONS

The main objective of this study was to evaluate
three different feeding regimens in horses by studying
esophageal wound healing, following an esophageal mucosal
resection and anastomosis. The feeding regimens chosen were
ones which could be applied in a clinical situation. It was
not the purpose of the study to directly evaluate surgical
techniques for esophageal mucosal resections because only
one surgical technique was used. However, some comments
will be made regarding a resection and anastomosis of the
esophagus. It was also my intention to develop some
quantitative techniques to better evaluate the healing
process occurring in biological issues like the eSOphagus,
following surgical intervention and in doing so, to better
relate structural and functional changes that occur in the
esophagus following this type of surgery. Unfortunately,
not all of these objectives were met. By discussing various
aspects of the project in chronological order, I shall
attempt to highlight where these objectives were achieved,

and where they fell short.

83

84

Clinical Observations

 

Within 3 days postoperatively, the mucosal
anastomoses had completely dehisced in all those horses fed
per os 2 days after surgery (Group I). The mucosa retracted
from the underlying submucosa. completely obliterating the
lumen in 3 of these horses and they showed ptyalism and
esophagismus. Because they were unable to eat, they were
humanely destroyed. Using this surgical technique and
probably for any surgical technique involving resection of
esophageal tissue, this feeding method is unacceptable.
However, variable degrees of dehiscence occurred in the
mucosal anastomoses in the other 2 groups also.
Unfortunately, due to the difficulty of maneuvering the
endoscope in the esophagus with a large nasogastric tube in
place, the horses in the nasogastric tube-fed group were not
examined before the 14th postoperative day. Due to the
formation of strictures in all the groups and the
predominance of fistulae and sinus tracts in the second
group, it is reasonable to assume that mucosal dehiscence
also occurred in this group. Extraoral alimentation
facilitated healing in the second and third groups of

horses.

Surgical Technique and Related Complications

 

According to the endoscopic appearance of the

mucosal anastomosis, dehiscence began as early as 1 day

85
postoperatively. This was too early for infection or
compromised blood supply to be contributing to tissue
weakness and necrosis. Tension on the mucosa, as a result
of the 3 cm resection, and tissue edema caused early tearing
out of the suture. The type of suture material,
polypropylene, and the pattern may have contributed to the
degree of breakdown in the anastomosis. Continuous suture
lines have a greater effect on decreasing vascular supply to
the ends of the tissues, but tension on the anastomosis
appears to contribute predominantly to the breakdown, at
least in the initial stages. In an experiment comparing
one-layer simple interrupted chromic catgut and non-
absorbable Teflon-coated polyester fiber, after a 5 cm
resection was performed, deaths were attributed to
anastomotic leaks.. There was a higher occurrence in the
group in which non-absorbable material was used. However,
the dehiscence was related to loosening and cutting through
to the suture, rather than to the anastomotic

92

technique. Strictures may develop without luminal

leakage due to excessive tension, leading to mucosal

separation.33'34’38'41

In this project, tension on the
anastomosis appeared to contribute greatly to early mucosal
dehiscence.

Significantly more horses survived in the

nasogastric tube—fed group than in the esophagostomy

tube-fed one. This data alone suggested that nasogastric

86
tube feeding was superior to esophagostomy tube-feeding
following this technique of esophageal mucosal resection.
The second surgical incision in the necks of the third group
of horses and the presence of the esophagostomy tube
predisposed these horses to a variety of other
complications, several of which were fatal. Sepsis appeared
to play a major role in the pathogenesis of these problems.
There was accumulation of saliva and purulent discharge
which together resulted in phlegmon development.

All but one of the horses in the esophagostomy
tube-fed group needed extra drainage at the surgical site.
The suction drains were not able to cope with the exudate
that developed as a result of incisional inflammation. The
drains themselves may have contributed to the infection
rate. The decision not to use prophylactic antibiotics in a
surgery that invaded the lumen of the gastrointestinal tract
may have contributed to the complications. Cellulitis
however, was not only related to breakdown of the muscularis
externa incision and leakage, with the subsequent formation
of fistulae; since only one horse in the third group
developed a fistula, yet sepsis was a major problem in this
group. All but one of the surgical incisions in this group
required additional drainage. The increased complication
rate in this group appeared to be a result of the presence

of two incisions in the same area.

87

I planned to remove the drains by the third
postoperative day, or at such time as the quantity of
exudate was approaching 40 mls per day, which is compatible
with the presence of the drain acting as a foreign body in
the wound. However, the quantities either never decreased
to this volume, or the character of the drainage became
purulent necessitating the opening of the distal portion, or
all, of the incision. The presence of the drains themselves
may have contributed to the sepsis rate. Additionally, this
surgical procedure required some tissue trauma with the
placement of large retractors in the wound for a prolonged
period and the tissues were generally exposed to the
operating room environment in excess of 2 hours.
Contamination from the lumen and the environment therefore
also contributed to the complication rate.

Skin and mucosa healed in a shorter time in the
esophagostomy tube-fed group due to the larger number of
fistulae and sinus tracts present in the nasogastric
tube-fed group. Indeed, once the mucosal ulcer had healed,
these strictures were often only demonstrable
radiographically, in the esoPhagostomy tube-fed group.

The incidence of fistulae and sinus tracts which
indicate the extent of mucosal and muscular leakage were
highest in Group II (nasogastric tube-fed). The presence of

traction diverticuli was greatest in the naSOgastric

88
tube-fed horses. It appears then that leakage predisposes

to traction diverticulum formation.

Intraluminal Tubes

 

The presence of a nasogastric tube in the esophageal
lumen appeared to predispose to leakage at the surgical
site, due to the high incidence of wall defects in the
second group. The statements by some authors that
intraluminal tubes place greater pressure on some parts of
the wall compared to other parts cannot be proven by this

study.79'91

It does seem however, that the tube either
increases the intraluminal pressure sufficiently to cause a
breakdown in the suture line, or simply by mechnical trauma
it delays healing at the mucosal anastomosis and
subsequently causes a breakdown in the muscularis externa
incision. It does not seem to minimize movement of the
organ and encourage healing.90 According to the

absolute stricture and percentage stricture diameter
measurements, there was no difference in the esophageal
lumen diameters between Groups II and III. However, the
mean diameter of the lumen 60 days postoperatively in the
nasogastric tube-fed group of horses was 78.4 i 2.2 percent
of maximal cranial diameter, but was 52.5 i 10.0 percent in
the esophagostomy tube—fed-horses (Group III). Some of this

increase in lumen diameter in the second group is due to the

traction diverticulum formation at the mucosal anastomosis.

89

However, the presence of a large bore tube across the
surgical site appeared to keep the esophageal walls further
apart as the mucosa begins to regenerate. This lends
credence to the experiment that showed that in piglets
undergoing long gap correction, strictures formed but were
less severe if a transanastomotic tube was
present.36’39'45

Esophagi of the second and third groups of horses
showed an increase in lumen diameter particularly between 30
and 45 days postoperatively. The natural bougienage of
eating must have caused the increase in lumen diameter that
occurred. At this stage, the collagenous tissue is
presumably still amenable to stretching but by 60 days, the
collagen is dense and mature histologically, and the
fibroblastic activity, with proliferation of reticulin
fibrils is greatly decreased. Also there is no change in

the traction diverticulum shape and size by 60 days

postoperatively, in experimental work done on ponies.19

Esophagostomy Complications

 

In the esophagostomy tube-fed group of horses, there
were 6 of 10 fatalities (Table l). The first horse had to
be euthanatized immediately on recovery from anesthesia due
to bilateral forelimb paralysis which was attributed to a
rhabdomyositis/radial neuritis. This loss was not directly

related to the surgical procedure and was probably the

90
result of poor positioning or padding. The same can be said
of the third horse that died as a result of a
phenoxybenzamine overdose which was used as a treatment for
this horse's acute laminitis. However, laminitis is not an
uncommon sequela to surgery and anesthesia in horses.

The second horse died of unknown causes 28 days
postoperatively. However, at necropsy, there was a septic
left jugular thrombophlebitis and pulmonary
microabscessation. Two of the horses asphyxiated as a
result of bilateral laryngeal paralysis. Evidence of a
respiratory problem did not develop in these horses until
about the fourth postoperative day. This coincided with the
development of the septic incisional complications and
purulent drainage. The suppurative cellulitis, associated
with the trauma of the surgery, probably affected the
function of both the left and right recurrent laryngeal
nerves. The same factors probably resulted in the
retropharyngeal swelling and left laryngeal hemiplegia seen
in the other two horses in the study. These complications
can be related to the combination of two surgical procedures
on the same animal in close proximity, resulting in
increased inflammation and infection. The occurrence of
postoperative ileus in one of the horses in Group II could
be a result of vagal inflammation and edema. This can again

be related to sepsis.

91

Difficulty encountered on re—passage of
esophagostomy tubes has been previously reported.18
However, on the occasions that horses in the second group
dislodged their nasogastric tubes, no difficulty was
encountered on their re—passage through the anastomotic
site.

The use of three different radiographic measurements
did not highlight any greater differences between groups
than absolute stricture diameter alone. There was no real
advantage using the percentage stricture diameter
measurement. However, it did permit a better appreciation
of the relative degree of compromise of the esophageal
lumen. It was anticipated that some of the variation
between horses might be eliminated using this technique
because even under the influence of xylazine, it is not
always possible to uniformly and consistently distend the
esophagus under barium or barium/air positive pressure, and

simultaneously radiograph each horse.

Pressure-Volume Measurements

 

The pressure-volume measurements were an attempt to
correlate a measure of esophageal function with the degree
of structural change we knew existed at the surgical site.
We were unable to demonstrate any difference between the
compliance of the surgical site and the sites cranial and

caudal to it, in any of the horses. This was surprising, as

92
there was a large amount of fibrosis, both in the esophageal
wall at the surgical site and in the more ventral
periesophageal tissues. Presumably due to the
two-compartmental nature of the model we were testing, there
was enough normal esophageal and periesophageal tissue to
mask the less compliant nature of the fibrous tissue. The
only significant difference we identified was between the
inflation and deflation limbs of the pressure-volume curve.
The deflation limb was the more compliant. The esophagus
undergoing deflation distended more than when undergoing
inflation under the same pressure. This phenomenon, called
hysteresis, is also observed in tissues like the lung. The
fact that we were unable to demonstrate a change in the
pressure-volume curve at the surgical site, is consistent
with the way the esophagus behaved under barium positive

pressure.

Qualitative Histopathology

 

Histopathologic changes occurred in all layers of'
the esophageal wall, even in those sections taken away from
the longitudinal muscularis externa incision. Since all
mucosal anastomoses dehisced to varying degrees, all the
surgical incisions were heavily colonized by bacteria
particularly in the submucosa, and underwent the chronic
suppurative reaction seen in the 3 horses in the first group

less than 5 days postoperatively. This explains the dense

93
fibrous tissue and chronic inflammation seen in the 60 day
postoperative esophageal sections. The presence of an ulcer
in one horse and the presence of a fistulous tract on
longitudinal section in another horse in Group II show that
healing was still incomplete as the chronic suppurative
reaction had still not resolved. Even by 60 days, most of
the horses had hypertrophic, hyperplastic mucosae,
illustrating that healing was not complete. Both the
severity of the initial acute reaction and presence of
non-absorbable suture material in the presence of an
infection could contribute to the long healing time in the
esophagus. However, the suture material in the mucosa
usually sloughed as soon as the mucosal anastomosis
dehisced. In Group II, the irritation of the
transanastomotic nasogastric tube for the first 14 days
postoperatively would also contribute to the slower healing
of the mucosa, as well as the presence of fistulae and sinus
tracts. In support of this, the mucosa in the Group III
horses, although still hypertrophic and hyperplastic, was
completely keratinized, indicating the stage of healing was
more advanced.

A chronic periesophagitis was present in all horses
but was accentuated in the adventitia of the esophagostomy
tube-fed group. These horses suffered a more severe septic
reaction in their surgical incisions, as discussed

previously. The severe degenerative changes, particularly

94
evident in one of the group characterized by adventitial
nerve fiber degeneration, indicate that there is a
functional abnormality in the esophagus as well as the
structural changes already described.

The foreign bodies - talc and non-absorbable suture
material - induced a severe granulomatous reaction,
characterized by giant cell formation. The presence of
fragments of suture material would invite one to question
whether the non-absorbable suture material used retains its
strength for 60 days after insertion. However, in the
presence of such a severe reaction in the esophagi of most
of these horses, the cellular and enzymatic response was
probably more pronounced than in most other surgical
procedures, that might be used to assess suture material

characteristics.

Morphometrics

 

Morphometric analysis, Where the thickness of
individual layers in the esophageal wall were measured in ”
cross section, did not demonstrate changes in cross-
sectional esophageal width between horses on different
feeding regimens. However, the technique is reproducible

and accurate,94

and in a surgical procedure where
healing by primary intention is achieved, it may be more

helpful. Quantitation of pathologic change is necessary so

95
that structure and function can be better related, and the
degree of structural alteration assessed.

The density-dependent image analysis was used to try
and assess the relative contributions of reticulum and
elastic fibers to the healing esophageal wall. Laidlow's
reticulum is a silver-based stain. Argyrophilic fibers,
consisting of very fine fibrils seen close to the cell
surface of the fibroblast have been termed eticulin.
Collagenous reticulins are found in association with
actively synthesizing fibroblasts. The staining reaction
appears to be due to associated with carbohydrate or lipid
components.79 The very dense collagen seen
histologically in these tissues, 60 days postoperatively is
apparently mature and not undergoing any significant
remodeling. This would suggest that the structural form of
the esophageal wall at 60 days postoperatively will not
undergo much change or at most, very slow change. The
larger quantities of reticulum and elastin fiber staining in
the Group III horses, is probably more a reflection of
tissue preparation and staining procedure, than a real
difference between groups.

As can be seen from the radiographic measurements,
most of the increase in stricture diameter occurred between
30 and 45 days postoperatively. Not much increase occurred
after that time. This is consistent with the presence of

the mature dense collagen 60 days after surgery. This

96
change in lumen size that occurs between 30 and 45 days
postoperatively has important clinical applications. In
horses presenting with circumferential mucosal and/or
submucosal esophageal erosion, there is an advantage in
encouraging horse owners to persist in feeding their horses
a sloppy mashed ration until 30 to 45 days after the
original injury. At this time the esophageal lumen should
enlarge, often enough to permit maintenance on a pelleted

diet, if not hay.

Conclusion

 

The survival rate alone in the first group makes the
oral method of feeding unacceptable. Although no
significant increase could be demonstrated in lumen diameter
between the nasogastric and tube-fed horses, the trend was
for the nasogastric tube—fed group to have larger mean lumen
diameters over all time periods and specifically at 60 days
postoperatively. Additionally, all the nasogastric tube-fed
horses (Group II) survived to the end of the project, which
outweighs the complications of fistula and sinus tract
development which predominated in this group over the
esophagostomy tube-fed one. Using this surgical technique
of esophageal mucosal resection and anastomosis, the
nasogastric tube-feeding technique is the preferred one to
use. Of the three feeding techniques tested, this one best
permits return toward normal, and most importantly, provides

the best survival rate.

SUMMARY

All mucosal anastomoses dehisced to varying degrees.
All horses developed strictures, subsequent to
anastomotic dehiscence.

Esophageal lumen diameter enlarged predominantly between
30 and 45 days postoperatively.

Barium and barium—air positive pressure esophagrams are
very useful for delineating strictures, fistulae and
sinus tracts.

The presence of a nasogastric tube across an esophageal
mucosal resection and anastomosis predisposes to
fistula, sinus tract and traction diverticulum
formation.

The presence of both an esophagostomy and an esophageal
mucosal resection and anastomosis in the same horse
greatly increases the complication rate.

Of the three feeding regimens used, nasogastric tube
feeding resulted in the best postoperative healing,
following the esophageal mucosal resection and
anastomosis.

Compliance measurements failed to reflect the degree of

esophageal structural damage.

97

10.

98
Morphometric procedures - cross-sectional wall
compartment dimensions and density dependent image
analysis, did not assist in differentiating between
effects of the different feeding regimens.
Histologic changes at the surgery site consisted of
extensive fibrosis, abscess and granuloma formation in

all esophageal layers beneath the mucosa.

LI ST OF REFERENCES

10-

11.

12.

LIST OF REFERENCES

Vaughn JT, Hoffer RE: An approach to correction of
cervical esophageal stricture in the equine. Auburn Vet
63-66, 1963.

Haasjes C: Esophageal diverticulum. J Am Vet Med Assoc
109:277, 1946.

Raker CW, Sayers A: Esophageal rupture in a
standardbred mare. J Am Vet Med Assoc 133:371-373,
1958.

Freeman ED, Naylor JM: Cervical esophagotomy to permit
extraoral feeding in the horse. J Am Vet Med Assoc
172:314—319, 1978.

Scott EA, Snoy P, Prasse KN, Hoffman PE, Thrall DE:
Intramural esophageal cyst in a horse. J Am Vet Med
Assoc 171:652—654, 1977.

Saurer TE, Greeley RE: Surgical approach to the equine
esophagus. Southwestern Vet 283-285, 1969.

Demoor A, Wouters L, Mouens Y, Verschooten F: Surgical
management of traumatic esophageal rupture in a foal.
Equine Vet J 11:265-266. 1979.

Hackett RP, Dyer RM, Hoffer RE: Surgical correction of
esophageal diverticulum in a horse. J Am Vet Med Assoc
173:998-1000, 1978.

Aanes WA: The diagnosis and surgical repair of
diverticulum of the esophagus, in Proceedings. let
Annual Meeting, Am Assoc Equine Pract, 211-222, 1975.

Lowe JE: Esophageal anastomosis in a horse. A case
report. Cornell Vet 54:636-641. 1964.

Fretz PB: Repair of eSOphageal stricture in a horse.
Mod Vet Pract 52:31-34, 1972.

Hoffer RE, Barber SM, Kallfelz FA, Petro SP: Esophageal

patch grafting as a treatment for esophageal stricture
in a horse. J Am Vet Med Assoc 171:350-354, 1977.

99

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

100

Freeman DE: The esophagus. In Mansmann RA, McAllister
ED (eds): Equine Medicine and Surgery, ed 3, Vol 1.
Santa Barbara, CA., American Veterinary Publications,
Inc., 476-497, 1982.

Wagner PC, Rantanen NW: Myotomy as a treatment for
esophageal stricture in a horse. Eq Pract 2:40-45.
1980.

Wingfield Digby NJ, Burguez PN: Esophageal rupture in
the horse. Equine Vet J 15:169-179, 1982.

Suann CJ: Esophageal resection and anastomosis as a
treatment for esophageal stricture in the horse. Equine
Vet J 14:163-164, 1982.

Derksen FJ, Stick JA: Resection and anastomosis of
esophageal stricture in a foal. Eq Pract 5:17-20, 1983.

Stick JA, Krehbiel JD, Kunze DJ, Wortman JA: Esophageal
healing in the pony: Comparison of sutured vs.
non-sutured esophagotomy. Am J Vet Res 42:733-734,
1981.

Stick JA, Derksen JF, Scott EA: Equine cervical
esophagostomy: Complications associated with duration
and location of feeding tubes. Am J Vet Res 42:727-732.
1981.

Stick JA, Slocombe RF, Derksen FJ, Scott EA:
Esophagotomy in the pony: Comparison of surgical
techniques and feeding regimens. Am J Vet Res (in
press).

Gundvall RL, Kingrey BW: Choke in Shetland ponies
caused by boluses. J Am Vet Med Assoc 133:75-76, 1958.

Adkins PC: Tumors of the esophagus, in Sabiston DC
(ed): Textbook of Surgery, The BologIEal Basis of
Modern Surgical Practice, ed 12. Philadelphia, PA., WB
Saunders Co., 1981, PP 841-855.

Talbert JL: Corrosive strictures of the esophagus, in
Sabiston DC (ed): Textbook of Surgery, The Biological
Basis of Modern Surgical Practice, ed 12. Philadelphia,

PA., WB Saunders Co., 1981, PP 834-841.

Parker EF, Brockington WS: Esophageal resection with
end-to-end anastomosis: Experimental and clinical
observations. Ann Surg 129:588-605, 1949.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

101

Postlethwait RW, Weinberg M, Jenkins LB, Brockington WS:

Mechanical strength of esophageal anastomosis. Ann Surg
133:472-476, 1951.

Shek JL, Prietto CA, Tuttle WM, O'Brien EJ: An
experimental study of the blood supply of the esophagus
and its relation to esophageal resection and
anastomoses. J Thorac Surg 19:523-533, 1950.

Nuangsombut J, Hankins JR, Mason GR, McLaughlin JS: The
use of circular myotomy to facilitate resection and
end-to-end anastomosis of the esophagus. J Thorac
Cardiovasc Surg 68:522-529. 1974.

Rossello PJ, Lebron H, Franco AAR: The technique of
myotomy in esophageal reconstruction: An experimental
study. J Pediatr Surg 15:430-432, 1980.

Carveth SW, Schlegel JR, Code CF, Ellis FH: Esophageal
motility after vagotomy, phrenicotomy, myotomy, and
myomectomy in dogs. Surg Gynecol Obstet 114:31-42,
1962. ’-

Postlethwait RW, Deaton WR, Bradshaw HH, Williams RW,
Winston—Salem NC: Esophageal anastomosis: Types and
methods of suture. Surg 28:537-542. 1950.

Macmanus JE, Dameron JT, Paine JR: The extent to which
one may interfere with the blood supply of the esophagus
and obtain healing on anastomosis. Surg 28:11-23, 1950.

Henderson RD, Fung K, Dube P, Marryatt G: Esophageal
reconstruction: An experimental approach to the control
of reflux after esophageal resection. Can J Surg
18:165-169, 1975.

Pearlstein L, Polk HC: Esophageal anastomotic
integrity. Review of Surgery 34:137-139, 1977.

Livaditis A, Okmian L, Bjorck G, Ivemark B: Esophageal
suture anastomosis: An experimental study in piglets.
Scan J Thor Cardiovasc Surg 3:163-173, 1969.

Livaditis A, Ivemark B: Esophageal anastomosis in
piglets: Histologic and microangiographic aspects of
early phases of healing. Scand J Thor Cardiovasc Surg
3:174-180, 1979.

Livaditis A, Bjorck G, Kangstrom L: Esophageal
myectomy: An experimental study in piglets. Scand J
Thor Cardiovasc Surg 3:181-185, 1969.

37.

38.

39.

400

41.

42.

43.

44.

45.

46.

47.

480

102

Kornfalt SA, Okmian L, Jonsson N: Healing of esophageal
anastomosis after release of tension by myotomy. Z
Kinderchir 12:444-456, 1973.

Livaditis A, Radberg L, Odensjo G: Esophageal
end-to-end anastomosis: Reduction of anastomotic tension
by circular myotomy. Scand J Thor Cardiovasc Surg
6:206—214, 1972.

Kornfalt SA, Okmian L, Jonsson N: Mucosal defects and
esophageal stricture formation: An experimental study in
the piglet. Z Kinderschir 13:392-400. 1973.

Kornfalt SA, Okmian L, Jonsson N: Healing of circular
esophageal mucosal defects: An experimental study in the
piglet. Z Kinderchir 13:184-197. 1973.

Kornfalt SA, Nelson 0, Olin T: X—ray evaluation of the
experimental esophageal anastomosis. Z Kinderchir
15:280-283, 1974.

Livaditis A, Radberg L, Jonsson N, Odensjo G:
Experimental evaluation of esophageal muscular cuffs in
the repair or esophageal defects. Scand J Thor
Cardiovasc Surg 9:155-161, 1975.

Nelson 0, Okmian L: Single layer and two layer
esophageal end-to-end anastomosis: An experimental study
in the piglet. Z Kinderchir 19:6-15, 1976.

Nelson 0, Okmian L: Intra— and extraluminal suture
knots in the esophageal end—to-end anastomosis: An
experimental study in the piglet. Z Kinderchir
19:235-246. 1976.

Nelson 0, Okmian L: Healing of esophageal end-to-end
anastomoses one, two and three weeks postoperatively: An
experimental study in the piglet. Z Kinderchir
19:353-364, 1976.

Kornfalt SA, Okmian L: Oesophageal clamp anastomosis
and stricture formation: An experimental study in the
piglet. Scan J Thor Cardiovasc Surg 10:263-266. 1976.

Nelson 0, Okmian L, Claesson G: A contribution to the
appraisal of stricture formation in the esophageal
end-to-end anastomosis: An experimental study in the
piglet. Z Kinderchir 18:141-154. 1976.

Nelson 0, Okmian L: Polyglycolic acid sutures (Dexon)
in esophageal end-to-end anastomosis: An experimental
study in the piglet. Z Kinderchir 18:253-265. 1976.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

103

Takada Y, Kart G, Filler RM: Circular myotomy and
esophageal length and safe esophageal anastomosis: An
experimental study. J Pediat Surg 16:343-348, 1981.

Janssens J, DeWerer I, Vantrappen G, Agg HO, Hellemans
J: Peristalsis in smooth muscle esophagus after
transection and bolus deviation. Gastroenterology
71:1004-1009, 1976.

Bergstrom S, Lundh B: Healing of esophageal
anastomosis: Animal experiments. Ann Surg 150:142-148.
1959.

Czerny: Neve operationen: vor laufige mittheilung. Zbl
Chir 4:433, 1877.

Bilroth T: Ueber die resection des oesophagus. Arch
Klin Chir 13:65, 1871.

Evans A: Rubber es0phagus. Br J Surg 20:388, 1933.

Torek F: The first successful case of resection of the
thoracic portion of the esophagas for carcinoma. Surg
Gynecol Obstet 16:614—617, 1913.

Ohsawa T: The surgery of the esophagus. Jap Chir
10:604-695, 1933.

Marshall SF: Carcinoma of the esophagus: Successful
resection of lower end of esophagus with reestablishment
of esophageal gastric continuity. Surg Clin North Am
18:643. 1938.

Haight C, Towsley HA: Congenital atresia of the
esophagus with tracheoesophageal fistula: Extrapleural.
ligation of fistula and end-to-end anastomosis of
esophageal segments. Surg Gynecol Obstet 76:672-688.
1943.

Ellis FH: The esophagus: Historical aspects of anatomy,
in Sabiston DC (ed): Textbook of Surgery: The
Eiological Basis of Modern Surgical Practice.
Philadelphia, PA., WB Saunders Co., 1981, pp 793-795.

Brewer LA 111: One-stage resection of carcinoma of
cervical esophagus with subpharyngeal
esophagogastrostomy. Ann Surg 130:9-20, 1949.

May IA, Samson PC: Esophageal reconstruction and
replacements. Ann Thorac Surg 7:249-277, 1969.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

104

Brewer LA III: History of surgery of the esophagus. Am
J Surg 139:730-743, 1980.

Grabb WC, Smith JW: Plastic Surgery. Little Brown and
Co., Boston, MA., 1973, pp 423-450.

Belsey R: Reconstruction of the esophagus with left
colon. J Thorac Cardiovasc Surg 49:35-55, 1965.

Wilson GP: Ulcerative eSOphagitis and esophageal
stricture. JAAHA 13:180-185. 1977.

Akiyama H, Miyazono H, Tsurumaru M, Hashimoto C, et a1:
Use of stomach as an esophageal substitute. Ann Surg
188:606-610, 1978.

Pavletic MM: Canine axial pattern flaps utilizing the
omocervical, thoracodorsal and deep circumflex iliac
direct cutaneous arteries. Am J Vet Res 42:391-406,
1981.

Swensen O, Magruder TV: Experimental esophagotomy.
Surg 15:954-963. 1944.

Hiebert CA, Cumming GO: Successful replacement of the
cervical esophagus by transplantation and
revascularization of a free gastric antrum. Ann Surg
154:103-106, 1961.

Wilkins EW, Skinner DB: Surgery of the esophagus
(concluded). N Engl J Med 278:887-891, 1968.

Edgarton MT: One-stage reconstruction of the cervical
esoPhagus or trachea. Surg 3:239—250, 1952.

Ariyan S: Pectoralis major, sternomastoid, and other
musculocutaneous flaps for head and neck reconstruction.
Clin Plast Surg 7:89-109, 1980.

Petrovsky BV: The use of diaphragm grafts for plastic
operations in thoracic surgery. J Thor Cardiovasc Surg
41:348-355. 1961.

Hopper CL, Howes EL: Strength of esophageal anastomoses
repaired with autogenous pericardial grafts. Surg Gynec
Obst 117:83-86, 1963.

Salamon J, Nudelman F, Kissin L, Gassner S, Levy MJ:

Experimental segmental replacement of esophagus by
biological tissue. Israel J Med Sci 13:272—277, 1977.

76.

77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

105

Kolt I, Gassner S, Luria M, Urea I: The tissue tube as
replacement of esophagus. Med Chir Dig 6:273-275, 1977.

Daido R, Kano M, Hirotani K, Ohashi I, Nakamura M: A
new method of esophageal anastomosis by means of the
deviced ring. Jap J Surg 1:63-71, 1971.

Strahan RW, Sajedee M, DeVal MK: The leaking esophageal
suture line. Am J Surg 106:570-574, 1963.

Pavletic MM: Reconstructive esophageal surgery in the
dog: A literature review and case report. JAAHA
17:435-444. 1981.

Peacock EE, Van Winkle W: Healing and repair of
viscera, in Wound Repair, ed 2. Philadelphia, PA., WB
Saunders Co., 1976, pp 629-635.

Stick JA: Surgery of the esophagus. Vet Clin North Am:
Large Animal Pract 4(1):33-39, 1982.

Hofmeyer CFB: The digestive system, in Oehme FW, Prier
JE (eds): Textbook or Large Animal Surgery. Baltimore,
MD., Williams and Wilkins Co., 1974, pp 364-449.

Nickel R, Schummer A, Seiferle E, et al: The alimentary
canal of the horse, in The Viscera of the Domestic
Mammals, ed 1. New York, NY., Springer-Verlag, 1973.

Filston HC, Merton DF, Kirks DR: Initial care of
esophageal atresia to facilitate potential primary
anastomosis. Southern Med J 74:1530-1532, 1981.

DeLorimier AA, Harrison MR: Long gap esophageal
atresia. J Thorac Cardiovasc Surg 79:138-141, 1980.

Ricketts RR, Luck SR, Raffensburger JG: Circular
esophagomyotomy for primary repair of long-gap
esophageal atresia. J Pediatr Surg 16:365-369. 1981.

Davidson JS: Circumferential esophageal myotomy. Brit
J Surg 59:938-947, 1972.

Rosin E: Surgery of the esophagus. Vet Clin North Am
5:557-564, 1975.

Orminger MB, Appleman HD, Argenta L, Bone E, Cimmin V:
Polypropylene suture in esophageal and gastrointestinal
operations. Surg Gynecol Obstet 144:67-70, 1977.

90-

91.

92.

93.

940

95.

96.

97.

98.

99.

100.

101.

102.

106

Baker GJ, Hoffer RE: Surgical correction of esophageal
stenosis in the dog. J Am Vet Med Assoc 148:44-47,
1966.

Sumner-Smith G: Esophagotomy and esophageal resection.
J Small Anim Pract 14:429-439. 1973.

'Borgstrom S, Lundh B: Healing of esophageal

anastomosis: Animal experiments. Ann Surg 150:142-148,
1959.

Belin RP, Lieber A, Segnitz RH: A comparison of
techniques of esophageal anastomosis. The American
Surgeon 38:533-536, 1972.

Slocombe RF, Todhunter RJ, Stick JA: Quantitative
ultrastructural anatomy of esophagus in different
regions of the horse: Effects of alternate methods of
tissue processing. Am J Vet Res 43:1137-1142, 1982.

Lantz GC, Cantwell HD, VanVleet JF, Blakemore JC, Newman
S: Pharyngostomy tube induced esophagitis in the dog:
An experimental study. JAAHA 19:207-212, 1983.

Hoffman P: Panel on practice tips. Proc Am Ass Equine
Practnr 13:134-32, 1965.

Teeter SM, Stillions MC, Nelson WE: The tube feeding of
anorectic horses. Proc Am Ass Equine Practnr 13:75-79,
1967.

Stohl EL, Holinger PH, Diffenbaugh WG: Nasogastric
intubation: Indications, complications, safeguards, and
alternate procedures. Am Surg 24:271-277, 1958.

Graham J, Barnes N, Rubenstein AS: The nasogastric tube
as a cause of esophagitis and stricture. Am J Surg
98:116-119, 1959.

McCredie JA, McDowell RFC: Oesophageal stricture
following intubation in a case of hiatus hernia. Br J

Edison B, Holinger PH: Traumatic pseudodiverticulum in
the new-born infant. J Pediatr 82:483-485. 1973.

Kassner EG, Baumstark A, Balsam D, et al: Passage of
feeding catheters into the pleural space: A radiographic
sign of trauma to the pharynx and esophagus in the
newborn. Am J Roentgenol 128:19-22, 1977.

103.

104.

105.

106.

107.

108.

109.

107

Douglas WK: Oesophageal strictures associated with
gastroduodenal intubation. Br J Surg 43:404-409. 1956.

Quick CB, Rendano VT: Equine radiology: The esophagus.
Mod Vet Pract 59:625-631. 1978.

Greet TRC: Observations on the potential role of
oesophageal radiography in the horse. Eq Vet J
14:73-79, 1982.

Luna LE (ed): Manuals of Histologic Staining Methods of
the Armed Forces Institute of Pathology, ed 3. New
York, McGraw-Hill Book Co., 1968, pp 1-46.

Sheehan DC, Hrapchak BB: Theory and Practice of
Histotechnology, ed 2. Toronto, CV Mosby Co., 1968, pp

Gurr E, Florey HW: Staining Animal Tissues: Practical
and Theoretical. London, Leonard Hill Books Ltd., 1962,
pp 267-268.

Steel RGD, Torrie JH: Principles and Procedures of
Statistics, ed 2. New York McGraw-Hill Book Co., Inc.,
1980.