the antrum in patients with duodenal gastric ulcersj. schrager, r. spink, ands. mitra stomachs,...

Gut, 1967, 8, 497

The antrum in patients with duodenal andgastric ulcers

J. SCHRAGER, R. SPINK, AND S. MITRA

From the Laboratory, Royal Albert Edward Infirmary, Wigan

EDITORIAL COMMENT In this histological study of surgical specimens it has been shown that duodenalulcer is associated with mild inflammation and gastric ulcer with marked changes and degenerationwith metaplasia.

Very few detailed descriptions of the antrum areavailable in the literature, yet the importance of thisorgan has been stressed by many writers in the past(Konjetzny, 1928; Faber, 1935; Hebbel, 1943;Magnus, 1954; Dean and Mason, 1964; and others).An attempt has been made in this laboratory to studythe gastric antrum in normal stomachs and induodenal and gastric ulcer in greater detail.

METHODS

One hundred and thirty specimens were examined.Seventy-five of these were surgical specimens from casesof duodenal ulcer and 40 from cases of gastric ulcer.Sixteen were from post-mortem specimens removedwithin a few hours of death, in which as far as could beascertained, there was no question of gastrointestinaldisease. The surgical specimens were opened immediatelyafter operation by cutting along the greater curvature andthey were pinned out with care on cork mats. They wereformalin fixed immediately. The specimens with duodenalulcer had a cuff of duodenum some 5 cm. in width. Theamount of duodenum in gastric ulcer specimens was muchreduced, but all the specimens showed an intact sphincter.The proximal portion of the specimen consisted of gastricmucosa.

After fixation, the specimens were photographed. Thesize of the photograph was exactly the same as the sizeof the stomach. A tracing was made from the photographonto thin paper. The whole of the specimen was cut intolongitudinal strips 1-Ij cm. broad. Transverse cuts werethen made through the strips at three places, the resultbeing 30-40 shorter pieces measuring approximately6 cm. in length. A plan of these cut strips was made onthe tracing from the stomach. The strips were numberedand these numbers entered on the plan. The strips werethen processed, blocked in wax, and sectioned on a sledgemicrotome. The boundary between the antrum and thebody mucosa was studied and mapped out and thesurface area of the antrum measured with a planimeter.This procedure enabled us to study thoroughly all of the

gastric mucosa and get acquainted with the whole of itscellular topography.The stains used in this study were haematoxylin and

eosin and Zimmerman's stain using the method describedby Marks and Drysdale (1957); also 30 specimens werestained with Van Gieson stain. The Zimmerman stainwas used to show up well the parietal, peptic, andmucous cells. The Van Gieson stain was used to show theincrease in fibrous tissue in the mucosa and submucosa.

RESULTS

CLASSIFICATION OF THE INFLAMMATORY CHANGES Theinflammatory changes in the antral mucosa wereclassified as follows:1 Changes showing inflammatory reaction and nodestruction of pyloric glands, and according toseverity, divided into grades Al and A2 (Figs. 1 and2); 2 changes resulting in damage and destructionof the pyloric glands with increasing amounts ofintestinalization. The final stage showed completeatrophy. According to severity this degree ofchange was subdivided into Bi (less than 25%reduction in glands), B2 (25-50% reduction inglands), and B3 (50-75% or more reduction inglands), and is shown in Figures 3, 4, and 5.

THE ANTRUM IN DUODENAL ULCER Of the 75 casesof duodenal ulcer, 71 were in males and four were infemales.The duodenal ulcer was situated at the distal

boundary of the antrum. In about one-third of thespecimens, the ulcer was situated adjacent to thepyloric mucosa, the remainder most often occurring1 cm. or so distal to the sphincter. The rest of theduodenal mucosa appeared normal. In six cases theulcer was found to lie across the boundary and toextend into the distal part of the antrum. Thesecases were mostly those in which the ulcer lay along

497

on March 6, 2020 by guest. P

rotected by copyright.http://gut.bm

j.com/

Gut: first published as 10.1136/gut.8.5.497 on 1 O

ctober 1967. Dow

nloaded from

http://gut.bmj.com/

FIG. 1.

FIG. 2.

FIG. 1. Al. Mild cell infiltration (lymphocyte andplasmacell) but no destruction ofpyloric glands.FIG. 2. A2. Similar changes as found in Fig. 1, but amore severe inflammatory reaction.FIG. 3. BJ. Inflammatory reaction and some destructionof the pyloric glands (less than 25% reduction in theglands). Two enlarged lymph follicles, one having agerminal centre.FIG. 4. B2. Similar changes to those found in BJ, buta much greater destruction in the pyloric glands.FIG. 5. B3. Severe inflammatory reaction, with intesti-nalization, and complete disappearance of pyloric glands.

FIG. 5.

HUe. J..



j.com/


ctober 1967. Dow

nloaded from

http://gut.bmj.com/

The antrum in patients with duodenal and gastric ulcers

the line of the lesser curvature. In these cases thesphincter had been eroded, and the more super-ficially situated antral mucosa destroyed.

Table I shows the distribution in 57 specimens.Specimens in which the deformity was so marked asto make the position of the ulcer uncertain wereexcluded.

TABLE ISITE OF TBE DUODENAL ULCER IN 57 CASES

Site of Ulcer

Anterior wallPosterior wallAnterior and posterior wallSingle along the line of the lesser curvatureSingle along the line of the greater curvature

No. of Cases

91819172

The inflammatory changes were diffused andinvolved the whole of the antral mucosa. Most ofthe cases fell into grade Bi (Table II).

TABLE IIINFLAMMATORY CHANGES'

Grade

A2 Bl

FIG. 6. Sharp boundary zone between antrum and bodymucosa. Diffused inflammatory reaction of the antrumstops abruptly at the boundary of normal body mucosa.

B2 B3

Duodenal ulcer 15 46 12 2(75cases) (17%) (64%) (16%) (3%)(average age at (3 with shal- (9 with shal- (2 with shal-

operation 40 yr.) low antral low antral low antralulcers) ulcers) ulcers)

Gastric ulcer - 10 19 11(40 cases) (25%) 47 %) (27%)(average age at (7 with (3 with (3 with

operation 55 yr.) chronic duo- chronic duo- chronic duo-denal ulcer or denal ulcer or denal ulcer orscar) (2 with scar) (4 with scar) (1 withshallow antral shallow antral shallow antralulcers) ulcers) ulcers)

'Microscopic erosions not included.

The changes were more severe in the mucosa alongthe lesser curvature near the boundary zone betweenantrum and body, this part being furthest removedfrom the region of the ulcer. In the B grades therewas found in this area an increasing proportion ofstomachs with erosions and shallow ulcers. Thesediffused inflammatory changes were in very markedcontrast with neighbouring body mucosa whichappeared entirely normal. The line of transitionfrom inflamed antral to body mucosa was suddenand easily discernible (Fig. 6).

In the more severe grades of inflammation, therewas an increase in fibrous tissue in the submucosa.

THE ANTRUM IN GASTRIC ULCER Of the 40 cases ofgastric ulcer, 23 were in males and 17 in females.The most severe inflammatory changes were found

in gastric ulcer cases. These showed widespread

destruction of pyloric glands, increasing intestinal-ization, and in a few cases complete mucosalatrophy. A number were found in grade Bi, but ascan be seen from Table II, at least three-quarters ofthe cases occurred in grades B2 and B3. There wasalso very marked fibrosis in the submucosa. Thiswas marked in areas furthest from the ulcer.The ulcer itself was situated, as others have noted,

near to the proximal boundary of the antrum alongthe lesser curvature (Magnus, 1954; Minoru Oi,Kingo Oshida, and Sadao Sugimura, 1959). It wasfound either entirely surrounded by antral mucosaor else partially surrounded by it with the proximalpart lying in the boundary zone. It could morejustifiably be termed an antral ulcer. It was almostinvariably single, only one specimen showing twoulcers, both in this region. The adjoining bodymucosa showed a zone of severe gastritis extendingfor a number of centimetres around it. In a quarterof the cases, the ulceration with its accompanyingzone of gastritis had occurred high up the lessercurvature. Although examination could not be asthorough with the ulcer in this situation, there seemslittle doubt that these ulcers occur at the boundaryof body with cardiac mucosa, and that ulceration ofthis kind is not uncommon.

In seven stomachs the ulceration had been severeenough to have produced stenosis of the pyloricantrum. In spite of the existence of stasis in these

499



j.com/


ctober 1967. Dow

nloaded from

http://gut.bmj.com/

J. Schrager, R. Spink, and S. Mitra

stomachs, the main inflammatory changes were stillto be found in the antrum.Magnus (1954) has recorded that of 439 cases of

chronic gastric ulcer, 54% of the ulcers occurredover 7-5 cm. from the pylorus in women in com-parison with 28% in men. The record of40 specimensin this collection is found in Table III.

TABLE IIIDISTANCE BETWEEN LOWER EDGE OF CHRONIC GASTRIC

ULCER AND THE PYLORUS

Distance (cm.) Men Women

Number Number

0- 2-525- 5050- 7-57-5-10-010-0 and overTotal

1 (4%°)4 (17%)8 (33%)8 (33%)3 (12%)

24

3 (18%)1 (6%)5 (35%)2 (11%)5 (30%)

16

Although the numbers are small, the preponder-ance shown is very slight, 45% occurring over 7 5 cm.in women in comparison with 41 % in men.

It might be remarked here that in this study of theantral mucosa, both in the normal and in cases ofdisease, apart from the transitional zone at theboundary between pyloric and body mucosa, none

of the slides showed the presence of parietal andpeptic cells. These were confined entirely to thebody mucosa. These findings do not confirm thestatement made by some authors, for example,Radasch (1921) and Palmer (1954). that parietal cellsare to be found in antral mucosa in small numbers.There was in the cases of gastric ulcer a rather less

abrupt transition from antrum to body mucosa,although in the Bi grade the transition was oftenas sharp as it was in the duodenal ulcer cases. Theremainder of the body mucosa showed a patchygastritis of variable degree, but there did not appearto be a reduction in parietal and chief cells.

Table II also shows that the gastric ulcer wasaccompanied in a high proportion of cases by anactive duodenal ulcer or ulcer scar. Since theduodenum was not always available for adequatehistological examination, it seems likely that hada sufficient amount of duodenum been furnishedwith all specimens, changes of this kind would havebeen found in most of them. The figures, althoughincomplete, nevertheless show a decrease in theproportion of combined ulcer cases with increase inthe severity of inflammation. This is likely to berelated to the diminution in secretion.

SURFACE AREA OF THE ANTRUM The antra of 45stomachs obtained at necropsy were investigated.

TABLE IVNORMAL STOMACHS

Case Sex Age GradeNo. (yr.)

Measurements alongLesser Curvature (cm.)

Sphincter Width of Sphincterto Boun- Boundary to Cardiadary Zone Zone

Measurements alongGreater Curvature (cm.)


Distance Distance Totalbetween between AreaSphincter Sphincter (sq. cm.)and Boun- and Boun-dary Zone dary Zoneas Percen- as Percen-tage of tage ofTotal Mea- Total Mea-surement surementalong Les- alongser Curva- Greaterture Curvature

Normals: good histology1 M - Al 92 M 76 A2 73 M 63 Bi 94 M 52 A2 75 M 22 A2 86 F 73 Bi 107 F 66 B2 108 M 72 A2 109 F 51 Al 810 M 66 Bi 811 M 53 B2 10Normals: indifferent histology12 M 62 - 913 M - - 714 M - - 715 M 73 - 716 F 58 - 617 M 51 - 718 M 68 - 7

907-0

7-5:.0).5D-0D-0B-0B-5D-O

*0

7-07-07-05-07-07-0

500

Antral AntralArea Area as(sq. cm.) Percen-

tage ofTotalArea

0-81 50-80-82-003

1 2071.0

2424222020262623191925

4-54040304.56-08-06-53.54-54.5

0-60-60-604070 5

0-60*50-5

4040383640404236

3642

440420330420350365425415290320600

45404045505510055555085

1010101014152313171514

4034434050423848455040

44353035303535

1010111014152018

1210

101010129912

22222521211920

40404.54.5303.54.5

42405038364036

455355675440325430375



j.com/


ctober 1967. Dow

nloaded from

http://gut.bmj.com/

TABLE VRESECTED STOMACH OF ULCER CASES

Case Sex Age Grade Measurements along Measurements along Distance Distance Total Antral AntralNo. (yr.) Lesser Curvature (cm.) Greater Curvature (cm.) between between Area Area Area as

Sphincter Sphincter (sq. cm.) (sq. cm.) Percen-Sphincter Width of Sphincter Sphincter Width of Sphincter and Boun- and Boun- tage ofto Boun- Boundary to Cardia to Boun- Boundary to Cardia dary Zone dary Zone Totaldary Zone Zone dary Zone Zone as Percen- as Percen- Area

tage of tage ofTotal Mea-Total Mea-surement surementalong Les- alongser Curva- Greaterture Curvature

--- - 80- - - - 40-- - - 65- - - - 35

- - 50

- - 70-- - - 50

- - - - 35-- - - 80- - - - 50

- - - - 110_ _-__ 45

95

- - - - 100- - - - 40

- - - - 100_-___ 75

- - - - 7555

- - - - 110_-__ 75

1006560

- - - - 70- - - - 70

65_-___ 45

7570

_ _ _ - 95

- - - - 7050

7060

- - - - 6550

65_ _ _ 60

_ -_ _ 75- - 70

- - -- 120_ _ 45

_ _ - 90- 95

- - - - 706570

_ 4055

- - - - 70_ _ 90

_ _ _ - 90

75- - 120

-- - - 55

__ - 10595

80- - - - 60

_-__ _ 85_ _ _ 70

- 95

_-__ - 90- - - - 50

-- - - 8590

_-_ _ 45_ _ 90

- - - - 55

70_ _ _ - 105

- - - - 80- - - - 65

_ _ - 90

Duodenal Ulcer2 F -3 M 554 F 435 M 44

21 M 2722 M 4025 F 4028 M -29 M -33 M 2638 M 4141 M -42 M -46 M 4049 F -53 M 5454 M 7257 M -59 M 6474 M 4863 M 3478 M 4379 M 3289 M 7292 M 45105 M 47107 M 386 M 5410 M 4511 M 5114 M 5115 M -19 M -17 M 2731 M 4834 M -35 M -37 M 7347 M 5750 M 3655 M -58 M 5266 F 6381 M 2983 M -92 M 45102 M 25118 M 5339 M -

111 M 3524 M -48 M -30 M -

Gastric Ulcer7 F 588 F 49

70 F 6588 M 58

I F -9 M18 F 5062 M 5564 M 6068 M 4973 F 5477 F 5680 F 5582 M 6287 M 5799 M 49100 F 68103 F 53110 M 46113 F 55116 M 6323 M 52

8

BiB2BiB1B2BIBIB2B2A2B2B1BiA2BiB1BiBIA2BiA2BiBiA2B1B1BiB2BiB1BiBIBiBIB2BiA2BiBiBIB2BIBIBiB2B1B1B2B3BiBIB1B1

B3BiB2B3B3B2B3B3B2B2B2B2BiBiB2B3B2B2B2B2B2B2

12-05-0652-56-58-06-05-58-06-58-05-59-08-07-08-07-07-06-57-07-57-07-07-07-07-57-56-08-5657-070706-09-0805-56-0709-07-0905.58 5707080805-5656-09090

9.511-08-012012010-06-58-59.57.5

11 08-08-07-070

10-04-55-0

10-012-08070

0-51-008251-0041-2041-7060o5021.0080 60602040*50-4070o50 303040-504

1.00-50-5

1-2

071-01-0061-2

1*5

0-5

0-5

0o5

0702

- 7030

- 60- 5-0

4.56-05*040

- 70- 40

7.5- 40- 70- 80

40- 8-0- 65-

70- o50- 10-0- 65- 7.5- 40- so0

5*0- 70

60

- 7060

- 70- 5.5- 60

5.5- 6-0

-7ss50.55*0

6-o

- s5.60

- 70-10-0

45s_ 90

70-60

75s- 60- 30

4.5

60- 70

- 7.5_7.- 7s0

7.5

-9.5

76080

- 6535s

- 7.5970-5s80

- 653.5

- 14-0_ 4.5_9-0

- 85- 80- 7.5- 85

0o50 7081 51-0040-70-404040o50-31.00 30-4040o50 3040 60706040o51-20407

1-01-00-7

04

071-01-01-00-6

1-0

1-2

03

0 6

0-50-2



j.com/


ctober 1967. Dow

nloaded from

http://gut.bmj.com/


TABLE VISTOMACHS OF ULCER CASES STUDIED AT NECROPSY

Case Sex Age GradeNo. (yr.)

Measurements alongLesser Cvrvature (cm.)


Measurements alongGreater Curvature (cm.)


Distance Distance Totalbetween between AreaSphincter Sphincter (sq. cm.)and Boun- and Boun-dary Zone dary Zoneas Percen- as Percen-tage of tage ofTotal Mea-TotalMea-surement surementalong Les- alongser Curva- Greaterture Curvature

Antral AntralArea Area as(sq. cm.) Percen-

tage ofTotalArea

FM

M

M

FM

M

M

6463

5472785972

_ 23_ 23

_ _ 2316

_ 27- - 26

2321

As far as one could ascertain there was no history ofgastrointestinal disease. The normal stomachsshowing reasonably clear histology were not easilyobtained due to the speed with which autolysisoccurs after death. Only 13 stomachs showedreasonably good histology, and the surface area ofthe antra of these was accurately charted and theboundary with the body mucosa was established.The area of the antrum in the controls is comparedwith the area found in cases of duodenal and gastriculceration in Fig. 7 and Fig. 8 compares the totalarea of the stomach in the normals with the totalarea found incidentally in 18 cases of ulcer atnecropsy. The mean differences in Fig. 8 were slight.

If the average for the normal antral area is put at50 sq. cm., about two-thirds of the duodenal ulcercases show areas larger than this. In gastric ulcercases the area is shown to be greater still and almostinvariably twice the size of the normal controls.These measurements suggest that the antrum inulcer cases is larger than in normal cases. This dirne-rence in antral size is shown also by lesser and greatercurvature measurements.The normal stomachs showing reasonably good

histology provide the most complete information(Table IV). The measurements include the distancebetween the sphincter and boundary zone along thelesser and greater curvatures, and the distancebetween sphincter and cardia along the lesser andgreater curvatures.The sphincter boundary zone distance expressed

as a percentage of the total sphincter-cardia distancealong the lesser curvature is a little over 40%, andthe same percentage along the greater curvature isusually a little over 10%. These figures correspondto those given by Landboe-Christiensen (1944).

It was not found possible to obtain complete

3644454545404445

- - - 550

- - - 540435

- - - 550_ _ 600- - - 420- - - 580

__ 495

measurements in resected stomachs, but thesphincter-boundary zone distance along the lesserand greater curvatures could be measured in suitablecases. These could be compared with the sphincter-cardia distances found in eight post-mortemstomachs showing ulcers which were found suitablefor measurement. These show that the sphincter-boundary zone distances expressed as a percentage ofthe total sphincter-cardia distance along the lessercurvature gives about the same figure as in thenormals, whereas the figure for the greater curvatureis in the region of 18 % (Tables V and VI).From Table V it will be seen that the sphincter-

boundary zone distances in ulcer cases along thelesser curvature show considerable variation from5 to 12 cm. The ulcer was found in most cases at theboundary, and it is apparent that it may vary con-siderably from the position of the incisura whereverthat may be placed. The sphincter-boundary zonedistance along the greater curvature also showedconsiderable variation, but in no case in either ulceror normal stomachs was the boundary zone foundat the pyloric sphincter but always several centi-metres proximal to it.The general outline of the antral mucosa in the

normal was a saddle shape, and in the ulcer casesit was larger and more roughly cylindrical, the lattershape the one depicted in some anatomical textbooks,e.g., Gray, Morris, Cunningham.

Included in Table V are measurements which weremade where feasible of the width of the boundaryzone. This showed considerable variation as mightbe expected, for since there was marked inflammationpresent in most, it was difficult to be certain if thepyloric glands in conjunction with gastric (body)glands had not arisen as a process of pylorisation.In some cases the width was very narrow. The mean

49

111618191410

502



j.com/


ctober 1967. Dow

nloaded from

http://gut.bmj.com/


I Normal

]D Duodenal ulcer

]1 Gastric ulcer

60 65 70 75 80 85 90 95 100 105 110 115 120 125 130

Surface area in sq.cm.

E Normal

[ Ulcer

250-300 300-350 350-400 400-450 450-500 500-550 550-600 600-650 650-700Surface area in sq.cm.

was little different on lesser and greater curvaturesand measured approximately 0 4 cm. This is a littlegreater than that found by Oi and others (0 3 cm.).The width did not differ materially between normaland ulcer stomachs, and the figures were too smallto affect the sphincter-boundary distances.

It will be seen from Table IV that in the normalstomachs the proportion of the antral area incomparison with the area of the whole of the stomachmucosa was found to vary from 10 to 15 %. Thiscorresponds to the average of 12% given byLandboe-Christiensen (1944). Unfortunately, it wasnot found possible to obtain corresponding figures

FIG. 7. Surface area of the gas-tric antrum in normal controls andin duodenal and gastric ulcerpatients.

FIG. 8. Total area from post-mortem stomach in normal andulcer cases.

for ulcer cases. If the mean size of the ulcer stomachis taken to be 500 sq. cm., about one-fifth of theulcer stomachs would show an area greater than20%.The increase in area was found in both sexes, and

no significant difference in variation in size ingastric and duodenal ulceration could be foundbetween them.

It would also appear that the increase in area isassociated with an increase in severity of inflam-mation. This is found in grades A2 and Bi (Fig. 9)in the cases showing mainly duodenal ulcerationand is more obviously shown in grades B2 and B3

E 80)0.

0

0)4

a)

E3

2

0

FIG. 7.

8

c

uu4)Q

co0

.D 4

z I2

FIG. 8.

503



j.com/


ctober 1967. Dow

nloaded from

http://gut.bmj.com/

J. Schrager, R. Spink, and L. Mitra

I A2

U0 B

I,35 40 45 50 55 60 65 7 580 85 90 95 100 105 110

FIG. 9. Relation betweenantral area and grade ofgastritis in duodenal ulcer.

Surface area in sq.cm.

which are cases showing mainly antral ulceration who had access to fresh surgical specimens, maintain(Fig. 7). that the inflammatory changes are confined to the

antrum. The term antritis would be a better one inEFFECT OF MUCOSAL DAMAGE ON SECRETION A describing the essential pathology of duodenal andnumber of the cases studied were found to have had gastric ulcer.secretory investigations performed on one (and Faber (1927-35) found the changes to be confinedsometimes more than one) occasion before to the antrum, and his material included resectedoperation. It was therefore possible to relate the specimens in which a gastroenterostomy had beenbasal secretion to the histological grading. The performed previously. In these cases the changesresults are shown in Table VII. were confined to the antrum and did not involve the

body where the gastroenterostomy opening lay.TABLE VII Konjetzny (1928), writing at the same period,

BASAL SECRETION RELATED TO HISTOLOGICAL GRADING described the changes as being confined to theType of Secretor Duodenal Ulcer Gastric Ulcer antrum and quotes other German authors (Puhl,

(55 Cases) (30 Cases) Orator) in support.A2 B) B2 B3 A2 Bl B2 B3 Magnus and Rodgers (1938), of 92 cases of gastric

Hypersecretor 5 21 1 0 0 2 1 0ulcer and found thechanges confined to theantrum in

Normosecretor 4 10 1 0 4 5 0 70% and in 20 cases of duodenal ulcer the inflam-Hyposecretor 0 4 7 1 0 0 11 7 mation was invariably in the antrum. Hebbel (1943)

did not find the inflammatory changes in the body inMost of the cases in grades A2 and Bi showed cases of duodenal ulcer, and in gastric ulcer it was

hypersecretion. Grades B2 and B3 included cases confined to the region of the ulcer. Obstruction didwhich showed mostly a marked hyposecretion. The not result in damage to the body mucosa. Meyerscases with reduced secretion showed a substantial (1948), examining material from 25 cases of duodenalreduction in the number of pyloric glands. This ulcer and 25 of gastric ulcer obtained at operation,diminished the basal secretion of free acid to a low found the inflammation to be confined to the antrum,level, and in most cases it was zero and Magnus (1952) based similar conclusions onOn referring to Table II, it is seen that most of the the examination of 635 partial gastrectomy speci-

combined ulcers occur in the Bi grade, which is the mens. Of 284 specimens of active gastric ulcer, ingrade to show the highest number of hypersecretors. 76% of the cases the inflammation was confined

to the antrum and in 256 cases of duodenal ulcer itDISCUSSION was confined to the antrum in 100%. Thompson

(1959) found inflammatory changes to be confinedThe term gastritis, when used to describe to the antrum and boundary zone.inflammatory changes in cases of duodenal and All authors have found the inflammatory changesgastric ulcer, suggests that the changes are to be to be more intense in cases of gastric ulcer. Thefound throughout the organ. Workers in this field, amount of metaplasia in cases of gastric ulcer is

8

a0E0

0co

0

.0Ez

6

4

2

504

-lq.9

I



j.com/


ctober 1967. Dow

nloaded from

http://gut.bmj.com/


much increased, and is not so often found in duo-denal ulcer. The maximum changes occur in theregion of the lesser curvature.Having localized the changes to the antrum,

most authors have held the view that the inflam-matory changes have preceded the ulceration, andthat the inflammation of the antrum is the morefundamental change, upon which the ulceration hasbeen superimposed. This view was certainly takenby pathologists who based their conclusions onexamination of post-mortem material before surgicalspecimens became available in large amounts.Stewart (1922-23), as the result of painstakingexamination of 1,500 stomachs at necropsy and 120operation specimens, expressed the view that theulcer was preceded by microscopic erosions andmacroscopically visible shallow (subacute) ulcersboth in the duodenum and antrum. Most of thesewould heal, but other factors, not necessarily thoseproducing the erosions, would result in a chronicallyinflamed ulcer. This, although often multiple in theduodenum, was usually single in the stomach. Thechronic ulcers tended always to heal in time, andgastric ulcer scars were commonly found in thismaterial. He noted that healing only infrequentlyproduced stenosis.

Faber (1927) and Konjetzny (1928) described allstages in the evolution of the chronic ulcer fromnumerous small inflammatory erosions and heldthe view that the inflammatory changes precededthe ulceration. Faber also discussed the experienceof surgeons who had furnished him with cases inwhich there had been a diagnosis of duodenal ulceron radiological evidence with marked hyperchlor-hydria. At operation no ulcer was found, but asevere antral inflammation with multiple erosions.These cases do not occur with great frequency, andindividual cases have been written up and are foundscattered throughout the literature (for example,Papayannopoulos, Dedes, Kasralis, and Arealis,1957; Gilmour, 1961). Faber noted that erosions didnot occur with great frequency and this may be theresult of most of the stomachs removed at operationbeing in a more or less quiescent stage. Magnusfound (1946) a similar state of affairs on examining20 normal stomachs at necropsy and 100 partialgastrectomy specimens. The antral mucosa showedall stages from microscopic erosions to completeintestinalization. The intestinalization was character-istic more of the gastric than the duodenal ulcer. Hepointed out that erosions were not as frequentlyfound as one might expect; he found it in 4% of hiscases. Magnus (1952), on the basis of an examinationof 284 specimens with active gastric ulcer, found it in7%. Thompson (1959) examined 35 cases of gastriculcer, 54 of duodenal ulcer, and 12 of combined

ulcer. He found all grades of inflammatory changefrom multiple microscopic erosions to extensivemetaplasia as a final result. Erosions occurredroughly in 17% of his cases. This is about the sameas has been found in this collection.Hebbel (1943) examined 78 cases of duodenal

ulcer, 16 of gastric ulcer, and 13 of combinedulcer. He concluded that ulceration developed onthe'basis of a degenerated mucosa. His system ofgrading showed the mucosa in cases of gastriculcer to be severely affected in most cases in contrastto cases of duodenal ulcer. He found erosions inonly three cases.The gastric ulcer was described by Stewart (1923

a and b) as lying almost invariably along the lessercurvature near to or at the incisura (Hurst andStewart, 1929). More recently authors who havemade a detailed study of the histology of the antralmucosa describe the gastric ulcer as being invari-ably situated in the antrum. Magnus (1954), inan analysis of 421 cases of gastric ulcer, comes tothe same conclusion. This point was also made byOi et al. (1959) who examined 149 cases of gastriculcer. The same conclusion was reached by Dean andMason (1964) as the result of examining 117 partialgastrectomy specimens. Du Plessis (1963), however,found on examining 41 cases that 70% occurred inthe antrum and 29% in the fundic mucosa. Heattributes the findings of other authors to thepylorization accompanying the development of agastric ulcer in the body mucosa, which would makeit appear that the ulcer lay in the pyloric gland area,whereas in reality it lies in a more proximal position.

It has been tacitly assumed in the past thatalthough it is likely that the gastric ulcer developson the basis of an inflammatory degeneration of themucosa, the duodenal ulcer is a different manifesta-tion of the disease, and it is only by chance that thetwo are occasionally found to coexist (Tanner,1954; Johnson, 1955). However, authors whohave collected statistics for combined ulcers areusually of the opinion that the number of coexistentulcers in any numerous set of gastric and duodenalulcers is too large to be the result of chance.Weisberg and Glass (1963) in their review of thistopic have included figures from radiological,surgical, and post-mortem studies by a dozenauthors. The post-mortem figures collected overmany years are particularly impressive (Watkinson,1960). The figures for combined ulcers in gastriculcer cases show a considerably high percentageranging from 30 to 40, whereas, as might be expected,those from cases of duodenal ulcer are very variableand considerably lower. It is pointed out by manyauthors interested in this aspect, perhaps becausemost of them are surgeons, that it is not an uncom-

505



j.com/


ctober 1967. Dow

nloaded from

http://gut.bmj.com/


mon experience for a surgeon on clinical and otherevidence to regard the case as one of duodenal ulcer,and at operation several years later to find anactive gastric ulcer with evidence of duodenalscarring. This makes it likelythat in many, if not mostcases of ulcer, a duodenal ulcer develops first anda gastric ulcer develops later after the duodenalulcer has healed. Unfortunately this leaves un-answered why only a small number of patients incomparison with the total develop a gastric ulcer.The healing of the duodenal ulcer has been ascribedto the decrease in secretion in the later stages of thedisease.The results of this investigation substantiate most

of the findings of previous investigators that themain inflammatory changes were to be found in theantrum. These changes were of varying degrees ofseverity from the relatively mild changes in cases ofduodenal ulcer to the severe degeneration anddiffused metaplasia found in most cases of gastriculcer. The changes, although involving the whole ofthe antral mucosa, stopped abruptly at the proximalborder. The widespread involvement did not appearto be secondary to the ulcer. It is just as intense incases where there is a healed scar only as in caseswith a large active ulcer. The inflammatory reactionwas more pronounced at the proximal border thanin the proximity of the ulcer. These findings suggestthat the inflammatory reactions are a primarymanifestation and that the ulcer is a sequence tothese more profound changes.The pathology of the surgical specimens fell into

two groups:

1 INFLAMMATORY CHANGES Lymphocyte and plasmacell infiltration involving in varying degree thewhole depth of the mucosa, but the pyloricglands remaining intact: these changes wereassociated with excess secretion of acid and pepsinand the development of duodenal ulcer.

2 EPITHELIAL CHANGES Increase in connective tissue,degeneration of the pyloric glands and metaplasia:these changes present a broad spectrum, namely,progressive degeneration of the pyloric glands,patchy metaplasia; complete disappearance of thepyloric glands and complete intestinalization andthinning of the mucosa with fibrosis of the sub-mucosa. These changes were usually associated withhyposecretion of acid and pepsin and gastric ulcer.Cases with complete achlorhydria usually show anantrum with complete metaplasia.

There is both in the duodenal and gastric ulcersa wide range of histological appearances of theantral mucosa, corresponding to those given in theclassical accounts, with a high proportion of cases

showing concomitant ulceration. This stronglysuggests that the two groups correspond to an earlyand late stage of what is essentially one continuouspathological process, proceeding with increasingseverity over the years. The body mucosa appears tohave a high degree of resistance, it appears normalin cases of duodenal ulcer, and shows only patchysuperficial inflammatory reaction in cases of gastriculcer. There is no apparent decrease in its parietaland peptic cell population. The area showing theweakest resistance appears to be the zones formingthe boundary between the two cellular populations.Three such zones are to be found: (1) the boundarybetween the cardiac glands and the body mucosa;(2) the boundary between body mucosa and antrum;and (3) the boundary between antrum andduodenum.The early changes consist of the development of

inflammation and ulceration of the duodenal mucosaadjacent to the sphincter. The severity of the diseasemay not for many reasons be so severe that antralresistance is completely broken down, but in somecases there is deterioration to a point beyond whichno repair is possible and metaplasia occurs. Chroniculceration will occur at a site nearest the acid-secreting body mucosa.As a result the inflammatory changes are not found

in the duodenum as a whole and stop abruptly at theproximal boundary of the antrum, and the chroniculceration when found is always at the distal marginof the proximal and distal boundaries. Thesefindings lend strong support to the idea that the acidsecretion from the body mucosa is a major factor inthe production of the inflammatory changes.

It is possible that the factor of stasis plays a part,but this is difficult to assess. It may be significantthat when organic obstruction is present, theinflammatory changes are most marked distal to theobstruction, which would suggest that the factor isnot ofgreat consequence.

It seems likely that the hyposecretion found in thelater stages of the disease results more fromdegeneration of the antral mucosa than injury tothe body mucosa (for an opposing view see Marksand Shay, 1959). Although degenerative changesare found in the region of the ulcer, the body mucosaappears to be intact when sufficient of the specimenis present for adequate histological examination. Inthe material examined here the body mucosa ingastric ulcer cases did not differ in structure from themucosa in duodenal ulcer cases, and except for agastritis of patchy and variable character, therewas little difference from the normal. It was alsoof interest to note that in the cases in which a fullfractional gastric analysis had been done, it wasunusual for there to be no response to ordinary

506



j.com/


ctober 1967. Dow

nloaded from

http://gut.bmj.com/

The antrum in patients with duodenal and gastric ulcers 507

doses of histamine, and it seems likely that indoubtful cases a more marked response might havebeen forthcoming had the dose of histamine beenincreased. In fact in most cases the response wasmarked. This corresponds to the results of otherauthors who have investigated the hypochlorhydriafollowing gastric ulcer (James and Pickering, 1949;Watkinson and James, 1951). It indicates that mostof the body glands are still functioning. The degene-ration of the antral mucosa must produce a decreasein its endocrine activity and in the late stages allactivity would be destroyed. Also the submucosalfibrosis may well interfere with innervation. It seemsunlikely that abnormal production of hormone isresponsible for the development of a gastric ulcer(for an opposing view see Dragstedt et al., 1954).The measurement of the total extent of the gastric

mucosa in stomachs with and without ulcerationshows that the rather larger size of the stomachswith ulcer is the result of increased size of theantrum, which is present in both duodenal andgastric ulcer cases but is more marked in gastriculcer. Although some cases of gastric ulcer showextensive pylorization of the body mucosa near tothe boundary, this does not prevent the boundaryfrom being mapped out with reasonable accuracy.The enlarged antrum in these cases may explain theleaving behind of appreciable amounts of antrumafter partial gastrectomy in some cases. It is ofinterest to consider that from the earliest days ofgastric surgery, a less radical operation commonlypractised for the treatment of gastric ulcer would beunsuitable for the treatment of a duodenal ulcer.This is very likely due to the marked changes foundin the antral mucosa in cases of gastric ulcer, that is,degeneration of pyloric glands, metaplasia, andatrophy. It is reasonable to assume that these changeshave affected the secretory activity of the antrum,reducing or destroying it. The unsuccessful resultsin duodenal ulcer cases are due to the leaving behindof a more or less intact antral mucosa with thepersistence of hypersecretion, and tendency to recur-rent ulceration. A patient with a partial gastrectomyand recurrent ulceration often shows that someantral mucosa has been left behind (Dean and Mason,1964).

It seems from the findings here that increased sizeis in some way connected with more severe grades ofdegeneration of the antral mucosa and eventualhyposecretion. Before more concrete suggestions canbe made as to the significance of these anatomicalfindings, much more will have to be known about thephysiology of gastric secretion in man.

SUMMARY

Seventy-five surgical specimens of duodenal ulcer

and 40 of gastric ulcer were sectioned and gradedaccording to the severity of the inflammatorychanges in the antral mucosa. The main changeswere found only in the antrum, and were primaryand independent of the ulcer. Duodenal ulcerationwas associated with mild inflammation, and gastriculceration with marked changes and degenerationwith metaplasia. The degeneration was accompaniedby diminution in gastric secretion. The size of theantrum in cases of ulcer was found to be increasedwhen compared with that in 16 normal stomachs.The possible significance of this finding is discussed.

We are indebted to the late Professor H. A. Magnus forreading the manuscript and for his valuable advice andcriticism. The late Mr. Ian Orr, F.R.C.S., Royal Infirmary,Preston, provided most of the surgical specimens. We wishto thank Mr. Ramsden, of the Royal Infirmary, Preston,and Mr. Molyneux, of the Royal Infirmary, Leigh, whoprovided the photographs, and Mrs. P. Rimmer for thearduous job of section cutting, Dr. Woodcock of theGeneral Infirmary, Bolton, who provided many of thepost-mortem specimens, together with other members ofthe Wigan Infirmary staff. This investigation wassupported by a financial grant from the Regional HospitalBoard, Manchester.

REFERENCES

Dean, A. C. B., and Mason, M. K. (1964). The distribution of pyloricmucosa in partial gastrectomy specimens. Gut, 5, 64-67.

Dragstedt, L. R., Oberhelman, H. A. Jnr., Evans, S. 0., and Rigler,S. P. (1954). Antrum hyperfunction and gastric ulcer. Ann.Surg., 140, 396-404.

Du Plessis, D. J. (1960). Some aspects of the pathogenesis and surgicalmanagement of peptic ulcers. S. Air. med. J., 34, 101-108.(1963). The importance of the pyloric antrum in peptic ulcera-tion. S. AJr. J. Surg., 1, 3.(1965). Pathogenesis of gastric ulceration. Lancet, 1, 974-978.

Faber, K. (1927). Schorstein lecture on chronic gastritis: its relationto achylia and ulcer. Ibid., 2, 901-907.(1935). Gastritis and its Consequences. Oxford University Press,London.

Gilmour, J. (1961). The surgical aspects of gastritis and its role ingastric pathology. Brit. J. Surg., 49, 278-288.

Hebbel, R. (1943). Chronic gastritis: its relation to gastric and duo-denal ulcer and to gastric carcinoma. Amer. J. Path., 19, 43-71.

Hurst, A. F., and Stewart, M. J. (1929). Gastric and Duodenal Ulcer.Oxford University Press, London.

Johnson D. H. (1955). The special significance of concomitant gastricand duodenal ulcers. Lancet, 1, 266-270.

James, A. H., and Pickering, G. W. (1949). The role of gastric acidityin the pathogenesis of peptic ulcer. Clin. Sci., 8, 181-210.

Konjetzny, G. E. (1928). Die Entzundung des Magens. In Handbuchder speziellen Pathologischen Anatomie und Histologie, editedby F. Henke and D. Lubarsch, Bd. IV, Th. 2. Springer, Berlin.

Landboe-Christiensen (1944). Extent of the pylorus zone in the humanstomach. Acta path. microbiol. scand., suppl., 54, 671.

Magnus, H. A. (1946). The pathology of simple gastritis. J. Path.Bact., 58, 431-439.

(1952). Gastritis. In Modern Trends in Gastro-enterology, editedby F. A. Jones, pp. 323-351. Butterworth, London.(1954). The pathology of peptic ulceration. Postgrad. med. J.,30, 131-136.and Rodgers H. W. (1938). The mucosa of the body of thestomach in chronic gastro-duodenal ulceration. St. Bart.Hosp. Rep., 71, 129-140.

Marks, I. N., and Drysdale, D. A. (1957). A modification of Zimmer-man's method for differential staining of gastric mucosa.Stain Technol., 32, 48.



j.com/


ctober 1967. Dow

nloaded from

http://gut.bmj.com/

508 J. Schrager, R. Spink, and S. Mitra

Marks, I. N. and Shay, H. (1959). Observations on the patho-genesis of gastric ulcer. Lancet, 1. 1 107-1 111.

Meyers, W. C. (1948). A study of gastric mucosa in various diseasesaffecting the upper part of the gastro-intestinal tract. Gastro-enterology, 10, 923-938.

Oi, M., Oshida, K., and Sugimura S. (1959). The location ofgastric ulcer. Ibid., 36, 45-56.

Palmer, E. D. (1954). Gastritis: a revaluation. Medicine (Baltimore),33, 199-290.

Papayannopoulos, G., Dedes, D., Kasralis, E., and Arealis, E. (1957).A study of antral gastritis in the resected stomachs. Gastro-enterology, 32, 428-433.

Radasch, H. E. (1921). The distribution of acid cells of the stomach.Int. J. Gastroent. 1, 24-34.

Stewart, M. J. (1922). The healing of gastric ulcer. Brit. med. J.,2, 1164-1166.

(1923a). The pathology ofgastric ulcer. Ibid., 2,955-957, 1021-1025.(1923b). The morbid anatomy of gastric and duodenal ulcer.Int. Clin., ser. 33, 4, 1-13.

Tanner, N. C. (1954). Surgery of peptic ulceration and its complica-tions. Postgrad. med. J., 30, 448-465.

Thompson, H. (1959). Gastritis in partial gastrectomy specimens.Gastroenterology, 36, 861-876.

Watkinson, G. (1960). The incidence of chronic peptic ulcer foundat necroscopy. A study of 20,000 examinations performed inLeeds in 1930-49, and in England and Scotland in 1956. Gut,1, 14-30.and James, A. H. (1951). Twenty-four hour gastric analysisin patients with histamine achlorhydria. Clin. Sci.,10, 255-266.

Weisberg, H., and Glass, G. J. B. (1963). Coexisting gastric andduodenal ulcers: a review. Amer. J. dig. Dis., 8, 992-1007.



j.com/


ctober 1967. Dow

nloaded from

http://gut.bmj.com/

the antrum in patients with duodenal gastric ulcersj. schrager, r. spink, ands. mitra stomachs,...

Documents