1035

better in my life," " New lease of life," " Not so well for

years," " Christmas dinner with all the trimmings "-these are fair sample comments. Our one poor resultwas in a woman whose ulcer did not show on barium mealbut was discovered on gastroscopy. Postoperative gastricbiopsies show that she has a concomitant gastritis, andthis rather than the ulcer is probably the cause of hersymptoms.

3 patients have since died. Their deaths have beenunconnected with surgery. Cerebral thrombosis claimedone, a carcinoma of the splenic flexure with hepaticsecondaries claimed another at the end of two and a halfyears, and the 3rd died of coronary thrombosis associatedwith chronic myeloid leukaemia after twenty-one months.

Partial Gastrectomy as Operation of ChoiceAlmost two-thirds of our patients had gastric ulcers.

Here the choice was simple : it was either gastrectomyor nothing.The most contentious part of this paper is the selection

of partial gastrectomy as the surgical procedure of choicefor duodenal ulceration.

Partial gastrectomy is the long-established rationaltreatment, whereas gastro-enterostomy is largely dis-credited at present. But gastro-enterostomy is oftenrecommended for the aged ; it is said to be quicker,safer, and equally effective. Yet speed is not an impor-tant factor in a gastric operation, and shock is no longera complication of gastrectomy. Gastro-enterostomy isnot safe, because it is not equally effective. After

operation the patient may get relief of symptoms but issubject to the same lethal complications of pepticulceration as before. He is subject to the same compli-cation that has led to the abandonment of gastro-enterostomy in younger people-the hazard of marginalulceration. We could trace no evidence that older

people are less liable to marginal ulceration. The issuewith which it has been confused is that there is a decreasein gastric acidity coincident with the advance of age,especially in men (Vanzant et al. 1932). But pepticulceration in general is as common in the aged as it isin other age-groups :

"

10% cases of peptic ulceration occur in persons over sixty.The age distribution of the general white population of theUnited States is just 10.8% over the age of sixty." Mulsow1941.)Our opinion is that, although the gastric acidity is

decreased, the mucosal resistance is also decreased, andthe end-result is peptic ulceration.The proof that jejunal ulceration develops after gastro-

enterostomy is from fact and inference :Kiefer and McKell (1947) found 4 jejunal ulcers following

gastrojejunostomy in patients aged more than 65. 2 of theseulcers occurred in patients whose initial symptoms of pepticulceration began at the age of more than 70.One of us did a low Polya gastrectomy as an emergency for

bleeding from a duodenal ulcer in a man aged 71 ; a yearlater he had a further operation for closure of a perforatedanastomotic ulcer.

Charles Wells (personal communication) ivrites : ‘‘ Even the

development of pyloric stenosis is an indication of continuedactivity. The element of fibrosis is by no means an indicationof diminished virulence. The incidence of marginal ulcerationafter gastro-enterostomy is no less in old people than it is inthe young."

Ivor Lewis (personal communication) says that, if theduodenum is opened as the first step of the operation in casesof pyloric stenosis, one will nearly always find an actual ulcerstill present. This proves the dangers of stomal ulceration,which he found " fairly frequently."Decreased acidity is often caused by the gastritis

which is secondary to pyloric hold-up. When this isovercome by gastro-enterostomy without attention to itsprecipitating factor, peptic ulceration, the gastritisresolves and the acidity increases again.

Conclusions

Given adequate indications and careful preparation,surgical treatment for peptic ulceration in old people issafe.

Partial gastrectomy has given promising results in ourhands.

SummaryPeptic ulceration, recurrent or new, is common in the

aged ; they stand its complications poorly.When elective surgery is indicated, we believe that

partial gastrectomy is the operation of choice.30 consecutive elective partial gastrectomies in patients

aged more than 70 are reviewed. There were no deathsdue to operation. All of these patients have beenfollowed up.

We wish to thank Professor Wells and Mr. Ivor Lewis fortheir personal communications.

REFERENCES

Carp, L. (1948) Int. Abstr. Surg. 87, 1.Cole, W. H. (1953) Ann. Surg. 138, 145.Fraenkel, G. J., Truelove, S. C. (1955) Brit. med. J. i, 999.Kiefer, E. D., McKell, D. M. jun. (1947) J. Amer. med. Ass. 133, 1055.Mulsow, F. W. (1941) Amer. J. digest. Dis. 8, 112.Vanzant, F. R., Alvarez, W. C., Eusterman, G. B., Dunn, H. L.,

Berkson, J. (1932) Arch. intern. Med. 49, 345.

SPORADIC NON-ENDEMIC GOITROUS

CRETINISMHEREDITARY TRANSMISSION

JAMES H. HUTCHISON

O.B.E., M.D. Glasg., F.R.C.P., F.R.F.P.S.VISITING PHYSICIAN, ROYAL HOSPITAL FOR SICK CHILDREN,

GLASGOW

E. M. MOGIERM.B. Glasg., M.R.C.P., F.R.F.P.S.

UNIVERSITY LECTURER IN MEDICINE, MUIRHEAD DEPARTMENTOF MEDICINE, ROYAL INFIRMARY, GLASGOW

ENDEMIC cretinism results from a severe lack ofiodine over several generations and is almost invariablyassociated with goitre. The common type of sporadicnon-endemic cretinism, on the other hand, is notassociated with goitre ; indeed it is usually considered tobe due to an embryonic defect in the development of thethyroid gland (Dorff 1934), which is often assumed to beabsent. Nevertheless radioactive-iodine (1311) studies ofcretins who, on clinical examination, seem to have nothyroid tissue at all have sometimes revealed enough131I in the neck to suggest that some thyroid tissue is

present (Quimby and McCune 1947, Silverman andWilkins 1953), which may be either ectopic or unde-scended (McGirr and Hutchison 1955), confirming anearly anatomical observation in cretinism (MacCallumand Fabyan 1907).The underlying cause of sporadic cretinism remains

unknown. Childs and Gardner (1954), who, seekingevidence of a hereditary tendency, investigated thefamilies of 90 cretins, found that, though there seemedto be such a tendency, a familial prevalence was distinctlyuncommon. Consanguineous matings among the parentsof the cretins was not strikingly commoner than in acontrol series of normal children.

In contrast to this type of sporadic cretinism is whathas been described as " sporadic cretinism associatedwith goitre." In this type of case familial prevalence andconsanguinity of the parents are common (Osler 1897,Lerman et al. 1946, Hurxthal 1948, Stanbury and Hedge1950, Hubble 1953, McGirr and Hutchison 1953, Wilkinset al. 1954, Hutchison and McGirr 1954, Sexton andMack 1954, Stanbury et al. 1955a, 1955b), but isolated

1036

cases without anyapparent familial pre-valence have also been

reported (Osler 1897,Hamilton et al. 1943,Stanbury 1951, McGirrand Hutchison 1953,Wilkins et al. 1954,Hutchison and McGirr

1954).None of these cases

of sporadic non-endemicgoitrous cretinism hasbeen due to deficiencyof iodine or to the

ingestion of goitrogens.Obviously in all of themthe hypo thyroid statewas due, not to a

simple lack of thyroidtissue (as it is in thecommon sporadic non-goitrous cretin), but toan inability on the part of the thyroid gland to meetthe body’s requirements, even after considerable enlarge-ment of the gland, most probably as the result of stimula-tion by thyrotrophin produced by the pituitary glandin response to a hypothyroid state.

Our understanding of the pathogenesis of goitrouscretinism has been greatly furthered by the use of radio-active iodine (1311). Excessive avidity of the thyroidgland for iodine has been a constant feature of thesecases, but evidence has also accumulated which suggestsa block in the synthesis of the thyroid hormone. Dis-

crepancies between the findings of different workersindicate that the block is not always at the same stagein every case. It seems likely that this block is due toenzyme deficiencies at different levels of synthesis indifferent cases.The high prevalence of goitrous cretinism in the families

of our patients led us to postulate that the conditionmay be due to an inborn error of metabolism (Hutchisonand McGirr 1954). A similar opinion has been expressedby Wilkins et al. (1954) and Childs (1954).We have since been able to add considerably to the

family tree of our largest family group (cases 1-7 and13 of Hutchison and McGirr 1954) and to correct someof the inaccuracies previously reported to us. The fullerdetails of this remarkable family seem worthy of record.

Y-McX Family GroupSuch details of the family tree as have been discovered

are shown in the accompanying figure. The membersof this family group are itinerant tinkers living in

Kintyre, Argyll, and the Isle of Islay, Scotland. Fromtime to time they travel in other parts of Scotland, andvery probably other members of the family group arescattered throughout the country, although we have notfound it possible to trace any with certainty. These

people are always on the move ; and this fact, combinedwith the distance from Glasgow of their usual haunts,explains our difficulty in collecting accurate data. Evennow we have reasons (see below) to believe that the familytree given here by no means exhausts the ramificationsof the Y-McX genes throughout Scotland.The original male Y came from Ireland to Scotland

about 160 years ago and settled as a blacksmith at

Southend, Kintyre. He married a McX, his full cousin.A female Y (ii.3) married another McX, who was her fullcousin. We have not been able to obtain a satisfactorypedigree of the main McX family, which is probablyscattered widely throughout the West and North ofScotland. So far as we know, none of the marriages ingeneration i of the Y family was consanguineous, but in

Pedigree of itinerant tinkers, showing appearance of goitrous cretinism and Werdnig-Hoffmann paralysis.

generations 11 and 111 the amount of intermarriage hasbeen astonishing, and possibly largely brought about bythe increasing demarcation and isolation of these peopleby the rapidly widening gulf between their mode ofexistence and that of the other inhabitants of the

country.The family tree (see figure) shows that 10 known

goitrous cretins (in.13, 21-23, and iv.9 and 10 and

16-19) have appeared among 31 persons in four lots ofsiblings (m.13-20; III.21-27 ; iv.9-15; andiv.16-24):6 females and 4 males-an approximately equal sex-distribution. The results of radioactive iodine studies in8 of these have already been published (cases 1-7 and 13of Hutchison and McGirr 1954). The 9th cretin (m.21)is the elder sister of IIL22 and ill.23 (cases 7 and 6 ofHutchison and McGirr 1954). One of us has examinedher clinically in Kintyre but she has not yet been sub-mitted to detailed study. The 10th (IV.10) is the youngerbrother of iv.9 (case 5 of Hutchison and McGirr 1954)and was first seen at the age of 4 weeks as a goitrousinfant without signs of hypothyroidism. Such signshave recently appeared at the age of 2 years. The resultof i31I studies on this child suggests that the goitrouscretinism in this family is due to a deficiency of theenzyme dehalogenase (McGirr and Hutchison 1956).

*

There have also been 4 cases of Werdnig-Hoffmamiparalysis (IV.1-4) in another lot of siblings. 3 patientsdied after hospital investigations in Glasgow and

Edinburgh had established the diagnosis beyond reason-able doubt. The 4th case (IV.4) has recently been

diagnosed. None of the parents of the cretins or of thepatients with Werdnig-Hoffmann paralysis showed anyevidence of disease. We have personally seen all theseparents. We have not, however, examined all thesupposedly healthy siblings of the cretins.

Discussion

It is obvious from the high familial prevalence ofsporadic goitrous cretinism noted by ourselves and othersin the United Kingdom and in the United States thatthis form of hypothyroidism must be genetically deter-mined. In none of the present cases was there anythingto suggest iodine deficiency or the ingestion of goitrogenicsubstances. Endemic cretinism usually appears in thefamily of a mother who is goitrous from iodine deficiency,whereas in the present cases the parents were, withoutexception, healthy.

It is therefore most unlikely that the appearance ofgoitrous cretinism in the Y-McX family group is relateddirectly to their common environment. On the otherhand, the pedigree (see figure) satisfies three of the

1037

criteria usually accepted as indicating simple recessiveinheritance (Roberts 1940) :

(1) The parents of the affected persons are normal to

outward appearances.(2) There is a striking familial prevalence : three of the four

lots of siblings contain more than one affected child. Theratio of normal to affected persons (21 : 10) shows the usualexcess over the 3 : 1 ratio which is typical of recessive inheri-tance because we have adequate information only about setsof siblings in whom at least one person is affected. A numberof normals in unaffected sibships must be unknown to us.

(3) All the affected persons are the children of con-

sanguineous marriages.Other characteristic, but rarely encountered, criteria, of

autosomal recessive inheritance are as follows :

(4) Affected persons who marry normals will have onlynormal offspring in most instances.

(5) Affected persons who marry affected persons will haveaffected offspring only.These last two criteria have not been satisfied in the

present pedigree, but in view of the predilection for con-sanguineous mating it is not inconceivable that the

thyroid treatment instituted by us may alter this positionin time.

Though it may properly be claimed that simple recessiveinheritance cannot be proved on the strength of a singlepedigree, the occurrence of familial prevalence and con-sanguinity in the families of other similar cases which wehave cited makes it extremely probable that the presentcases were so determined. The appearance within thesame pedigree of Werdnig-Hoffmann paralysis-a diseaseknown to be inherited as a mendelian recessive (Gates1946)-is interesting. Here also the first three criteriaof recessive inheritance mentioned above are fulfilled.

We have alluded to the possibility that the geneproducing goitrous cretinism in the Y-McX family maybe operating elsewhere in Scotland. We have alreadyrecorded the appearance of 3 siblings with goitrouscretinism in another family (Hutchison and McGirr 1954).These also were tinkers. They were referred to us byProf. J. L. Henderson, of Dundee. Careful inquiry byProfessor Henderson failed to produce any evidence thatthey were related to the Y-McX cretins, but (like us) hemet some resistance in taking a family history from thesepeople. We have recently had the opportunity of

applying radio-iodine tests to 2 more cretinous siblingswith goitre, who were referred to us by the Renfrewshirelocal authority from a home for deprived children. Theygave results which were identical with those found in ourearlier cases. The parents of these 2 children were alsoitinerant tinkers but could not now be traced. In viewof the unique and self-demarcated mode of life of tinkers,who tend to intermarry and to live a life apart from therest of the community, it seems at least probable thatthe various cretinous siblings whom we have studied owetheir condition to the same recessive gene occurring inthe homozygous state.

SummaryThe pedigree of a group of itinerant tinkers is presented.It is noteworthy for the remarkable frequency of

consanguineous mating.Within this family group there were 10 cases of goitrous

cretinism and 4 cases of Werdnig-Hoffmann paralysis.Goitrous cretinism and Werdnig-Hoffmann paralysis

are each due to the effect of single but quite separateautosomal recessive genes.

We wish to thank Dr. Catherine A. Brown, executive schoolmedical officer, Argyll County Council, for her great help intracing the various members of the family group ; andProfessor Henderson for his efforts to find the connecting linkbetween his cretinous tinkers and ours.

Childs, B. (1954) Amer. J. med. Sci. 228, 681.— Gardner, L. I. (1954) Ann. human Genetics, 19, 90.

Dorff, G. B. (1934) Amer. J. Dis. Child. 48, 1316.Gates, R. R. (1946) Human Genetics. New York.Hamilton, J. G., Soley, M. H., Reilly, W. A., Eichorn, K. B. (1943)

Amer. J. Dis. Child. 66, 495.Hubble, D. (1953) Lancet, i, 1112.Hurxthal, L. M. (1948) Med. Clin. N. Amer. 32, 122.Hutchison, J. H., McGirr, E. M. (1954) J. clin. Endocrin. 14, 869.Lerman, J., Jones, H. W., Calkins, E. (1946) Ann. intern. Med.

25, 677.MacCallum, W. G., Fabyan, M. (1907) Johns Hopk. Hosp. Bull.

18, 341.McGirr, E. M., Hutchison, J. H. (1953) Lancet, i, 1117.

— — (1955) J. clin. Endocrin. 15, 668.— — (1956) Lancet, i, 106.

Osler, W. (1897) Trans. Congr. Amer. Phys. & Surg. 4, 169.Quimby, E. H., McCune, D. J. (1947) Radiology, 49, 201.Roberts, J. A. F. (1940) An Introduction to Medical Genetics.

London.Sexton, D. L., Mack, R. (1954) J. clin. Endocrin. 14. 747.Silverman, S. H., Wilkins, L. (1953) Pediatrics, 12, 288.Stanbury, J. B. (1951) J. clin. Endocrin. 11, 740.

— Hedge, A. N. (1950) Ibid, 10, 1471.— Kassenaar, A. A. H., Meijer, J. W. C., Terpstra, J. (1955a)

Ibid, 15, 1216.— Ohela, K., Pitt-Rivers, R. (1955b) Ibid, p. 54.

Wilkins, L., Clayton, G. W., Berthrong, M. (1954) Pediatrics,13, 235.

VASOPRESSIN TANNATE IN OIL AND THE

URINE CONCENTRATION TEST

H. E. DE WARDENERM.B.E., M.D. Lond., M.R.C.P.

SENIOR LECTURER IN MEDICINE,ST. THOMAS’S HOSPITAL MEDICAL SCHOOL, LONDON

THE maximal ability of the kidney to concentrate theurine is generally considered to be a good indication ofrenal function, but to approach this concentration, atleast twenty-four hours’ deprivation of fluid is necessary(Miles et al. 1954). Deprivation of fluid, however, isseldom strictly controlled and depends on the variableamounts of water in the food and on the vigilance of thenurses. For the patient the test is nearly always unpleas-ant and occasionally, if a severe negative fluid-balancedevelops, it may be hazardous. The patient’s emotionalreaction to being deprived of fluid may also cause theresults to be quite misleading (Miles and de Wardener1953). Furthermore, because of its unpleasantness andpotential danger, the concentration test is nearly alwaysperformed on only one occasion, although its value wouldbe much increased if it were repeated. This naturaldisinclination to repeat the test is probably its greatestdisadvantage.To avoid these disadvantages Miles et al. (1954)

investigated the possibility of substituting an injection ofvasopressin tannate in oil for a long period of fluid

deprivation. The concentration of the urine after

deprivation of fluid was compared in the same subjectswith that obtained after the intramuscular injection of5 units of vasopressin tannate in oil. In 10 subjectsconvalescing from non-renal diseases the concentrationof the urine after deprivation of fluid was usually greaterthan after the injection of vasopressin ; the discrepancyin 3 of them being substantial; in 9 subjects with renaldisease there did not appear to be any difference betweenthe two methods. It was considered from these fewresults that vasopressin tannate was unlikely to be asufficiently reliable substitute for deprivation of fluidin testing the kidney’s maximal ability to concentrate theurine. Since these findings were reported, however, thesame experiment has been extended to 43 more subjects,of whom 17 were normal (Jones and de Wardener 1956)and 26 were patients, of whom 18 had renal disease and8 were convalescent from non-renal disease. The collectedresults from all these 62 subjects are given below ; theyconfirm that the concentration of urine after deprivationof fluid is greater than after the injection of vasopressin,but they also show that the difference is such that for