Debatable points in the anatomy of the lower oesophagus · barium meal examination. In adult...

Thorax; (1966), 21,487.

Debatable points in the anatomy of the loweroesophagus

G. W. FRIEDLAND', D. H. MELCHER, F. R. BERRIDGE,AND G. A. GRESHAM

From the Department of Radiodiagnosis, Addenbrooke's Hospital, Cambridge,and the Pathology Department, University of Cambridge

Although there are a number of accounts of theanatomy of the distal oesophagus in the literature,none is definitive. The arrangement of the musclefibres, the structure and nature of the sphinctermechanism, the mode of attachment of themucosa to the muscle, the cause of the 'mucosaljunction' ring, the factors which keep the stomachin the abdomen, and the anatomical criteria of asliding hiatal hernia are all still sub judice. Forthis reason, a careful anatomical study was carriedout on 132 post-mortem specimens of oesophagus,stomach, and diaphragm. One hundred andtwenty-five specimens were adjudged normal.There were seven cases of hiatal hernia, one ofwhich had associated columnar-cell lining of theoesophagus and Barrett's ulcers. A wide spectrumof ages was represented, ranging from new-bornto 102 years.In this paper, certain practical aspects of the

anatomy of the distal oesophagus are discussedand a number of new findings are presented. Noattempt will be made to review the literature.

MUSCLES OF THE OESOPHAGUS

The muscularis propria of the oesophagus consistsof two principal muscle coats, an inner transverseand outer longitudinal layer. Additional musclebundles which occur in this region are the slingfibres of the stomach, the 'constrictor cardiae',and the inconstant bracket fibres.

THE TRANSVERSE AND LONGITUDINAL MUSCLELAYERS The transverse muscle layer is really areticular layer, for each muscle bundle both con-tributes to and receives from adjacent musclebundles (Figs la and b). At irregular intervalsthe transverse layer contributes muscle bundles to

'Present address: The Hospital for Sick Children. Great OrmondStreet. London, W.C. 1. and University College Hospital, London,W.C.1

the longitudinal layer (Fig. 2). This anatomicalcontinuity may account for the fact that these twolayers contract simultaneously in living persons.

THE SLING FIBRES OF THE STOMACH The sling fibresof the stomach hook around the notch betweenthe lower end of the oesophagus on its left lateralaspect and the fundus of the stomach internal tothe inner transverse layer (Figs 3 and 4) and thenpass downwards, anteriorly and posteriorly,parallel to the lesser curve of the stomach. Asstressed by Botha (1962), they are a reliable guideto the oesophago-gastric junction. This relation-ship between the sling fibres of the stomach andthe oesophago-gastric junction is of considerablepractical importance because these fibres may berecognized in living persons as a notch during abarium meal examination. In adult patients anabnormal location of this notch enables a confi-dent diagnosis of sliding hiatal hernia to be made.This observation will be discussed in greater detailin a further paper.

THE CONSTRICTOR CARDIAE The constrictor cardiaeis a small muscle band about the same size as atransverse muscle bundle, lying internal to theinner muscle layer, which completely encirclesthe lower end of the oesophagus. It runs from theleft lateral aspect of the oesophagus above theupper limit of the sling fibres obliquely down-wards and to the right, crossing over the trans-verse muscle bundles, and, about 1-2 cm. lower,changing its direction again and running parallelto the transverse muscle fibres of the lesser curveof the stomach (Fig. 4). This band was alwaysfound when searched for.

THE BRACKET FIBRES Internal to the transversemuscle layer the thin, branching, oblique bundlesof the bracket fibres (Laimer's fibres) were foundin about one-sixth of cases (Fig. la).

487

copyright. on M

arch 27, 2020 by guest. Protected by

http://thorax.bmj.com

/T

horax: first published as 10.1136/thx.21.6.487 on 1 Novem

ber 1966. Dow

nloaded from

http://thorax.bmj.com/

G. W. Friedland, D. H. Melcher, F. R. Berridge, and G. A. Gresham

(bb

FIG. 1. The transverse muscle layer of the oesophagus and the bracket fibres.(a) Specimen of lower oesophagus opened longitudinally. The mucosa, the con-nective tissue between some of the transverse muscle bundles, and part of thelongitudinal muscle layer have been removed to demonstrate that the transversemuscle layer is a reticular layer. Bracket fibres are visible in the upper part ofthe specimen on the viewer's left. (b) Close-up view of the same specimenshowing the reticular arrangement of the muscle bundles.

488 ,

copyright. on M



/T


ber 1966. Dow

nloaded from


Debatable points in the anatomy of the lower oesophagus

FIG. 2. The lower oesophagus has been everted and the mucosa and part ofthe transverse muscle layer removed. A muscle bundle of the transversemuscle layer on the viewer's right joins the longitudinal muscle layer.

FIG. 3. Coronal section ofthe lower oesophagus and upper

stomach showing the posterior parts of these structures.The mucosa has been removed. The sling fibres of thestomach lie in the notch between the lower end of theoesophagus and the fundus of the stomach. A partiallycontracted inferior oesophageal sphincter is present in thedistal oesophagus.

FIG. 4. The lower oesophagus and upper stomach havebeen everted and the mucosa removed. The lesser curve ofthe stomach lies on the viewer's left. Transverse musclebundles are present in the lower oesophagus and the lesser

curve of the stomach. The anterior band of the sling fibresruns internal to the transverse muscle layer parallel to thelesser curve of the stomach. Above the sling fibres andinternal to the transverse muscle layer is the constrictorcardiae, which runs obliquely d9wnwards across thetransverse muscle layer towards the lesser curve of thestomach.

489

copyright. on M



/T


ber 1966. Dow

nloaded from



SPHINCTERS OF THE DISTAL OESOPHAGUS

An area of muscular contraction is recognized bythickening of the muscularis propria and themuscularis mucosae, large longitudinal mucosalfolds, and narrowing or obliteration of the lumen.According to these criteria, no evidence of

muscular contraction in the oesophagus could bedetected in just over one-half of the specimens,but two distinctly different parts of the oesophaguscould be distinguished-namely, a long, proximal,tubular and a short, distal, fusiform portion.Radiographs of these specimens confirmed thatthe proximal tubular oesophagus expanded in afusiform manner at its lower end (Fig. 5a).

In about one-third of the cases in this seriesa localized area of contraction, 1-2 cm. above thehiatus, was found at the junction of a proximaltubular and distal fusiform part of the oeso-phagus. This is the 'inferior oesophageal sphincter'of Lerche (1950). When viewed from the stomaChthe inferior oesophagcal sphincter may be seen asa distinct narrowing of the lumen (Fig. 6). Onopening such a specimen by means of a longi-tudinal incision, one may be surprised to discoverthat a considerable narrowing of the lumen wascaused by a now insignificant-looking ridge (Figs3 and 7). There are two important reasons forthis apparent paradox: (1) For every 10% in-crease or decrease in a circumference of a circle,the area of the circle increases or decreases by

FIG. 6. Same specimen as in Fig. 5b. Lower oesophagutsviewed from the stomach. The inferior oesophagealsphincter causes a narrowing of the lumen above therelaxed vestibule.

21% respectively. (2) A relatively small muscularcontraction causes the oesophageal mucosa toform large, longitudinal fo:ds, and it is these foldswhich eventually narrow or obliterate the lumen.Hence, areas of contraction are best judged bymeans of transverse sections (Figs 8a, b, and c).Radiographs of these barium-distended specimensrevealed that the inferior oesophageal sphinctercauses a cons:riction in the form of a ring with arounded margin at the junction of the tubular andfusiform portions of the oesophagus (Fig. Sb).

il I (b) VC

FIG. 5. Specimens of oesophagus and stomach distended with barium. (a) Inferior oesophageal sphincter andvestibule relaxed. The proximal tubular oesophagus expands in a fusiform manner at its lower end. Thetransverse mucosal fold between the lower end of the oesophagus and the fundus of the stomach forms a

ledge-like fi!ling defect. (b) Inferior oesophageal sphincter partially contracted and vestibule relaxed. Thesphincter causes a contriction in the form of a ring with a rounded margin. Note the filling defect caused bythe transverse mucosal fold. (c) Inferior oesophageal sphincter and vestibule contracted. Barium caught in thegrooves between the longitudinal folds in the contracted vestibule forms longitudinal linear shadows.

490

copyright. on M



/T


ber 1966. Dow

nloaded from



~~~~-'S.-"

FIG. 7. (Above) Longitudinal section through inferior oesophageal sphincter. (Below) Histological section ofsame specimen showing thickening of the muscularis propria.

The term 'vestibule' is used to denote the poten-tially more distensible portion of the terminaloesophagus between the inferior oesophagealsphincter and the sling fibres of the stomach, orbetween the junction of the tubular and fusiformdistal oesophagus and the sling fibres of thestomach.Both the inferior oesophageal sphincter and the

vestibule were found to be contracted in approxi-mately one-sixth of the cases, i.e., the transverseand longitudinal muscle layer and the muscularismucosae were thickened between the inferioroesophageal sphincter and the sling fibres of thestomach, and prominent mucosal folds occluded

the lumen. Contraction of the inferior oesopha-geal sphincter and the vestibule was recognizedon radiographs of the specimens by a long narrowsegment at the distal end of the oesophagus inwhich the barium caught in the grooves betweenthe longitudinal mucosal folds was seen as longitu-dinal linear shadows, usually three in number(Fig. Sc).

Berridge (1961) has described three similarradiographic appearances seen during cinefluoro-graphic studies of the distal oesophagus in adultpatients. The sequence of events which heobserved was as follows: (1) The tubular oeso-phagus was seen to become continuous with a

491

-, !I.. -

4w .0 .40 A. ILI...

copyright. on M



/T


ber 1966. Dow

nloaded from


(h)FtIG. 8

copyright. on M



/T


ber 1966. Dow

nloaded from



(c)FIG. 8. Same specimen as in Fig. 5b. Transverse histological sections (a) above inferior oeso-phageal sphincter, (b) at inferior oesophageal sphincter; thickening of both layers of themuscularis propria and the muscularis mucosae, some longitudinal folding of the mucosa, andnarrowing of the lumen; (c) below inferior oesophageal sphincter. The inferior oesophagealsphincter in this specimen is only moderately contracted.

dilated fusiform distal end. No constriction wasseen between the tubular and fusiform portions.(2) A ring appeared between the tubular and fusi-form segments. (3) The fusiform segment con-tracted, becoming narrow and tubular with promi-nent longitudinal mucosal folds.

In the past, three conflicting views have beenexpressed concerning sphincters in the distaloesophagus, namely (1) that no sphincter exists,(2) that there is a short sphincter situated 1-2 cm.above the hiatus (the inferior oesophageal sphinc-ter), and (3) that there is a long sphincter extend-ing for several centimetres from the oesophago-

2Q

gastric junction proximally. In fact, all threeappearances exist and are merely different stagesof the complex action that occurs in the distaloesophagus during swallowing. The first appear-ance is due to relaxation of the sphincter and thevestibule, the second to contraction of the inferioroesophageal sphincter above a relaxed vestibule,and the third to contractions of both the sphincterand the vestibule (Fig. 9). The inferior oeso-phageal sphincter is a physiological sphincter, forthere is good anatomical and radiological evidencethat it is capable of contracting independentlyfrom the rest of the oesophagus, and, when

493

copyright. on M



/T


ber 1966. Dow

nloaded from



31 2

FIG. 9. Diagrammatic representation of the anatomy of the lower oesophagus and upper stomach showing(1) vestibule contracted, (2) vestibule relaxed, (3) inferior oesophageal sphincter contracted, vestibule relaxed.m, mucosa; mj., mucosal junction; t.m.f., transverse mucosal fold; t.l., transverse muscle layer; U.1., longi-tudinal muscle layer; i.o.s., inferior oesophageal sphincter; s.f., sling fibres; p.o.m., phreno-oesophageatmembrane; h, hiatal margin.

relaxed, no trace of it can be found. Similarly,the vestibule when contracted forms a sphincter.Radiological studies in the living indicate thatthese two sphincters, possibly reinforced by a

third, the sling fibres, act together in regulatingflow through the distal oesophagus into thestomach and in closing it. This will be discussedin a further paper.

THE MUCOSA OF THE OESOPHAGUS

THE MUCOSAL JUNCTION The junction between thesquamous epithelium of the oesophagus and thecolumnar epithelium is clearly visible to the nakedeye as an irregular line, the well-known Z-line(Fig. 10). This mucosal junction is variable inposition and length. So great is this variation thatthe mucosal junction is no guide to the positionof the oesophago-gastric junction.A mixture of gelatin, barium, and India ink

was injected into the submucosa of the oesophagusand allowed to gravitate downwards. In 19 cases

where it had been injected in the correct planethe mixture was arrested at the mucosal junction.This finding indicates that the mucosa is firmlyattached to the muscle layers at the mucosaljunction.

LONGITUDINAL MUCOSAL FOLDS Both in the regionslined by squamous and by columnar epitheliumthe mucosal folds of the oesophagus are longitu-dinal in areas of muscular contraction and whenthe oesophagus is relaxed and not distended(Fig. 10). On a radiograph of a specimen, oeso-phageal squamous and columnar epithelium areindistinguishable.The mucosa is attached more firmly to the

underlying muscle layers in the grooves betweenthe longitudinal mucosal folds. When the mucosais stripped off the muscle layer, greater resistanceto stripping is encountered along the longitudinalgrooves. Furthermore, a mixture of gelatin, Indiaink, and barium injected into the submucosa tendsto run longitudinally for several centimetres along

494

copyright. on M



/T


ber 1966. Dow

nloaded from



wall of the oesophagus is formed by the mucosa.Remove this mucous membrane plug, and agaping hole remains. The mucosa and musclethus act in unison in closing the lumen, the speedof closure and mechanical effectiveness of muscu-lar contraction being enhanced by the mucosa.As the diameter of the distended, relaxed vesti-

bule exceeds that of the rest of the oesophagus,it has a greater area of mucosa per unit length,so particularly fleshy folds form when it contracts.

TRANSVERSE MUCOSAL FOLDS When specimens inwhich the vestibule is relaxed are distended, analmost horizontal mucosal fold forms at the levelof the sling fibres of the stomach (Fig. 11). Thisfold lies below the mucosal junction in themajority of cases. In barium-filled specimens, thetransverse fold forms a ledge-like ring at theoesophago-gastric junction, usually most promi-nent and sometimes only on the left (Figs 5a andb). Radiological studies in living persons haveshown that a similar ring may form at the samelevel as the notch caused by the sling fibres of the

FIG. 10. Fresh specimen of lower oesophagus and upperstomach opened longitudinally. The mucosal junction lieswell above the oesophago-gastric junction. The longestdigitation of squamous epithelium measured 15 cm.Longitudinal mucosal folds of the oesophagus are presentboth in the regions lined by squamous and by columnarepithelium. Gastric columnar epithelium forms rugae.

a fold between two grooves before spreadinglaterally. This attachment of the mucosa to themuscle along the longitudinal grooves is not dueto the deep oesophageal glands and ducts, as oftquoted in the literature, for histological examina-tion reveals that the glands and ducts may lie bothin the grooves and in the folds.

Serial sections of distended gullets show thatthe mucosa is flattened in areas where the musclesare relaxed and thrown into longitudinal folds inareas of muscular contraction. The attachment ofthe mucosa to the muscle layer at fixed pointsalong the circumference of the wall of the oeso-phagus encourages the formation of longitudinalfolds as the muscularis propria contracts. Arelatively small change in the lumen of the musclelayer results in a large change in the lumen of theoesophagus, because the longitudinal folds rapidlyincrease in size as the muscular ring narrows.These folds produce a stellate oesophageal lumenon cross-section. A further decrease in the dia- FIG. 11. Specimen oflower oesophagus and upper stomachmeter of the muscular ring now obliterates this opened longitudinally. Part of the mucosa has beenstellate lumen by increasing the size of the folds removed demonstrating the relation between the slingfibresand also by pressing the folds together. At this of the stomach, the transverse mucosal fold, and thestage, a surprisingly substantial proportion of the mucosaljunction.

495

copyright. on M



/T


ber 1966. Dow

nloaded from



stomach in patients with a hiatal hernia. Thissubject will be discussed more fully in a furtherpaper.Frequent reference is made in the literature to

a 'mucosal junction ring'. All illustrations showthis ring to have a perfectly straight outline. Ithas long been accepted that the mucosal junctionis not a straight line but has digitations of varyinglength. If this ring were to follow the line of themucosal junction it, too, should have a jaggedoutline on radiographs. This is not the case.The. transverse mucosal fold lies at the oeso-

phago-gastric junction and not in the oesophagus.When oesophageal squamous epithelium is grippedwith a pair of forceps and raised, it never formsa transverse mucosal fold in a fresh specimenirrespective of the mode of traction. A flat,triangular fold always forms with its long baserunning parallel to the longitudinal folds of theoesophagus and its apex at the point at which thetraction is being applied. Gastric mucosa, on theother hand, either assumes a stellate configuration,or the base on the triangle may be dependent onthe direction of the gripping force.

Histological sections of the transverse mucosalfold show that in most cases it is covered bycolumnar epithelium. Rarely, a digitation of thesquamous epithelium may run on to the base ofthe fold and a histological section taken at thispoint shows part or all of the upper surface of thefold to be covered by squamous epithelium andthe rest by columnar epithelium. In only one casedid a digitation of squamous epithelium cover thewhole of the fold. The site at which the sectionis taken thus determines the-histological appear-ance. Similarly, in living persons, the type ofepithelium found on histological examination willdepend on the exact point in the circumferenceof the oesophagus at the particular level at whichthe biopsy is taken; this is particularly true in per-sons who have a Z-line with long digitations.

ANATOMICAL STRUCTURES WHICH KEEP THESTOMACH IN THE ABDOMEN

The oesophagus is attached-to the diaphragm bya fibro-elastic membrane Io%wn as the phreno-oesophageal membrane or ligarnent. The normaloesophagus in an adult patient can slide up anddown in the hiatus over a length of about 2 cm.When the nerves, vessels, peritoneum, and fatattached to the stomach are cut it is not possible,by pulling on the oesophagus, to displace the slingfibres of the stomach above the hiatus. However,once the attachments of the phreno-oesophageal

membrane to the oesophagus are cut, the slingfibres of the stomach can be pulled into the chest.

After arising from the under surface of thediaphragm the phreno-oesophageal membranedivides into two limbs. The upper limb passesthrough the hiatus, where it is reinforced by thesupradiaphragmatic fascia before inserting bymeans of multiple slips into the muscles of theoesophagus. The lower limb is inserted into thelower oesophagus and upper stomach below thehiatus. The membrane allows the oesophagus toslide up and down in the hiatus rather like atendon in a tendon sheath.

FIG. 12. Specimen of diaphragm, phreno-oesophageaimembrane, and lower oesophagus. The structures visibleare the upper surface of the diaphragm immediatelyadjacent to the hiatus in the lower part of the photograph,the upper limb of the phreno-oesophageal membrane as awhitish area, and the external surface of the oesophagus.The line of insertion of the upper limb of the phreno-oesophageal membrane is very irregular.

The line of insertion is irregular. Some digita-tions of the membrane may extend for up to 2 cm.further along the wall than adjacent digitations(Fig. 12). Contrary to some current views, thereis no anatomical evidence that the inferior oeso-phageal sphincter is not a sphincter at all but iscaused by traction on the wall of the oesophagusby the upper limb of the phreno-oesophagealmembrane, for the digitations of the membrane

496

copyright. on M



/T


ber 1966. Dow

nloaded from



are inserted below, into, or above the sphincter.Furthermore, the behaviour of the sphincter inliving persons indicates that it is due to a localizedcontraction of the muscle layers. Similarly, thering caused by the transverse mucosal fold is notdue to traction of the lower limb of the phreno-oesophageal membrane. As a corollary, the -upperand lower limbs of the phreno-oesophageal mem-brane do not define the limits of the vestibule.

HIATAL HERNIA

Seven cases of hiatal hernia were encountered inthe series, one of which had an oesophagus almostentirely lined by columnar epithelium and severalpenetrating ulcers on the postero-lateral wall ofthe oesophagus. Only the case of columnar-celllining of the oesophagus was diagnosed duringlife.Common to all the hernia cases was the presence

of a thin and lengthened phreno-oesophagealmembrane. Indeed, in some the phreno-oesopha-geal membrane could not be identified. In all, thesling fibres of the stomach could be drawn wellabove the hiatus by pulling on the oesophagus.On superficial inspection, the oesophagus in the

clinically unsuspected cases did not appear grosslyabnormal, but on close inspection small super-ficial mucosal scars could be detected in some.Difficulty was experienced in stripping themucosa off the muscle layers, and on histologicalexamination submucous fibrosis was present in all.It is likely that these unsuspected cases wouldhave been overlooked during routine post-mortemexamination. All specimens with a hiatal hernia(excluding the case of columnar-cell lining of theoesophagus) had a relaxed vestibule above thesling fibres and the transverse mucosal fold, andin none was the inferior oesophageal sphinctercontracted.The incidence of hiatal hernia in this series was

lower than is commonly reported by radiologists,possibly because minimal hiatal herniation shownradiologically is not associated with appreciableweakening of the phreno-oesophageal membrane.Unusual prominence of the sling fibres of thestomach was observed in the cases of unsuspectedhiatal hernia but not in the case with an oeso-phagus lined by columnar epithelium. As the slingfibres of the stomach are easily identified atnecropsy, a simple post-mortem test for hiatalhernia is now available. If the phreno-oesopha-geal membrane is left undisturbed and the slingfibres of the stomach can be pulled above thehiatus, a hiatal hernia is present.

SUMMARY

Some anatomical findings in 132 post-mortemspecimens of oesophagus, stomach, and diaphragmare discussed. There were seven cases of hiatalhernia, one of which had most of the oesophaguslined by columnar epithelium.The transverse muscle layer of the muscularis

propria is a reticular layer and contributes musclebundles to the longitudinal layer. Both musclelayers contract together, simultaneously shorten-ing and narrowing the contracted segment andcausing the mucosa to form large longitudinalfolds. In the notch between the left lateral aspectof the lower end of the oesophagus and the fundusof the stomach lie the sling fibres of the stomach.As they are a reliable guide to the oesophago-gastric junction, displacement of these fibres abovethe hiatus enables a diagnosis of hiatal hernia tobe made post mnortem and on barium studies inliving persons.

In just over half the cases there was no evidenceof muscular contraction in the oesophagus, andthe tubular oesophagus expanded in a fusiformmanner at its lower end before entering thestomach at the level of the sling fibres. Alocalized area of contraction at the upper limitof the vestibule (the inferior oesophageal sphinc-ter) was present in about one-third. The inferioroesophageal sphincter and the whole vestibulewere contracted in the remaining cases. It is thusclear why there has been so much controversyin the literature over the sphincters in the loweroesophagus. Each of these appearances is merelya phase of the normal action seen in the distalend of the oesophagus during swallowing (Fig. 9).The variability of the mucosal junction has been

confirmed. Longitudinal folds form in the oeso-phagus when the muscles contract, bzcause themucosa is attached to the muscle wall at fixedpoints in its circumference. These folds play animportant part in the closure of the oesophageallumen. A transverse mucosal fold forms at thesame level as the sling fibres of the stomach whenthe vestibule is relaxed and distended. This foldis not related to the mucosal junction.The oesophagus is anchored to the hiatus by the

phreno-oesophageal membrane, which permits theoesophagus to slide up and down in the hiatusand prevents the stomach from herniating into thechest. Like the mucosal j&nction, this membranehas irregular digitations and thus does not definethe limits of the vestibule or form either the in-ferior oesophageal sphincter or the mucosal foldat the level of the sling fibres of the stomach. In

497

copyright. on M



/T


ber 1966. Dow

nloaded from


G. W. Friedland, D. H. Melcher, F. R. Berridge, and G. A. G)resham

cases of sliding hiatal hernia the phreno-oesopha-geal membrane is attenuated.The way in which various structures at the distal

end of the oesophagus behave in living persons

will be fully discussed in a further paper.

We wish to thank the British Medical ResearchCouncil for their generous grant to Dr. G. W. Fried-land, who carried out the macroscopic anatomical dis-sections, Dr. Ifor Williams for suggesting some of the

experiments on the mucosa, and the Department ofPhotography, Addenbrooke's Hospital, for the photo-graphs.

REFERENCES

Berridge, F. R. (1961). The mechanism at the cardia: A symposium.III. Radiological aspects. Brit. J. Radiol., 34, 487.

Botha, G. S. Muller (1962). The Gastro-oesophageal Junction, p. 5.Churchill, London.

Lerche, W. (1950). The Esophagus and Pharynx in Action, p. 55.Thomas, Springfield, Illinois.

498

copyright. on M



/T


ber 1966. Dow

nloaded from


Debatable points in the anatomy of the lower oesophagus · barium meal examination. In adult...

Documents

Transcript of Debatable points in the anatomy of the lower oesophagus · barium meal examination. In adult...