31st annualmeeting ofthe british society of gastroenterology

Gut, 1970, 11, 1055-1067

31st Annual Meeting of the British Societyof Gastroenterology

The 31st Annual Meeting of the British Society of Gastroenterology was held in London from5 to 7 November, 1970, at the Royal College of Physicians, under the Presidency of Dr NelsonCoghill. An excellent dinner was held at the Connaught Rooms, and the occasion was taken topresent Dr T. C. Hunt with a specially bound copy of the Selected writings of Sir ArthurHurst. The President paid tribute to the Honorary Secretary, Dr J. E. Lennard-Jones, whohad completed his term of office. During his tenure of five years, the Society has more thandoubled in size, and Dr Lennard-Jones has been responsible for overcoming the many problemsarising from this period of rapid growth-a period which has demonstrated the vitality andscientific activity of present-day gastroenterology.

In addition to the scientific papers, of which theabstracts are printed in this number, the annualSir Arthur Hurst Memorial Lecture was given byDr Kurt J. Isselbacher on 'Intestinal absorption-structural and functional aspects'.An interesting innovation this year was the

introduction of educational symposia, in additionto the scientific papers. Professor A. W. Kay actedas Chairman for the symposium on 'Coordinationof gastroduodenal function', with E. L. Blair,I. E. Gillespie, and K. Wormsley taking part.Dr Roger Williams arranged a symposium onthe 'Treatment of fulminant hepatic failure', withcontributions from Dr P. T. Flute, Dr J. Winch,and from himself. In addition, a quite outstandingprogress report on 'Immunology and the gut'was given by Professor J. R. Hobbs, and it will longremain inthememory ofall those whowere present.

There were innovations also in relation to thedemonstrations. There were two tape and slidelectures from the American GastroenterologicalAssociation illustrated lecture series, both byDr A. F. Hofmann, one on 'Fat digestion' andthe other on 'The enterohepatic circulation ofbile acids'. A slide quiz had been arranged byProfessor B. N. Brooke and Dr G. Thompson,and a further slide-tape demonstration, 'Liverbiopsy in the diagnosis of acute jaundice', byDr P. J. Scheuer. As a last minute addition to theprogramme there was a colour film entitled'Duodenoscopy', and this was presented by DrW. M. M. Classen from Erlanger, West Germany.This was a brilliant film of great interest anddelightfully commented on by Dr Classen inperfect English.

In the Library, the Harveian Librarian, DrC. E. Newman, and his staff had arranged ahistorical demonstration, entitled 'Landmarks in

gastroenterology'. In this demonstration manywell known names in gastroenterology were to befound, with the descriptions of diseases andnatural phenomena being set out in the originaltext. Here one could browse over the books ofthose who had made such major contributions inthe past: Van Helmont, with his description ofdigestion as a process of fermentation; Wharton'sdescription of the submaxillary salivary gland;William Prout who first proved that gastric juicecontained free hydrochloric acid; Cruvheilier'sfine illustrations of gross pathology; the originalaccount of acute ulceration of the duodenum incases of burns by T. B. Curling-being but a fewof the many exhibits to be seen, and a demonstra-tion of the most remarkable collection ofhistoricalbooks which is in the possession of the RoyalCollege of Physicians of London.On display and on sale was the Selected

writings of Sir Arthur Hurst, the founder of theBritish Society of Gastroenterology. This publica-tion brings together much of the work of thisgreat twentieth century gastroenterologist wholaid the foundations of so much subsequent work.The Society will long be indebted to Dr T. C.Hunt for bringing together these historic paperswhich can all be re-read for the first time withinterest and advantage by all working in thisfield. Medical history, presented in this way, illus-trates so well the lessons of success and failurewhich can help later generations. This is a bookwhich should be in the possession ofevery memberof the Society and in every important libraryaround the world. The publication was devisedinitially to provide a presentation to those whogave the annual Sir Arthur Hurst memoriallecture, and specially bound copies have beenprovided for this purpose.

on June 6, 2022 by guest. Protected by copyright.

http://gut.bmj.com

/G

ut: first published as 10.1136/gut.11.12.1055 on 1 Decem

ber 1970. Dow

nloaded from

http://gut.bmj.com/

1056 The British Society of Gastroenterology

THE TJSE OF GASTRIC FUNCTION TESTS BY MEMBERS OFTHE BRITISH SOCIETY OF GASTROENTEROLOGY

J. H. BARON and J. ALEXANDER WILLIAMS (Departmentsof Surgery, Royal Postgraduate Medical School,London, and the General Hospital, Birmingham)A postal questionnaire was returned by 78% of 259clinician members of the British Society of Gastro-enterology and the replies of 108 physicians and75 surgeons are reported. Their choice of gastricfunction tests and rationale for their use have beenanalysed and compared with criteria suggestedrecently (Baron, 1970).

Pentagastrin (73 %) has largely replaced older drugs(histamine 16%, histalog 8%) as the stimulant ofchoice for evoking maximal acid secretion. Potentiallydangerous insulin tests are being used in situationswhere they are unlikely to be helpful. In the assess-ment of patients with proven or suspected ulcers of thestomach and duodenum significantly fewer physiciansthan surgeons measure gastric acid, and they use thesetests less often. The reluctance of physicians to testpatients with uninvestigated dyspepsia or gastriculcer seems justified but in patients having dyspepsiawith negative x-rays, or after gastrectomy or vagotomy,the greater investigative keenness of surgeons seemscommendable. It is to be regretted that only half thesurgeons have tried to assess the completeness of theirvagotomies and in only one-third of this half is ittheir usual practice. In patients with x-ray evidence ofduodenal ulcer, it is surprising that (a) 80% of thegastroenterologists measure acid at least sometimes;(b) these tests influenced more than half the physiciansand two-thirds of the surgeons in deciding whetheroperation should be advised; (c) half the surgeonswere influenced by these tests in deciding the type ofoperation.

REFERENCE

Baron, J. H. (1970). Scand. J. Gastroent., Suppl. 6, 9.

THE RELATIONSHIP OF GASTRIC SECRETION TO ANAEMIAOVER 15 YEARS FOLLOWING VAGOTOMY AND GASTRO-

ENTEROSTOMY

E. JEAN WHELDON, C. W. VENABLES, and IVAN D. A.JOHNSTON (Department of Surgery, Royal VictoriaInfirmary, Newcastle upon Tyne) In previous studieswe have shown that anaemia is common followingvagotomy and gastroenterostomy (43.5% men and84% women). The cause of this anaemia was mainlyiron deficiency but how it occurred was not clear.A study was planned in two groups of patients to

see if the anaemia was related to changes in gastricsecretion induced by the operation. One group con-sisted of 12 patients whose haemoglobin concentra-tion was less than 10-2 g%, the second group werepatients who, at review, were not anaemic and whohad never been treated for anaemia. The gastricsecretory response of acid and pepsin to pentagastrin(6 utg/kg im) and insulin (0-2 units/kg) was measured,secretion being collected by a 'washout' techniquewhich gave an estimate of losses. The mean acidoutput for the anaemic group was 1.12 m-equiv/hrand pepsin output 0.4 mg/hr. These studies wererepeated following correction of the anaemia and irondeficiency to confirm that hypochlorhydria did not

result from the iron deficiency. After correction, meanacid output was 1-56 m-equiv/hr and pepsin outputwas 25.7 mg/hr. In the second group the mean acidoutput in response to pentagastrin was 9-62 m-equiv/hr and the pepsin output 108.8 mg/hr. The differencein acid and pepsin output between the anaemic andnon-anaemic patients was highly significant (p < 0005).Only one patient without anaemia had an acid outputwithin the range of the anaemic group.These findings suggest that anaemia following

vagotomy and gastroenterostomy is related to changesin gastric secretion and makes us question whetherthe aim of surgery should be to produce the greatestreduction in acid secretion that is possible.

EFFECT OF LUMINAL pH ON INTESTINAL ABSORPTIONOF WATER AND ELECTROLYTES

B. ROUSSEAU and G. E. SLADEN These studies wereundertaken because it is often assumed that a reduc-tion of luminal pH may be in part responsible for thefluid diarrhoea, which occurs in some patients withthe Zollinger-Ellison syndrome and with variousforms of sugar intolerance. There is evidence thatsmall reductions ofpH may profoundly affect jejunaltransport of water and sodium(McHardyand Parsons,1957), but little is known about such effects of pHin the ileum or colon. In these studies, using ratsin vivo, isotonic fluid buffered at different pHs(5 6-7 6) was placed in tied loops of distal ileum(20-30 cm) or proximal colon (5-8 cm) and left forshort absorption periods (30-45 minutes).

In the ileum, appreciable absorption of water,sodium, and chloride took place from solutions ofpH 7.6; negligible absorption of water or sodiumoccurred at pH 6.6 or 5-6. Significant chlorideabsorption occurred, however, at all pHs and this isprobably mediated by a separate anion exchangemechanism. In the colon, net absorption took placeat all the pHs studied. The optimal pH was 6-6 andthe lowest absorption rates were from solutions ofpH 5-6.

These studies show that luminal pH may greatlyaffect ileal handling of fluid and electrolytes, whereascolonic function is less affected bypH changes of thesame magnitude. Changes in luminal pH, especiallyin the small intestine, may therefore play an importantrole in the production of fluid diarrhoea.

REFERENCE

McHardy, G. J. R., and Parsons, D. S. (1957). Quart. J. exp.Physiol., 42, 33.

STUDIES ON BIDIRECTIONAL SODIUM FLUXES ACROSS THEINTESTINAL MUCOSA IN CHOLERA PATIENTS

A. H. G. LOVE, J. E. ROHDE, and N. VEALL (Departmentof Medicine, The Queen's University, Belfast, N. Ireland,Pakistan-SEATO Cholera Research Laboratory, Dacca,Pakistan, and Radioisotopes Division, MRC ClinicalResearch Centre, Harrow) The diarrhoea of cholerareflects a net difference between the amount of sodiumpassing into the gut lumen and the amount absorbedfrom the gut. Whether it is due to impaired absorptionor increased secretion has remained a matter for


http://gut.bmj.com

/G


ber 1970. Dow

nloaded from

http://gut.bmj.com/


speculation which can only be resolved bybidirectionalflux measurements.

Studies have been carried out in normal subjectsand cholera patients using whole gut steady-stateperfusion techniques. Sodium isotopes were adminis-tered orally and intravenously simultaneously.Stochastic analysis of the subsequent activity curvesin the blood and stool allowed calculations to bemade of the bidirectional fluxes across the intestinalmucosa.The results obtained show that exogenous and

endogenous sodium are handled in a similar fashionby the intestine and that sodium ion exchange occurspredominantly in the upper region of the smallintestine. Both inward and outward fluxes are con-siderably reduced in cholera but absorption isreduced to a greater extent than secretion, resultingin diarrhoea. Segmental studies further confirm thisimbalance of sodium ion fluxes and also define moreclearly the region of the intestine maximally involved.

SODIUM, POTASSIUM, AND WATER ABSORPTION BYNORMAL HUMAN RECTUM

C. J. EDMONDS (MRC Department of Clinical Research,University College Hospital Medical School, LondonWC1) It has recently been claimed that normal hu-man rectum 'fails to absorb electrolytes and water'(Devroede and Phillips, 1970). However, by a newmethod which will be described, it has been possibleto show that not only do significant ionic movementsoccur but that rectal mucosa possesses a powerfulsodium-absorbing mechanism. The unidirectionalsodium efflux rate (plasma-to-lumen flux) was smallby comparison with that of proximal colon, probablyindicating restricted permeability to passive sodiummovement. The sodium influx rate (lumen-to-plasma-flux) was considerably greater and the striking charac-teristic was that sodium absorption occurred againstconsiderable electrochemical gradients. Thus, evenwhen luminal sodium concentration was only10 m-equiv/l and the epithelial charge was 20 to 40mV(blood side positive), the sodium absorption stilloccurred. Potassium also crossed the epithelium inboth directions, zero net movement usually beingfound only when the luminal concentration exceeded40 to 50 m-equiv/l. It is unlikely that passive processesdependent upon the electrochemical gradients accountfor the observations and active secretion of potassiuminto the lumen seems likely. Water absorption wasobserved in most experiments but changing the direc-tion of net water flux by altering the osmotic gradientdid not significantly affect sodium or potassiummovements or the electrical potential difference.

Thus, although total ionic movements are less thanin proximal colon, the rectal epithelium cannot beregarded as inactive in respect to electrolyte andwater transport. The particular properties of rectalmucosa, especially the ability to absorb sodium activelyagainst large electrochemical gradients, and probablyalso to secrete potassium, are well adapted to thenormal function of the distal gut.

REFERENCE

Devroede, G. 3., and Phillips, S. F. (1970). Gut, 11, 438.

SPONTANEOUS DUODENO-COLIC FISTULA

C. B. JONES (Department of Gastroenterology, Man-chester Royal Infirmary) Duodeno-colic fistulae areuncommon. In all previously reported cases theyhave arisen secondary to some other pathology, eitherin the duodenum, colon, or elsewhere. Three identicalcases of spontaneous communication between thethird part of the duodenum and transverse colon aredescribed.The three cases described and illustrated occurred

in two females and one male. The patients were allmiddle aged and had a history of diarrhoea, generalill health, and emaciation. Following removal of thefistulae, all symptoms were relieved.

In order to find the fistula a careful search wasnecessary; indeed one patient had previously under-gone a laparotomy when the fistula had not beendetected.

Microscopy showed that all coats of the bowel wererepresented within the fistulous track. The mucousmembrane and muscular coats of the large and smallbowel continued into one another. There was noevidence of any accompanying pathological process.

In the absence of any other pathological lesions,it is postulated that there is an embryological basisfor the development of this unusual type of internalintestinal fistula.

75Se ACTIVITY IN THE DUODENAL TEST MEAL ASPIRATE:A NEW PANCREATIC FUNCTION TEST

G. R. YOUNGS, I. A. D. BOUCHIER, J. E. AGNEW, andG. E. LEVIN (Departments of Medicine, Physics andChemical Pathology, The Royal Free Hospital, London)The pancreas selectively takes up 75Se-selenomethio-nine and incorporates the isotope into secretoryproteins. In this study we have evaluated the appear-ance of 75Se in human duodenal aspirate and haveused the information as a test of pancreatic exocrinefunction. Intravenous 75Se-selenomethionine in a doseof 3 gCi/kg body weight was given 10 minutes afterdrinking a Lundh test meal and a pancreatic scanand duodenal aspiration were then performed simul-taneously. In patients not undergoing a pancreaticscan one tenth of the dose of isotope was given.In normal subjects the level of 75Se in the duodenalaspirate remained low for one hour and then roserapidly, whereas in patients with diffuse pancreaticdamage no such increase occurred. There was a signi-ficant correlation between the 75Se activity at twohours and the trypsin concentration of the aspirate.The level of 75Se activity remained low in bile andgastric juice. The results suggest that the 75Se-seleno-methionine is incorporated into the enzymes releasedin the pancreatic juice and that measurement of 75Seactivity in the duodenal aspirate is a useful index ofpancreatic function. Under certain circumstances thismay prove a simpler test of pancreatic function thanestimation of trypsin ac:ivity.

GALLBLADDER INERTIA IN ADULT COELIAC DISEASE

T. S. LOW-BEER, K. W. HEATON, and A. E. READ (Depart-ment of Medicine, University of Bristol) Since thephysiological stimulus to gallbladder contraction ischolecystokinin released from the proximal small


http://gut.bmj.com

/G


ber 1970. Dow

nloaded from

http://gut.bmj.com/


intestine in response to food, we have investigatedthis process in patients with adult coeliac disease.All patients had subtotal villous atrophy on peroralbiopsy and were not on a gluten-free diet. The tech-niques used were measurement of radioactive bilesalt turnover (Austad, Lack, and Tyor, 1967) andmetabolism (Garbutt, Heaton, Lack, and Tyor,1968) and oral cholecystography.The results showed that, compared with control

subjects, labelled taurocholate remained in the bilefor longer and was metabolized more slowly in sixout of eight patients with coeliac disease. This suggeststhat bile salts were being sequestered in an essentiallysterile part of the enterohepatic circulation, presum-ably the gallbladder. Nevertheless, injected chole-cystokinin caused a copious flow of concentrated bileinto the duodenum of intubated patients.

Oral cholecystography in 13 patients with adultcoeliac disease showed normal concentration of thecontrast mediu.n in all patients but in eight of themthere was little or no contraction in response to a fattymeal. A control series matched for age and sex showedgood contraction in 17 out of 20 subjects.We therefore suggest that in many patients with

mucosal disease of the upper small intestine synthesisor release of endogenous cholecystokinin is impaired,resulting in gallbladder inertia. This could contributeto the production of steatorrhoea.

REFERENCES

Austad, W. I., Lack, L., and Tyor, M. P. (1967). Gastroenterology,52, 638.

Garbutt, J., Heaton, K. W., Lack, L., and Tyor, M. P. (1968).(Abstr.) Gastroenterology, 54, 1234.

IMMUNOLOGICAL STUDIES OF A COMMUNITY OUTBREAKOF HEPATITIS

A. B. AJDUKIEWICZ, R. A. FOX, F. DUDLEY, S. SHERLOCK,D. DONIACH, S. DEL PRETE, and D. CONSTANTINO(Department of Medicine, Royal Free Hospital,Department of Immunology, Middlesex Hospital,Istituto di Clinica Medica 11, Milan) Serial sera wereobtained from 72 patients with infective hepatitis inTilbury. The incubation period was two to threeweeks in the majority. Threequarters of those affectedwere less than 15 years of age. The sera were negativewhen tested for the usual hepatitis-associated antigen.Using an antiserum from Milan positive results wereobtained in 39 patients. Most patients gave positivetests in the first two weeks of illness. Sera obtainedlater in the course of the illness and convalescentsera were both negative. The incidence of smoothmuscle antibody was 68 %. Higher titres tended tooccur earlier in the course of the illness and in mostof these cases tests with the Milan antiserum werepositive. The titre of smooth muscle antibody gradu-ally fell. Mitochondrial antibodies were uniformlynegative. The results using the Milan antiserum suggestthat short incubation epidemic infective hepatitis isvirologically a different illness from that associatedwith the hepatitis-associated antigen, but that thetissue antibody responses are similar to those inhepatitis associated with hepatitis-associated antigen.

A TRIAL OF AZATHIOPRINE IN PRIMARY BILIARYCIRRHOSIS

A. ROSS and s. SHERLOCK (Department of Medicine,

Royal Free Hospital, London) A controlled trial ofimmunosuppressive therapy in patients with primarybiliary cirrhosis (PBC), has been in progress for twoyears. Thirty-one patients with PBC havebeenadmittedto the trial, and randomly allocated to either atreatment group, receiving azathioprine at an initialdose of 2 mg/kg body weight, or to a control group.Twenty-two patients have been followed for longer

than one year, three deaths have occurred, and fourpatients have been withdrawn from the trial becauseof complications. The clinical course of the two groupshas been comparable, but pruritus has decreased inthe treated group. Biochemical tests were not signifi-cantly different at inclusion in the trial; in the treatedgroup there has been a significant fall in alkalinephosphatase, cholesterol, and aspartate transaminase.Serum immunoglobulin estimations showed a signifi-

cant fall in IgM levels at the start of azathioprinetreatment, but mitochondrial antibody titres didnot alter significantly between the groups. Liver histol-ogy after one year has not shown any difference be-tween the groups.

Preliminary conclusions are that azathioprineimproves aspects of biliary excretion in primarybiliary cirrhosis, but, as yet, there is little evidence ofan effect on immunology or histology.

DIAGNOSIS AND PROGNOSIS OF PRIMARY MALIGNANTTUMOURS OF THE LIVER

P. SHARPSTONE, M. R. FLEISHER, M. 0. RAKE, K. B.SHILKIN, J. W. LAWS, and ROGER WILLIAMS (Liver Unit,King's College Hospital) Forty-one patients withhistologically proven primary malignant tumours ofthe liver have been seen by the authors. Thirty-fourof the patients had hepatocellular carcinoma; threehad Kupffer-cell sarcoma; three cholangiocellularcarcinoma; and one hepatoblastoma. In 26 patients,underlying hepatic cirrhosis was present; nine ofthem had haemochromatosis. Unusual metaboliceffects-hypercalcaemia and porphyria cutanea tarda-were produced by two of the hepatocellular carci-nomas.

Scintiscanning of the liver was the most valuablediagnostic technique. Technetium 99m sulphur-colloidscans showed filling defects in all but one of the 28patients examined. 75Se-selenomethionine scansshowed uptake of the isotope in the tumour area in14 of 20 cases. All the Kupffer-cell sarcomas andcholangiocarcinomas failed to take up selenomethio-nine. Selective hepatic arteriography showed diag-nostic abnormalities-tumour circulation or avascularareas-in 15 of 21 examinations.

Thirty-four patients have died; the mean survivaltime was 5-8 months. Three of the seven survivingpatients have lived for more than two years sincediagnosis. A patient with haemochromatosis and acholangiocellular carcinoma treated with local intra-arterial 5-fluororacil infusion and with irradia-tion, a haemochromatotic with a hepatocellularcarcinoma, who had no anti-tumour treatment, and apatient with a hepatocellular carcinoma in a cirrhoticliver, treated with parenteral 5-fluororacil, arealive at 56 months, 32 months, and 28 months,respectively.

THE PATHOGENESIS OF COELIAC DISEASE

MARGOT SHINER and D. H. SHMERLING (Central Middle-sex Hospital, London) The response of the jejunal


http://gut.bmj.com

/G


ber 1970. Dow

nloaded from

http://gut.bmj.com/


epithelial cells and their subepithelial tissues to asingle dose of gluten was studied in treated coeliacchildren and in children without coeliac diseaseby light and electron microscopy and by immuno-fluorescent techniques. Serial jejunal biopsies wereobtained before gluten challenge and from two to 96hours after.The results, based on ultrastructural findings, indi-

cate that following gluten challenge in the coeliacchildren there is an early thickening of the connectivetissue in the subepithelial spaces and around thesmall blood vessels, with infiltration of mononuclearcells. This was not observed in the children withoutcoeliac disease who, in contrast to the coeliac children,showed epithelial cell changes which may be relatedto the normal process of polypeptide ingestion andabsorption.

In an attempt to demonstrate immune complexformation after gluten challenge within the subepi-thelial connective tissue of villi and crypts, fluoresceinisothiocyanate-conjugated antisera (anti-human gam-ma globulins :-anti-IgA, anti-IgG, and anti-IgM) wereused. The results obtained in the coeliac children,when compared with the non-coeliac children, weresuggestive of binding of antigen and antibody in theconnective tissue as early as 21 hours followinggluten administration. The timing coincided with theearliest reactive changes in the subepithelial layersnoted by electron microscopy.

SMALL BOWEL PERMEABILITY IN ANIMALS AND MAN

C. A. LOEHRY and D. PARRISH (Royal Victoria Hospital,Bournemouth) Previous studies on rabbits havedemonstrated that the clearance of substances fromthe blood into the small intestinal lumen is dependentdirectly on their concentration in the plasma, andinversely on their molecular size (Loehry, Axon,Hilton, Hider, and Creamer, 1970).The blood-to-lumen transfer in man of creatinine,

uric acid, urea, and vitamin B12 has been studied withintestinal perfusion through a triple-lumen tube.Both normal subjects and patients in varying degreesof renal failure have been studied, and a direct rela-tionship demonstrated between rising blood levels andintestinal loss.The differential clearance of fractions of isotopically

labelled PVP in the faeces after an intravenous dosehas been studied using a Sephadex column. A pro-gressive fall in clearance is demonstrated in fractionsof rising molecular weight.To study the methods by which substances travel

from the blood to the lumen various iron preparationshave been injected directly into the radicals of themesenteric artery in rabbits, and the mucosa has beenstained for iron. The iron enters epithelial cells nearthe tips of the villi, but not at their base, or in thecrypts. Under certain circumstances iron also passesbetween epithelial cells near the tips of the villi.

REFERENCE

Loehry, C. A., Axon, A., Hilton, P., Hider, C., and Creamer, B.(1970). Gut, 11, 6.

THE AMINO ACIDS AND PANETH CELLS OF THE SMALLINTESTINE

A. E. GENT (Introduced by B. CREAMER (St Thomas'Hospital, London) The concept of a homeostatic

mechanism for free amino acids in the small intestinewas first proposed by Nasset (1957). He showed thatconsiderable amounts of endogenous protein andamino acids enter the intestine in response to food.The purpose of the present experiments was to deter-mine the contribution made by the mucosa to thefree amino acids appearing in the intestine. Initialexperiments used pilocarpine to stimulate the Panethcells in the rat small intestine, and showed that asignificant increase in amino acid concentrationsoccurred after its use. Experiments using 65Zn indi-cated that these amino acids are produced by Panethcells.The nature of the homeostatic mechanism was

investigated by using amino acid solutions for instilla-tion experiments in rats and perfusion studies inman. The rat experiments indicate that such a homeo-static mechanism does exist, and information fromthe human studies indicates the importance ofpancreatic and Paneth cell secretion. Evidence is.presented which suggests that biliary secretion is notan important factor in this mechanism.The conclusion drawn from these results is that

Paneth cell secretion plays an important part inmaintaining the homeostatic mechanism for aminoacids in the small intestine, second only in importanceto the pancreatic contribution.

REFERENCENasset, E. S. (1957). J. Amer. med. Ass., 164, 172.

THE LOCALIZATION OF ENTEROKINASE TO THE BRUSHBORDER MEMBRANE OF THE GUINEA PIG SMALL INTESTINE

R. W. LOBLEY and R. HOLMES (Department of Gastro-enterology, The Royal Infirmary, Manchester) Entero-kinase, or enteropeptidase, releases trypsin fromtrypsinogen and thus initiates the process leading toprotein digestion in the small intestine. Nordstromand Dahlqvist (1970) showed that enterokinase waspresent mainly in the epithelial cells at or near thevillous tips in the rat small intestine, but the sub-cellular location of enterokinase has not been clearlydefined.To determine the distribution of enterokinase

throughout the small intestine of guinea pigs, 5 cm-segments were homogenized and enterokinase levelsdetermined. Enterokinase activity was found onlyin the first quarter of the small intestine and was un-detectable in the distal jejunum and in the ileum.Brush borders were prepared from the proximal

small intestine of guinea pigs, and after Tris-disrup-tion, the resultant microvillous membranes wereseparated on a glycerol density gradient (Eichholzand Crane, 1965). There was a marked rise in specificactivity of enterokinase in the purified brush bordersover the mucosal homogenate, and a further increasein specific activity in the brush border membranes.This was similar to the progressive increase in specificactivities of sucrase in the same preparations. Theseresults suggest that enterokinase, like sucrase, is abrush border enzyme and that its action on trypsino-gen occurs initially at the epithelial cell surface andnot in the lumen of the small intestine.

REFERENCES

Eichholz, A., and Crane, R. K. (1965). J. Cell. Biol., 26, 687.Nordstrom, C., and Dahlqvist, A. (1970). Biochim. biophys. Acta,

198, 621.


http://gut.bmj.com

/G


ber 1970. Dow

nloaded from

http://gut.bmj.com/


AN EXPERIMENTAL MODEL OF HUMAN CHRONIC PANCREA-TITIS

HENRI SARLES (Institut de la Sante et de la RechercheMedicale, Marseille, France) Thirty-eight Wistar ratsreceiving ad libitum a diet normal in proteins witheither 10% or 34% of lipids were given a lethal doseof ethanol by gastric intubation for three to 20 days.No pancreatic lesion was observed.

Sixty-one Wistar rats were given ad libitum astandard diet with either 20% ethanol (45 rats) orwater (16 rats) as beverage during 20 to 30 months.Half of the rats drinking ethanol developed pan-creatic lesions indistinguishable at the optical micro-scope level from those observed in the human chroniccalcifying pancreatitis (Sarles, Sarles, Camatte,Muratore, Gaini, Guien, Pastor, and le Roy, 1965)with small foci showing atrophy of the acini,augmentation and dilatation of the ducts, calcifiedprotein plugs in the ducts, and sclerosis. In thepancreatic juice of these rats, a spontaneous precipita-tion of proteins was observed and the protein con-centration is significantly higher than in controls.

This experiment provides the first experimentalmodel of chronic pancreatitis identical to the humandisease.

REFERENCE

Sarles, H., Sarles, J. C., Camatte, R., Muratore, R., Gaini, M.,Guien, C., Pastor, J., and le Roy, F. (1965). Gut, 6, 545.

THE ROLE OF SERUM TRYPSIN INHIBITORS IN ACUTEPANCREATITIS

0. FITZGERALD and K. F. MCGEENEY (Department ofMedicine and Therapeutics, University College,Dublin) The rise in enzymatically determined serumtrypsin inhibitors in acute pancreatitis is well docu-mented. However, the behaviour of the individualserum trypsin inhibitors in this condition is notcompletely understood. The elevation of the serumtrypsin inhibitor level in acute pancreatitis is usuallyseen some time following the acute episode. At thetime of the acute episode the level of the serum trypsininhibitors drops and then comes up again some hourslater. The rise is mainly due to a1 antitrypsin which canreach a level double the normal value. On the otherhand the a2 macroglobulin inhibitor does not reboundbut remains low.The pancreatic proteases, notably trypsin, which are

bound to a, antitrypsin are enzymatically inactivewhile those bound to a2 macroglobulin form trypsinprotein esterase which is active against smallmolecular weight substrates. In a study in vitro,trypsin protein esterase was found to be an angio-tensinase. This would suggest that pancreatic pro-teases bound to a% macroglobulin may exert an im-portant effect.

AN ENDOCRINE TUMOUR IN KIDNEY AFFECTING SMALLBOWEL STRUCTURE, MOTILITY, AND FUNCTION

M. H. GLEESON, S. R. BLOOM, J. M. POLAK, K. HENRY, andR. H. DOWLING (Departments of Medicine and Pathol-ogy, RoyalPostgraduateMedical School, HammersmithHospital, London W12, and Department ofMedicine,Middlesex Hospital, London W1) Ectopic endocrine

polypeptide production by tumours is well recognizedbut their influence on small bowel structure and func-tion has rarely been studied. We report a unique casewith altered small bowel structure, motility, andabsorptive function associated with an endocrinerenal tumour.A housewife (aged 44) presented with hypoprotein-

aemic oedema, anorexia, and marked constipation,retained faeces producing abdominal distension.Investigations showed neither proteinuria nor protein-losing enteropathy. Despite constipation she hadsteatorrhoea (20 g/day), vitamin B12 malabsorption,gross indicanuria (520 mg/24 hours), and a diabeticGTT. Repeated barium studies showed grossly dilatedsmall intestine, thickened mucosal folds, and slowtransit. Intestinal hypertrophy was confirmed whensurgical biopsies showed marked enlargement ofvilli, easily visible to the naked eye, measuring up to1,350 , (N300-800 ,u).

Intravenous pyelography showed a renal tumour.After nephrectomy, histology suggested an endocrineneoplasm with secretory granules on electron micro-scopy. Histochemical studies confirmed that thetumour cells contained a polypeptide which showedimmunofluorescence with antiglucagon serum. Radio-immuno assay of tumour and pre-operative plasmashowed high glucagon levels. On bioassay tumourextracts showed glucagon-like activity.

Following nephrectomy the small bowel abnor-malities promptly disappeared. It is suggested that thetumour hormone, either glucagon alone and/oranother unidentified hormone, was responsible for thesmall intestinal effects.

INACTIVATION OF GASTRIN AND PENTAGASTRIN IN THEPORTAL CIRCULATION1

JOHN M. TEMPERLEY, JOHN H. WYLLIE, and BRIAN H.STAGG (Gastrointestinal and Surgical Units, UniversityCollege Hospital Medical School, London WCJ)The effect on synthetic human gastrin I (SHG) andpentagastrin (PG) of passage through the portal cir-culation has been studied.Acid secretion was measured in anaesthetized dogs

by washing out the cannulated stomach every tenminutes. SHG was infused (usually at 2 utg/kg/hr)alternately into the left external jugular vein and theportal vein in four dogs, or the superior mesentericartery in another four dogs.Acid secretion was similar during portal and jugular

vein infusions, but during superior mesenteric arteryinfusions acid secretion was 60% of that duringjugularvein infusions. The effect of portal and jugular veininfusions of PG was also studied-acid secretionfell to near basal levels during portal vein infusions.Thus, PG, but not SHG, is inactivated by passagethrough the liver. SHG has diminished potency afterpassage through the intestines.

Similarly, in the perfused rat stomach preparationfor the bioassay of gastrin (Smith, Lawrence, Colin-Jones, and Schild, 1970) ligating the superior mesen-teric vessels doubled the acid secretory response tostandard doses of SHG, but had no effect on theresponse to standard doses of PG.

In studies in vitro, the rates of inactivation of SHGand PG by homogenates of rat small bowel mucosaand liver were measured by bioassay to be 0.9 and0.04 nanomoles/min/mg tissue protein respectively'This work was supported in part by the Medical Research Council.


http://gut.bmj.com

/G


ber 1970. Dow

nloaded from

http://gut.bmj.com/


for SHG and 0 and 20-0 nanomoles/min/mg tissueprotein respectively for PG.

REFERENCE

Smith, G. M., Lawrence, A. J., Colin-Jones, D. G., and Schild,H. 0. (1970). Brit. J. Pharmac., 38, 206.

THE EFFECT OF PROTEIN, CARBOHYDRATE, OR FATON PLASMA GASTRIN CONCENTRATION IN HUMAN

SUBJECTS

P. C. GANGULI (Medical Research Council, ClinicalEndocrinology Unit, Edinburgh) Hansky, Soveny, andCain (1970) have suggested that protein, carbohydrate,and fat are all potent stimulants ofendogenous gastrinsecretion but the meals used by these authors wereoften a mixture of protein, carbohydrate, and fat.In the present study the protein meal consisted of250 g grilled lean beef steak (55% protein), the carbo-hydrate meal was 400 g rice and sugar (93 % carbo-hydrate), and the fat meal was 120 g double cream(93 % fat).

Five healthy human subjects were studied witheach meal and thus fifteen subjects took part in thisinvestigation. A sample of blood was taken from thefasting subjects and then they were given the meal;after it blood was sampled at 10, 20, 30, 45, 60, 75, 90,105, 120, and 150 minutes using an indwelling venouscatheter. Plasma gastrin was measured by the radio-immunoassay method of Ganguli and Hunter (1969)reported previously to this Society.

Following the protein meal, there was a rise inplasma gastrin concentration in all five subjects. Thepeak concentration was reached 30 to 60 minutes afterthe meal, and was four to five-fold higher than theirfasting level; this difference is highly significant(p< 0-001). Plasma gastrin concentration then gradu-ally decreased until it reached the fasting level 120to 150 minutes after the meal. The carbohydrate mealhad no significant effect on the fasting plasma gastrinconcentration in any of the five subjects, while afterthe fat meal there was a gradual fall in the plasmagastrin concentration, though this finding was notstatistically significant.The present findings on the effect of these three

different meals on the concentration of endogenousgastrin in healthy human subjects are consistentwith the earlier observations on the physiology of gas-tric secretion.

REFERENCESHansky, J., Soveny, C., and Cain, M. D. (1970). Proc. JVth

World Congr. Gastroent., 206.Ganguli, P. C., and Hunter, W. M. (1969). Gut, 10, 413.

REDUCTION IN MUCOSAL BLOOD FLOW AS A FACTORIN THE DECREASED RESPONSE TO GASTRIN AFTER

VAGOTOMY

A. A. HARPER, J. D. REED, D. J. SANDERS, and j. R. SMY(Department of Physiology, Medical School, TheUniversity, Newcastle upon Tyne) It has beensuggested that the decreased acid response of parietalcells to gastrin after vagotomy may result from areduction in the affinity of their receptors for gastrin(Grossman, 1970) with a consequent shift of the dose-response curve to the right (Konturek, Oleksy, andWysocki, 1968).

In groups of anaesthetized cats with the splanchnicnerves intact and the vagus nerves intact (A) or cut (B)there was no significant difference between the smallresting acid secretion, but the mucosal blood flow(MBF), measured by the amidopyrine method, wassignificantly higher (p < 0.001) in A than in B. Infusionof pentapeptide (0.1 ,ug/min) produced a significantlyhigher acid secretion (p = 0.02) and MBF (p < 0.001)in A than in B.As the proportion of injected gastrin reaching the

parietal cells is related to the MBF, the reduction inblood flow after vagotomy may be a factor in thedecreased response to gastrin and shift of the dose-response curve. This concept is supported by theobservation that in vagotomized animals small dosesof isopropylnoradrenaline, insufficient by themselvesto stimulate acid secretion, increase MBF and acidsecretion during infusion of small amounts of gastrin.

REFERENCES

Grossman, M. I. (1970). Frontiers in Gastrointestinal Hormone.Research, Nobel Symposium XVI.

Konturek, S. J., Oleksy, J., and Wysocki, A. (1968). Amer. J. dig.Dis., 13, 792.

THE EFFECT OF VAGOTOMY ON THE INTESTINAL TRANS-PORT OF WATER AND ELECTROLYTES

G. A. BUNCH and R. SHIELDS (Department of Surgery,Welsh National School of Medicine, Cardiff) Thecause of diarrhoea as a complication of vagotomyremains unexplained. The possibility that a defectin the intestinal handling of water and electrolytesmay explain this diarrhoea has been investigatedusing a technique of intestinal perfusion.Using physiological solution containing isotopes

and an inert marker, we have studied the intestinalhandling of water, sodium, and potassium by per-fusing 30-cm segments of jejunum using an orallyintroduced double lumen tube.

Eleven healthy patients, six patients immediatelyafter a non-intestinal operation, and five patientsimmediately after vagotomy were studied. Threepatients were studied again ten days after vagotomyand three patients who had severe postvagotomydiarrhoea were also studied.

Immediately after operation, the net absorptionof water and electrolytes was reduced in all post-operative subjects. The intestinal handling of waterandelectrolytes was the same in vagotomized and non-vagotomized patients.Ten days after vagotomy the net absorption of

water and electrolytes was no longer reduced and thetwo-way traffic across the mucosa was similar to thatof the healthy control subjects. The patients withestablished severe postvagotomy diarrhoea demon-strated no abnormality in their intestinal handlingof water or electrolytes.

RECURRENT ULCER AND POSTVAGOTOMY GASTRIC ACIDSECRETION

C. V. RUCKLEY, W. SIRCUS, C. W. A. FALCONER, W. P.SMALL, and A. N. SMITH (Gastro-Intestinal Unit,Western General Hospital, Edinburgh) The associa-tion between incomplete vagotomy and recurrent ulceris universally recognized. The practical value of thisobservation is limited by the fact that the majorityof patients with positive insulin tests do not developrecurrent ulcer.


http://gut.bmj.com

/G


ber 1970. Dow

nloaded from

http://gut.bmj.com/


Analyses of the insulin test by criteria other thanthose of Hollander (Bank, Marks, and Louw, 1967;Gillespie, Gillespie, and Kay, 1968) suggest the possi-bility of identifying the recurrent ulcer risk moreprecisely. Pre- and postoperative gastric secretion testsin 226 patients treated by vagotomy and drainage havebeen examined in respect of basal, maximal, andinsulin-stimulated secretion. Twenty-one were diag-nosed as having recurrent duodenal or jejunal ulcer.Each insulin test has been interpreted by five criteria(Gillespie et al, 1968) and scored accordingly. Patientswith a 3+ postinsulin response showed only a 10%/incidence of recurrent ulcer. When the response was5+ the incidence was 77 %Y. A 'late-positive' (secondhour) response was noted in 19% of patients withrecurrent ulcer. Significant differences in mean basaland 'maximal' acid outputs between recurrent ulcerand non-recurrent ulcer groups were also noted.Postoperative gastric secretion tests can be used notmerely to detect the patient at risk of developingrecurrent ulcer but also to forecast the degree of risk.This has important implications in the postoperativemanagement of these patients.

REFERENCES

Bank, S., Marks, I. N., and Louw, J. H. (1967). Gut, 8, 36.Gillespie, G., Gillespie, I. E., and Kay, A. W. (1968). Gut, 9, 470.

110 HIGHLY SELECTIVE VAGOTOMIES WITHOUT DRAINAGE(HSV) FOR DUODENAL ULCER

E. AMDRUP, D. JOHNSTON, and J. C. GOLIGHER (SurgicalDepartment I, Kommune Hospitalet, Copenhagen, andUniversity Department of Surgery, The GeneralInfirmary, Leeds) Previous attempts to treat duo-denal ulceration by truncal vagotomy alone withoutdrainage procedure gave poor results because of theoccurrence of gastric stasis (Dragstedt, Harper,Tovee, and Woodward, 1947). The operation ofHSV (Johnston and Wilkinson, 1970), in which thevagal nerve supply to the distal 6 cm of the stomachis preserve(d to impart 'vis-a-tergo' to the chyme,has been used previously in man (Holle and Hart,1967)-but never alone-without drainage or antrec-tomy.The insulin test soon after HSV was negative in

50 consecutive patients, but reverted to positive in40% of 50 patients re-tested two to 10 months later.Acid outputs, however, were very low. The positiveresponses may be attributable to vagal release ofgastrin. Reductions in spontaneous and maximal(pentagastrin) acid outputs in 100 patients weresimilar to those achieved by conventional vagotomy.There have been no deaths or recurrent ulceration.

Maximum follow up, however, is only 21 months.Transient food vomiting occurred in two patients,but there was no persisting stasis. Mild dumping ordiarrhoea was noted in less than 10 per cent. The resultsappear to be better than those achieved by vagotomyand drainage, but a prospective trial is needed toprove this.

REFERENCES

Dragstedt, L. R., Harper, P. V., Jr., Tovee, E. B., and Woodward,E. R. (1947). Ann. Surg., 126, 687.

Johnston, D., and Wilkinson, A. R. (1970). Brit. J. Surg., 57, 289.Holle, F., and Hart, W. (1967). Med. Klin., 62, 441.

THE APPLICATION OF NUMERICAL TAXONOMY TO THESEPARATION OF COLONIC INFLAMMATORY DISEASE

J. HYWEL JONES, W. CARD, M. CHAPMAN, J. E. LENNARD-JONES, B. C. MORSON, M. J. SACKIN, and P. H. A. SNEATH

(St Mark's Hospital, London, University DepartmentofMedicine in Relation to Mathematics and Computing,Glasgow, and Medical Research Council MicrobialSystematics Unit, University of Leicester) Someworkers consider there is such an overlap betweenproctocolitis and colonic Crohn's disease that noclear separation can be drawn between them. Toresolve this controversy a prospective analysis wasmade of 109 patients with non-specific colonicinflammatory disease consecutively admitted toSt Mark's Hospital. Clinical, pathological, andradiological data from these patients were analysedusing numerical taxonomic methods.

In the initial computer analysis many of the patientsdiagnosed as Crohn's disease fell into several smallclusters in close proximity to one relatively dense,large cluster which comprised most of the patientsdiagnosed as proctocolitis. When the analysis wasrepeated using a limited number of discriminatingcharacters, three relatively large clusters emerged.One contained 67 patients of whom 58 had beendiagnosed as proctocolitis; another contained 12patients, 10 having been diagnosed as Crohn's disease;and the third cluster contained eight patients ofwhom seven had been diagnosed as Crohn's disease.About half the patients regarded as having Crohn'sdisease were not included in the clusters.The results suggest that the disorder now termed

proctocolitis comprises a single relatively uniformgroup of patients which can be distinguished fromother types of colitis at present described as Crohn'sdisease.

REFERENCES

Lennard-Jones, J. E., Lockhart-Mummery, H. E., and Morson,B. C. (1968). Gastroenterology, 6, 1162.

Sokal, R. R., and Sneath, P. H. A. (1963). Principles ofNumericalTaxonomy. Freeman, London.

EVALUATION OF A BAYESIAN MODEL IN THE DIAGNOSISOF JAUNDICE

R. P. KNILL-JONES, J. D. MAXWELL, R. P. H. THOMPSON,and ROGER WILLIAMS (Liver Unit, King's CollegeHospital, London) Two hundred patients presentingwith jaundice were studied prospectively with finalallocation to one of nine separate diagnostic groupson the basis of histological, operative, or necropsyfindings. Data were analysed to show which pieces ofevidence were most useful in differentiatingbetween thediseases. Selected indicants were used in a Bayesianmodel to calculate the most likely diagnosis on theinitial findings for each patient.Three indicants obtained from the history (length

of illness, age, and duration of itching) were moreuseful as discriminators than any laboratory or scanmeasurement. This was reflected in the results fromthe model; for classification into nine diseases scanmeasurements gave a correct prediction in 43 % ofpatients, laboratory tests in 58 %, and clinical indicantsin 65 %. Classification into medical or surgicalcategories only was correct in 75%, 80%, and 82°%of patients respectively, comparing favourably withpublished reports of clinicians' diagnostic accuracyin a similar situation of about 85%. Combination ofall measurements produced some furtherimprovement.


http://gut.bmj.com

/G


ber 1970. Dow

nloaded from

http://gut.bmj.com/


BILE SALT METABOLISM IN THE NORMAL HUMANSMALL INTESTINE

T. C. NORTHFIELD, E. CONDILLAC, and I. MCCOLL (Guy'sHospital and St Bartholomew's Hospital, London)The steatorrhoea of the stagnant loop syndromehas been attributed to abnormal bile salt metabolisminvolving bacterial deconjugation (Tabaqchali, Hatzi-oannou, and Booth, 1968), but very little attentionhas been paid to bile salt metabolism in the normalhuman small intestine.We have studied small-intestinal samples obtained

by intubation from 11 subjects with no known gastro-intestinal disease, and at laparotomy in a further fivesubjects. Free (deconjugated) bile acids have beenconsistently demonstrated in ileal samples by thinlayer and gas/liquid chromatography. Their meanconcentration fell from 1.2 to 0-1 gmoles/ml followingampicillin (p < 0.02), suggesting deconjugation bybacteria, which colonize the normal human ileum(Drasar and Northfield, in press). Total bile acidconcentration, however, increased sevenfold followingampicillin (p < 0.05), suggesting impaired absorption.One physiological implication of these findings isthat bacterial deconjugation may be an importantfactor in the remarkable efficiency of the entero-hepatic circulation of bile salts, since animal experi-ments have shown that ileal absorption is considerablymore rapid for free than for conjugated bile acids(Dietschy, Salomon, and Siperstein, 1966). One clinicalimplication is that the presence of free bile acids isonly abnormal in jejunal juice, so that it is unnecessaryto screen ileal juice in order to exclude the stagnantloop syndrome in a patient with steatorrhoea.

REFERENCES

Tabaqchali, S.,Hatzioannou,J.,and Booth, C.C. (1968). Lancet,2, 12.

Drasar, B. S., and Northfield, T. C. In press.Dietschy, J. M., Salomon, H. S., and Siperstein, M. D. (1966).

J. clin. Invest., 45, 832.

INHIBITORY EFFECT OF NEOMYCIN ON CHOLESTEROLABSORPTION IN GERM-FREE PIGS

G. R. THOMPSON, K. HENRY, N. EDINGTON, and P. c.TREXLER (Departments of Medicine and Pathology,Royal Postgraduate Medical School, and Departmentof Pathology, Royal Veterinary College, London)Neomycin is an effective hypocholesterolaemic agentand in large doses also causes steatorrhoea. Thehypocholesterolaemic effect has been variously attri-buted to inhibition of bacterial 7oa-dehydroxylationof bile acids in the colon (Samuel and Sekowski,1969), damage to the small intestinal mucosa (Dob-bins, Herrero, and Mansbach, 1968), and intra-luminal precipitation of cholesterol during absorption(Thompson, MacMahon, and Claes, 1970). Thisstudy examines these mechanisms further bydetermining whether the hypocholesterolaemic actionof neomycin is dependent upon its antibiotic proper-ties.

Six germ-free piglets were given an intravenous doseof cholesterol-4-14C and their faecal excretion ofendogenous neutral sterols and bile acids measuredduring consecutive two-week periods on and offneomycin sulphate 1-2 g/day. Neomycin significantlyincreased the mean weekly excretion of neutral sterolsfrom 127 to 269 mg (p< 0.001) but bile acid excretion

remained unaltered at 224 and 213 mg/week. Faecalfat excretion was also increased to a slight but signifi-cant extent, from 51 to 64 m-equiv/week (P<0.02).Histology of intestinal mucosa showed no significantchanges on light microscopy.

These results support previous evidence that neo-mycin impairs cholesterol and fatty acid absorptionby precipitating mixed micelles within the intestinallumen. This effect is due to the polybasic nature ofneomycin, and not to its antibiotic properties.

REFERENCESSamuel, P., and Sekowski, I. (1969). J. clin. Invest., 48, 73.Dobbins, W. 0., Herrero, B. A., and Mansbach, C. M. (1968).

Amer. J. med. Sci., 225, 63.Thompson, G. R., MacMahon, M., and Clacs, P. (1970). Europ. J.

clin. Invest., 1, 40.

ABSORPTION OF BILE ACIDS BY THE COLON

JOHN S. MORRIS, K. W. HEATON, and A. E. READ (Depart-ment of Medicine, University of Bristol) The studyhas been designed to investigate the possibility ofcolonic absorption of bile acids. Patients over theage of 40 years who underwent surgery to the extra-hepatic duct system which required postoperativeT-tube drainage were studied. During operationsodium taurocholate-24-C14 or cholic carboxyl-C14acid (Tracerlab (GB) Ltd, Weybridge, Surrey) wereinjected into the lumen of the ascending or transversecolon. In one patient with permanent T-tube drainagecholic carboxyl-C"4 acid was introduced into thecolon through a sigmoidoscope.Over the succeeding three or four days all bile

draining from the T-tube was collected. After suitabledilution the contained radioactivity was measuredin a liquid scintillation counting system. Alcoholicextracts of the bile were separated by thin-layerchromatography, the separated bile acids were identi-fied, and the amount of radioactivity in the primaryand secondary bile acids was measured.

Studies in seven subjects show that the colon wasable to absorb substantial amounts both of cholicacid and sodium taurocholate. The radioactive labelwas recovered as conjugated cholate and deoxy-cholate. These studies expand the observations ofSamuel, Saypol, Meilman, Mosbach, and Chafizadeh(1968) who studied the colonic absorption of cholicacid alone.

Together with the observation that patients withouta colon have little or no secondary bile acids in theirbile or ileostomy effluent (Percy-Robb, Brunton, Jalan,McManus, Gould, and Sircus, 1969; unpublishedpersonal obervations), our data suggest that in man thelarge intestine is a major site of dehydroxylation ofbile acids. They also confirm the ability of the colonto absorb bile acids.

REFERENCES

Samuel, P., Saypol, G. M., Meilman, E., Mosbach, E. H., andChafizadeh, M. (1968). J. clin. Invest., 47, 2070.

Percy-Robb, I. W., Brunton, W. A., Telfer, Jalan, K. N., McManus,J. P. A., Gould, J. C., and Sircus, W. (1969). Gut, 10,1049. (Abstr.)

ABSORPTION OF TRITIATED VITAMIN K1 IN PATIENTSWITH FAT MALABSORPTION

M. J. SHEARER, C. N. MALLINSON, G. R. WEBSTER,and P. BARKHAN (Departments of Haematology,


http://gut.bmj.com

/G


ber 1970. Dow

nloaded from

http://gut.bmj.com/


Physiology, and Chemical Pathology, Guy's Hospital,and the Gastro-intestinal Unit, Greenwich Hospital,London) Previous work in man using tritiatedvitamin K1 (K1-3H) showed that, following an oraldose, peak blood levels were reached after two tofour hours and 8-19% of the dose was excreted in theurine after three days. About 50% of the dose wasrecovered from faeces, 20% as apparently unchangedK,-3H and 30% as more polar metabolites.

Radioactivity levels in blood and urine after oralK,-3H given with breakfast have been measured inpatients with steatorrhoea from four different causes.(1) In two of four adults with coeliac disease in whichradioactivity levels were reduced, effective treatmentrestored these levels to normal. (2) Three patientswith chronic pancreatitis showed markedly reducedlevels: in the two most severely affected patients,treatment with Cotazym resulted in normal levels.(3) In a patient with only 18 inches of small intestineremaining after resection, levels were low. (4) In apatient with complete biliary obstruction, no radio-activity was detected in blood or urine: 80% of thedose was recovered from the faeces, almost all ofwhich was unchanged K1-3H. Only in this last patientwas the prothrombin time prolonged. These studiessuggest that although vitamin K1 absorption isreduced in patients with fat malabsorption, this isprobably only clinically important when none isabsorbed. The faecal results suggest that the K1-3Hmetabolites originate from biliary excretion.

REFERENCE

Shearer, M. J., Barkhan, P., and Webster, G. R. (1970). Brit. J.Haematol., 18, 297.

FAMILIAL STEATORRHOEA WITH CALCIFICATION OF THEBASAL GANGLIA AND MENTAL RETARDATION

R. COCKEL, CHARLOYITE M. ANDERSON, EILEEN E. HILL,and C. F. HAWKINS (Queen Elizabeth Hospital andInstitute of Child Health, Birmingham, and the EastBirmingham Hospital) Two boys and fratemal twingirls, of a sibship of 16, were seen with steatorrhoea,mental retardation, and calcification of the basalganglia. The similarity of clinical features and investi-gative findings suggested a disease of recessive type.

Diarrhoea started in childhood and was associatedwith irregular passage of voluminous faeces. Episodicabdominal pain occurred with gross distension,sometimes simulating intestinal obstruction. Radio-graphs showed variable distension and atony of thesmall bowel, the duodenum being constantly abnormal.Absorption of fat and xylose were abnormal butglucose and disaccharides were absorbed normally.There was no anaemia and serum biochemicalvalues were normal apart from moderate reductionof cholesterol and ,B-lipoprotein. Intestinal biopsyand pancreatic secretion were normal. Early growthwas normal though attainment of developmentalmilestones was delayed.

Steatorrhoea was attributed to defective smallbowel motility with bacterial colonization causingineffective action of bile salts. In one patient there wasbasal gastric hypersecretion, with duodenal pHbetween 3 5 and 7-4. The duodenal juice containedKlebsiella aerogenes and deconjugated bile salts.One boy, aged 15, died after operations for suspected

intestinal obstruction; necropsy was performed. Thethree surviving patients are kept reasonably well on alow fat diet and antibiotics.

STUDIES OF COLONIC CARCINOMA ANTIGENS

MICHAEL D. TURNER, MARTIN S. KLEINMAN, and LEEHARWELL (Gastroenterology Unit, Department ofMedicine, University of Rochester School of Medicineand Dentistry, Rochester, NY, USA)' Antisera wereprepared in rabbits by immunization with perchlorateextracts of colonic carcinomata. On double immuno-diffusion these sera gave precipitin lines against tumourextracts which persisted after absorption of the serawith human red cells, plasma,normal colonic mucosa,and preparations of colonic bacteria. Two persistingprecipitin lines were observed with the original tumourextracts and with fractions after gel filtration andelectrofocusing. Antiserum donated by Dr Goldalso gave two lines of precipitation when tested againstsome of our extracts: one of these lines gave a reactionof identity with that given by the antiserum producedhere. The results suggest that perchloric acid mayextracttwo antigens from colonic carcinomata. The presentstudies confirm previous reports of the existence oftumour antigens in human colonic cancer (Gold andFreedman, 1965a and b; von Kleist and Burtin,1961) and lend support to the possibility of developingimmunological screening tests for this disease.'Supported by grant no. CA-09546 from the National CancerInstitute, U.S.P.H.S.

REFERENCES

Gold, P., and Freedman, S. 0. (1965a). J. exp. Med., 121, 439.Gold, P., and Freedman, S. 0. (1965b). J. exp. Med., 122, 467.Kleist, S. von. and Burtin, P. (1969). Cancer Res., 29, 1961.

INTESTINAL IMMUNOGLOBULIN LEVELS AND BACTERIALFLORA IN HYPOGAMMAGLOBULINAEMIC ADULTS IN

RELATION TO INTESTINAL ABSORPTIVE FUNCTION

D. M. PARKIN, D. B. L. MCCLELLAND, I. W. PERCY-ROBB,R. R. O'MOORE, and D. J. C. SHEARMAN (UniversityDepartments of Therapeutics and Clinical Chemistry,The Royal Infirmary, Edinburgh) Five patients withadult-onset panhypogammaglobulinaemia and onewith isolated IgA deficiency were studied togetherwith control subjects. Gastric and jejunal juice wasobtained for immunoglobulin estimations and immedi-ate aerobic and anaerobic culture. Jejunal aspirateswere also tested for deconjugated bile acids by thin-layer chromatography and individual organismswere tested for deconjugating ability in vitro.

All patients with panhypogammaglobulinaemiahad large numbers of Giardia lamblia in the jejunalaspirate: this organism was not found in the IgA-deficient patient or the controls. Bacterial counts,especially of 'enteric' organisms (coliforms andbacteroides), tended to be greater in the patients withpanhypogammaglobulinaemia than in controls andthe IgA and IgM levels were much lower, the oneexception being the IgA-deficient patient whosejejunal IgM level was high normal. Fat malabsorptionwas found in three of the panhypogammaglobulin-aemic patients, and deconjugated bile salts weredetected in the aspirates from all these patients. It isconcluded that the abnormal bacterial flora inpanhypogammaglobulinaemia relates to deficiency inintestinal immunoglobulin, although IgM appearsable to compensate in the case of isolated IgAdeficiency. The resultant abnormal flora may interfere


http://gut.bmj.com

/G


ber 1970. Dow

nloaded from

http://gut.bmj.com/


with bile salt metabolism and affect intestinal absorp-tion. The possible role of the Giardia in the presenceof malabsorption is not yet clear.

IMPAIRED INTESTINAL SUGAR TRANSPORT IN BLIND-LOOPRATS in vivo

MICHAEL GRACEY, VALERIE BURKE, ADEMOLA OSHIN, andCHARLOTTE M. ANDERSON (Institute of Child Health,University of Birmingham, Birmingham) The occur-rence of impaired D-glucose tolerance in patientswith the blind-loop syndrome (Donaldson, 1967)and temporary monosaccharide malabsorption insome babies with bacterial contamination of thesmall bowel (Gracey, Burke, and Anderson, 1969)suggests that impaired carbohydrate absorption maybe related to bacterial overgrowth in the small intes-tine.The absorption of Arbutin (p-hydroxyphenol- -

glucoside), an analogue of glucose which shares thesame active transport pathway (Alvarado and Crane,1964) but is not metabolized, was studied in femaleadult Wistar rats with midjejunal blind loops, one tothree months after operation. In the anaesthetizedrat, segments of jejunum proximal and distal to theblind loop and measuring approximately 20 cm inlength were used for perfusion. This was done forone hour at 10 mI/hour with Krebs-Henseleit bicar-bonate buffer containing 10 mM Arbutin and poly-ethylene glycol as a non-absorbable marker, at 37°Cand pH 7-4 and perfusate collected continuously. Theamount of Arbutin transported across the segmentwas calculated by difference and expressed as g moles/cm of intestine/hour.

Control rats (17) transported 2 02±0 27 g moles/cm/hour. In the blind loop rats the proximal segments(5) transported 1 51±0 58 (P<0 1) while the distalsegments (5) transported 1 21 ±0 36, which is signifi-cantly less than the control value (p < 0.001).

REFERENCESDonaldson, R. M., Jr. (1967). Fed. Pro., 26, 1426.Gracey, M., Burke, V., and Anderson, C. M. (1969). Lancet, 2,

384.Alvarado, F., and Crane, R. K. (1964). Biochim. biophys. Acta,

93, 116.

MANNITOL UTILIZATION IN THE STAGNANT LOOPSYNDROME

P. R. SALMON, A. B. AJDUKIEWICZ, J. R. CLAMP, andA. E. READ (Department of Medicine, University ofBristol) At the present time there is no simplescreening or absorption test which will predictablydemonstrate small-bowel bacterial contamination(Hamilton, Dyer, Dawson, O'Grady, Vince, Fenton,and Mollin, 1970).

This study was designed to investigate bacterialmetabolism of the poorly absorbed (Fordtran,Rector, Ewton, Soter, and Kinney, 1965) and non-metabolized sugar alcohol mannitol in patients withthe stagnant loop syndrome.

Six normal subjects and 17 patients with radiologicalevidence of a blind loop syndrome were each given2-5 g of D-mannitol labelled with a tracer of D-manni-tol-1-'4C. Radio-carbon breath excretion wasmeasured over five hours using the method describedby Salmon, Read, and McCarthy (1969). A furthersix normal subjects were given 2-5 g of D-mannitol

and urinary mannitol excretion was measured overfive hours by gas chromatography. The results werecompared with urinary indican excretion, bacteriologyof the small bowel contents, and in-vitro bacterialmetabolism of tryptophan and mannitol.

It was shown that normal subjects absorbed lessthan 10% and metabolized less than 2% of the inges-ted mannitol in five hours. On the other hand thosewith the stagnant loop syndrome metabolized onaverage 13-2% of the mannitol within five hours.These results correlated with in-vitro bacterial meta-bolism of mannitol. Urinary indican levels were nor-mal in six patients with bacterial contamination ofthe small bowel.

It is suggested that a test based on radio-carbonbreath excretion may be used to determine the pre-sence and metabolic activity of small bowel bacteria.

REFERENCES

Hamilton, J. D., Dyer, N. H., Dawson, A. M., O'Grady, F. W.,Vince, A., Fenton, J. C. B., and Mollin, D. L. (1970).Quart. J. Med.. 154, 265.

Fordtran, J. S., Rector, F. C., Jr., Ewton, M. F., Soter, N., andKinney, J. (1965). J. clin. Invest., 44, 1935.

Salmon, P. R., Read, A. E., and McCarthy, C. F. (1969). Gut,10, 685.

OSTEOMALACIA IN ADULT COELIAC DISEASE

PH. BORDIER, D. HIOCO, G. HEPNER, D. E. EDWARDS,F. H. DOYLE, and c. C. BOOTH (Unite de Recherchesur le Metabolisme Phospho-calcique, H6pital Lari-boisiere, Paris, France, and the Royal PostgraduateMedical School, London) Osteomalacia is a recog-nized complication of adult coeliac disease. Reporteddifferences in its incidence may reflect differences indiagnostic criteria (Melvin, Hepner, Bordier, Neale,and Joplin, 1970). We present an investigation of17 unselected adults with untreated coeliac disease.Iliac bone biopsy was studied by quantitative histo-logical techniques (Matrajt, Bordier, and Hioco, 1967).Total bone and osteoid were expressed as percentagesof total bone volume. Osteoid was also measured as apercentage of cancellous bone surface covered withosteoid. The proportion of osteoid surface associatedwith a calcification front was determined (Bordier,Matrajt, Hioco, Hepner, Thompson, and Booth, 1968).Using empirical histological criteria, patients were

assigned to three groups: normal (seven patients),mild osteomalacia (five patients), and severe osteo-malacia (five patients). Using statistical techniques ofdiscriminant analysis, analysis ofvariance and multipleregression, it was shown that: (1) When all variableswere used, each patient was assigned to the 'correct'group with an estimated P value > 0.999. (2) None ofthe biochemical measurements, either singly or ingroups, could discriminate effectively. (3) Severeosteomalacia was associated with higher levels of totalbone, osteoid and serum alkaline phosphatase. Theproportion of osteoid surface with a calcificationfront, serum calcium and inorganic phosphatedecreased with increasing severity of bone disease.(4) Osteoid and surface osteoid were moderatelycorrelated (P < 0-01) with serum alkaline phos-phatase. Total bone and the proportion of osteoidwith a calcification front were less well correlated withalkaline phosphatase (P < 0-05).

REFERENCES

Melvin, K. E. W., Hepner, G. W., Bordier, Ph., Neale, G., andloplin, G. F. (1970). Quart. J. Med., 39, 83.


http://gut.bmj.com

/G


ber 1970. Dow

nloaded from

http://gut.bmj.com/


Matrajt, H., Bordier, Ph., Hioco, D. (1967). In L'Osteomalacie,edited by D. Hioco, p. 101. Masson et Cie, Paris.

Bordier, Ph., Matrajt, H., Hioco, D., Hepner, G. W., Thompson,G. R., and Booth, C. C. (1968). Lancet, 2, 437.

THE MUCOSAL CONTRIBUTION TO NON-DIETARY LIPIDIN THE INTESTINAL LUMEN

PETER B. COTTON, introduced by B. Creamer (StThomas' Hospital, London) Endogenous (non-dietary)intestinal lipid may be derived from the mucosa(by cell loss and exudation) and from bile and bacteria.The mucosal contribution has been assessed by per-fusing the entire small intestine of fasting anaesthetizedrats, with bile excluded. The cell and lipid contentsof these perfusates, and of suspensions of isolatedintestinal cells, were measured by DNA analysis, andby a specially developed method of quantitative thin-layer chromatography. The rate of cell loss obtainedwas within the physiological range.

It was shown that lecithin, phosphatidyl ethano-lamine, cholesterol, cholesterol ester, and triglycerideenter the intestinal lumen only in cellular form, where-as some non-esterified fatty acid is freely exuded inde-pendently of cells.

In fasting rats, lipid exfoliation and exudationaveraged 3 mg/hour (250 mg/kg/day), the majorityof which must normally be reabsorbed.From published estimates of the lipid content of

human intestinal cells and of daily intestinal cellloss (Croft, 1970), it appears that 10-25 g of lipid isnormally exfoliated each day into the small intestineof fasting man. When cell loss is increased and ab-sorption impaired, steatorrhea may result partlyfrom exfoliation and exudation of endogenous lipid-a state of fat-losing enteropathy.

REFERENCE

Croft, D. N. (1970). Proc. roy. Soc. Med., in press.

MUCOSAL ABNORMALITIES AND DISACCHARIDASES INPERNICIOUS ANAEMIA

A. S. PEN.A, S. C. TRUELOVE, SHEILA T. CALLENDER, andR. WHITEHEAD (The Radcliffe Infirmary, Oxford)Patients with untreated pernicious anaemia may showmorphological abnormalities of the small-intestinalmucosa (Sauli, Astaldi, and Malossine, 1963; Foroo-zan and Trier, 1967). Abnormalities of the small-intestinal mucosa are frequently associated withdepressed disaccharidase levels. We have thereforestudied the disaccharidase levels in patients withpernicious anaemia, both before and after treatmentwith vitamin B12.

Six patients with untreated pernicious anaemia werebiopsied from the duodenojejunal flexure with aCrosby capsule immediately before treatment withvitamin B12 was started. Four of these patientsshowed partial villous atrophy on light microscopy;the other two had a normal villous pattern. The firstfour patients were re-biopsied at four to seven daysafter 1,000 gtg vitamin B12 had been given intra-muscularly, and all of them showed a marked improve-ment in the villous pattern.Another six patients with pernicious anaemia who

had been treated for varying periods of time were alsobiopsied and showed an entirely normal small-intesti-nal mucosa.The levels of the disaccharidases were low in the

patients with untreated pernicious anaemia. Whencompared with the treated group and with a group of19 normal volunteers there were significant differences.The four patients who were re-biopsied after treatmentwith vitamin B12 showed a rise in the values of all thedisaccharidases.These abnormalities in the small-intestinal mucosa

in untreated pernicious anaemia were not associatedwith significant clinical symptoms.Both the morphological abnormalities and the

depressed enzyme levels probably reflect the adverseeffect of vitamin B12 deficiency on DNA synthesis.

REFERENCES

Sauli, S., Astaldi, G., and Malossine, L. (1963). Acta vitamin.(Milano), 17, 43.

Foroozan, P., and Trier, J. S. (1967). New Engl. J. Med., 277, 553.Hepner, G., and Herbert, V. (1969) Fed. Proc., 28, 513.

THE FATE OF ILEORECTAL ANASTOMOSIS IN CROHN 'SDISEASE OF THE LARGE BOWEL

J. H. BURMAN, J. ALEXANDER WILLIAMS, and W. T.COOKE (The General Hospital, Birmingham) Anassessment has been made of the results of ileorectalanastomosis in 25 patients with Crohn's colitis usinganastomotic leak and disease recurrence as indicesof short- and long-term morbidity.Leaks occurred in eight patients (32%) despite

temporary ileostomy in two and irrespective ofwhetherthe anastomosis was made through diseased rectumor not. The occurrence of a leak had no long-termprognostic significance. The incidence of leakage andof recurrence was unaffected by the administrationof steroids, age at operation, or duration of illness.Sigmoidoscopic findings before operation gave noindication as to the likelihood of leak or futurerecurrence although the relatively normal appearancesmust have influenced the decision to perform anileorectal anastomosis.One patient (who also had severe amyloid disease)

died as a result of operation. Two patients died subse-quently: one with severe recurrence and one withhypertension. Two patients, who have had revisionileorectal anastomoses and five who now have ileos-tomies, are well. Nine patients have not developedrecurrence and are well (36 %). The overall diseaserecurrence rate was 60% and the mortality rate 12 %.The incidence of pre-stomal ileitis in 31 patientstreated by panprocto-colectomy alone is 52%.

REFERENCES

Jones, J. H., Lennard-Jones, J. E., and Lockhart-Mummery, H.(1966). Gut, 7, 448.

Hawke, W., and Turnbull, R. (1966). Gastroenterology, 51, 802.Lindner, A., Marshak, R., Wolf, B., and Janowitz, H. (1963).

New Engl. J. Med., 269, 379.

COELIAC DISEASE, VASCULITIS, AND CRYOGLOBULINAEMIA

W. F. DOE, D. J. EVANS, J. R. HOBBS, and c. c. BOOTH(Departments of Medicine, Pathology, and ChemicalPathology, Royal Postgraduate Medical School,London) Adult coeliac disease has previously beenassociated with a variety of skin disorders. This paperdescribes four patients with adult coeliac disease who


http://gut.bmj.com

/G


ber 1970. Dow

nloaded from

http://gut.bmj.com/

The British Society of Gastroenterology

developed vasculitis associated with cryoglobulin-aemia, a hitherto unrecognized association.

All four patients had a rash, similar to that foundin patients with cryoglobulinaemia, but in only twowere there episodes of Raynaud's phenomenon. Thehistological appearances of the skin showed a non-specific vasculitis with polymorphonuclear infiltrationand in one case there was a necrotizing arteritisinvolving skin and muscle alone. Serum from thesepatients separated at 37°C showed a cryoprecipitateon standing at 4°C. The cryoglobulin was shown tocontain polyclonal IgG and IgM in one case, andin two other patients the cryoprecipitate formed at

4°C contained a mixture of IgM, IgG, and IgA.Rheumatoid factor and direct antiglobulin testswere positive in all three patients studied. Thesefindings were typical of a secondary or mixed typecryoglobulinaemia. In two patients the rash respondedsatisfactorily to steroid therapy, but in the other twocases, both suffering from severe coeliac disease,unresponsive to continued treatment, steroid therapyhad no effect on the rash and both these patients died.These findings may be related to circulating immunecompiexes in patients with severe coeliac disease,or alternatively may reflect a biochemical abnormalitysuggestive of early malignant change.

The November 1970 IssueTHE NOVEMBER 1970 ISSUE CONTAINS THE FOLLOW1NG PAPERS

Quantitative measurement of iron stores withdiethylenetriamine penta-acetic acid MICHAELBARRY, GUISEPPE CARTEI, AND SHEILA SHERLOCK

Measurement of iron stores in cirrhosis usingdiethylenetriamine penta-acetic acid MICHAELBARRY, GUISEPPE CARTEI, AND SHEILA SHERLOCK

Ultrastructure of the liver in non-cirrhotic portalfibrosis with portal hypertension B. N. TANDON,R. LAKSHMINARAYANAN, S. BHARGAVA, N. C.NAYAK, AND S. K. SAMA

Pancreatic islet cell tumour with watery diarrhoeaand hypokalaemia D. J. STOKER AND V. WYNN

Corticosteroid or corticotrophin therapy inCrohn's disease (regional enteritis) w. T. COOKEAND J. F. FIELDING

A comparative radiographic and pathologicalstudy of intestinal vaso-architecture in Crohn'sdisease and in ulcerative colitis F. BRAHME ANDC. LINDSTROM

Surgery in amoebic colitis D. STEIN AND SIMMYBANK

Further studies on the perfusion method formeasuring intestinal absorption in man: Theeffects of a proximal occlusive balloon and amixing segment G. E. SLADEN AND A. M. DAWSON

Gastric acid secretion in Chinese with particularreference to the dose of histamine required formaximal stimulation WYE POH FUNG

Plasma insulin response to oral carbohydrate inpatients with glucose and lactose malabsorptionJ. D. MAXWELL, MARGARET T. MCKIDDIE, ANNE

FERGUSON, AND K. D. BUCHANAN

TechniqueA genteel device for collecting faeces E. S.PRYKE AND H. M. WHYTE

Progress report

New developments in the therapy of amoebiasisS. J. POWELL

Progress report

Gastrointestinal polyposis H. J. R. BUSSEY

Notes and activities

Copies are still available and may be oblained from the PUBLISHING MANAGER,

BRITISH MEDICAL ASSOCIATION, TAVISTOCK SQUARE, WC1H 9JR price 17s. 6D.

1067 on June 6, 2022 by guest. P

rotected by copyright.http://gut.bm

j.com/

Gut: first published as 10.1136/gut.11.12.1055 on 1 D

ecember 1970. D

ownloaded from

http://gut.bmj.com/

31st annualmeeting ofthe british society of gastroenterology

Documents