908 GASTRIC ULCER AND GASTRITIS Firstly, as far as we know, the designation " protein-calorie malnutrition of early childhood " was introduced in 1959 as SIR,-We have read with interest your leading article (Aug. a generic expression to cover a wide range of malnutrition on 27, p. 481), which evaluates recent evidence for a relation a community basis,’ including mild and moderate forms, as between gastritis, decreased gastric secretion, and gastric well as such severe degrees as kwashiorkor, nutritional ulcer. marasmus, and intermediate syndromes. The term continues Evidence for the association of chronic gastritis with gastric to have the advantage of emphasising the need for both ulcer is formidable,l2 and a causal relation between the protein and calories, in the diets of young children and also in gastritis and duodenal reflux has been inferred.3 But the the therapy of all forms and degrees of protein-calorie malnuquestion remains whether duodenal reflux invariably, or even trition, including the two severe syndromes of kwashiorkor in most cases, results in gastritis (which seems unlikely), or and marasmus. It would be unfortunate, for instance, if the whether it does so only when an abnormally susceptible gastric term " calorie deficiency " were to be used for human marasmucosa is present, such as would be expected in a candidate mus (although lack of calories is the principal error in the for gastric ulcer. Such abnormality could be caused by a poor diet), since this would imply to medical and other health quality blood-supply to the stomach resulting in impaired personnel that calories were all that were required to prevent secretory pattern, particularly of mucus or other protective or to treat this condition. Imprecise though the expression substances, which would then be replaced insufficiently P.C.M. is, it can have a definite value in assisting the thinking rapidly after the washing-off effect of the neutral or alkaline of those engaged in community nutrition. The term P.C.M. duodenal reflux suggested by du Plessis.3 covers both kwashiorkor, due dietetically to a low-protein diet In the context of animal work, duodenal reflux can probably containing carbohydrate calories, and nutritional marasmus, be considered along with substances such as alcohol, aspirin, due to a diet deficient principally in calories, but also in proand caffeine as ulcerogenic stimuli. Evidence for impaired tein. In other words, in practical terms in human malnublood-supply to the gastric region is provided by Elkeles,4 and trition, kwashiorkor can be considered as protein-calorie is an appropriate extension of the observations of Stehbens-5 malnutrition, and marasmus as calorie-protein malnutrition. on the abnormal microscopic appearances of certain bloodLack of, and dysbalance between, both nutrients figure in vessels in chronic peptic ulcer. This of course appears to be a each. Professor Platt prefers the designation " protein-calorie specific and important aspect of the general concept of lowered deficiency", which has been adopted by F.A.O./W.H.O., in mucosal resistance previously postulated. 67 part because it is tidier and more logical to consider malnuSubmaximal secretion is, as you suggest, perhaps at first trition of all nutrients as due to either " deficiency " or a surprising bed-fellow for gastric ulceration, especially sight " excess ". The main point, as Professor Platt remarks, is in the light of earlier beliefs. There is nevertheless adequate that for practical purposes protein and calories are always evidence for this association. 8-10 Although earlier beliefs interdependent, as stressed by the hyphen between the two were greatly concerned with the alleged effects of acid, to words. assume they implied that a role for acid in ulcerogenesis meant Secondly, though clear-cut polar cases occur, most children that gastric ulceration should be accompanied by hypersecrehave mixed deficiencies, either as intermediate severe syntion was going further than the evidence on which they were dromes or, more usually, as mild to moderate forms, which founded. defy short precise nomenclature. The protein deficiency and We have shown 11 that in the pylorus-ligated anaesthetised calories deficiency either may occur simultaneously with one guineapig of suitable strain a low dose of histamine acid or other predominating, or can, as it were, develop " in series " phosphate (2-5 mg. per kg. subcutaneously) will stimulate with one primary and initial, and the other secondary and both volume and acidity of gastric-juice secretion, causing at supervening. The latter may occur, as exampled in animals the same time only marginal acute gastric ulceration. When by Professor McCance and Dr. Widdowson, when loss of the dose is increased to 5 mg. per kg. subcutaneously, howappetite supervenes in a predominantly protein-deficient ever, severe ulceration occurs, but this is accompanied by child, or when a marasmic infant is stuffed with a high calorie gastric secretion which is significantly less in volume and intake on weaning. In addition, in any one child there are acidity than that stimulated by the non-ulcerogenic dose. often other complications, due to loss of protein and Pyloric ligation in this experiment ensures accurate measurecalories, confusing the picture still further. Diarrhoea, mal- ment of secretion volume, and permits the calculation of a rate absorption, and infection often occur in association with mal- of gastric secretion over the period of the experiment; furthernutrition, and further affect protein and calorie metabolism. more it also prevents the possibility of duodenal reflux which Pharmaco-genetic differences may play a part in influencing might otherwise confuse the results. the disturbances of protein and calorie metabolism in an We have interpreted the decrease in secretion as a maniindividual child by any specific degree of deficiency. Furtherfestation of mucosal exhaustion following the effects of a more, with the complex differences of diet in various parts of supramaximal dose of histamine on the gastric vasculature, the world, it may well be that imbalance of specific aminoand the ulceration as the result of the action of a decreased, acids or other factors as yet unknown may occur. Kwashiorkor but still acid, gastric secretion on the devitalised mucosa. in the Caribbean presents as a somewhat different clinical We wonder whether in man gastritis, however caused, syndrome from that usually seen in Uganda. supervenes only in a defective mucosa. It may be that a temIn summary, we welcome the warning that the truth must porarily defective mucosa can carry a temporary gastritis not be obscured by the abuse of too generic a term. Proteinwhich does not develop into ulceration because of restoration calorie malnutrition (or deficiency), however, remains a useful of the mucosa or withdrawal of the stimulus before ulceration descriptive term for the overall problem, particularly in the occurs, but that a chronically and inherently defective mucosa teaching of its prevention in the community and public-health may readily fail to overcome the gastritis, with the result that fields. All the same, individual children with protein-calorie in the continuing presence of the particular stimulus, chronic malnutrition (or deficiency) always need more specific diagnosis ulceration ensues. Mackay and Hislop2 found that gastric and grading into mild, moderate, or severe degrees of involve1. Deutsch, E., Christian, H. J., Fabregas, R. Am. J. Gastroent., N.Y. ment, and, in the last group, into the identifiable syndromes of 1962, 37, 168. 2. Mackay, I. R., Hislop, I. G. Gut, 1966, 7, 228. kwashiorkor or marasmus, where possible.2 Department of Pædiatrics and Child Health, Makerere Medical School, D. B. JELLIFFE P.O. Box 2072, J. P. STANFIELD. Kampala, Uganda. 1. Jelliffe, D. B. J. Pediat. 1959, 54, 227. 2. Jelliffe, D. B. The Assessment of the Nutritional Status of the Country. W.H.O. Monograph. Geneva, 1966.
3. du Plessis, D. J. Lancet, 1965, i, 974. 4. Elkeles, A. Lancet, 1965, i, 1069. 5. Stehbens, W. E. Archs Path. 1964, 78, 584. 6. Illingworth, C. F. W. J.R. Coll. Surg. Edinb. 1956, 2, 14. 7. Marks, I. N., Shay, H. Lancet, 1959, i, 1107. 8. Grossman, M.I. Am. J. Med. 1960, 29, 748. 9. Welbourn, R. B., Burns, G. P. ibid. 1965, i, 1069. 10. Hunt, J. N. ibid. 1950, ii, 397. 11. Anderson, W., Soman, P. D. J. Pharm. Pharmac. 1965, 17,
909 cancer is frequently impalpable, often because it thin sheet of cancer, or because there are several tiny
ulceration could heal in the presence of continuing widespread
think there is no good evidence for accentuat the expense of, for example, pepsin(s) or any other component of gastic secretion. Mucosal resistance to autodigestion has been ably discussed by Montagu.12 We therefore believe that the evidence from both clinical and experimental work 13 precludes at present the dissociation of gastric secretion and vascular factors, and the accentuation of one or other of them, in the ætiology of gastric ulcer. We also wonder to what extent a " normal " gastric secretion is a desirable accompaniment to therapy, and also whether wholesale attempts completely to eliminate or neutralise components of gastric secretion are enlightened. Department of Pharmacy, W. ANDERSON University of Strathclyde, P. D. SOMAN. Glasgow C.1.
ating the role of acid
CARCINOMA AND EARLY PERNICIOUS ANEMIA SIR,-Accepting Dr. Morson’s evidence that intestinal metaplasia may be premalignant (Oct. 1, p. 753), one still asks oneself what it is about this microscopic appearance that predisposes to cancer of the stomach. As one measure of turnover of gastric surface cells my co-workers and I 14 performed mitosis-counts in normal and abnormal gastric biopsies. Of the five highest mitosis-counts, four were in areas of intestinal metaplasia, as shown by the following results:
together. H. S. SHUCKSMITH J. A. DOSSETT.
James’s Hospital, Leeds 9.
REOVIRUS 3 AND LYMPHOBLASTIC LYMPHOMAS SIR,-The closing remark of one of your leading articles on the lymphoblastic lymphomas of childhood1 prompts me to communicate to you some of the recent developments in St. Louis, Missouri. I should like first to comment on the recent observations of Achong and Epstein.2 After a study of the fine structure of the Burkitt tumour, in biopsy material from Uganda, they reported the presence of nuclear projections in the tumour cells, which they regard as lymphoblasts. These projections of the nuclear envelope corresponded to those of cultured lymphoblasts of established in-vitro stains, and were thus interpreted as an inherent attribute of the Burkitt tumour and not the result of cultivation in vitro. Similar nuclear projections were noted by Stanley3 and his collaborators in lymphoblasts from a murine lymphoma induced by neonatal infection with reovirus 3. In April, 1966, laparotomy was performed on a two-year-old boy (from St. Louis, Missouri) with an abdominal tumour. ’A frozen-section diagnosis of malignant lymphoma, with a starry-sky pattern " was made on biopsy of a tumour nodule in the omentum. Electron microscopic examination of this tissue revealed the presence of lymphoblasts with nuclear projections identical to those described by Achong and Epstein (fig. 1). The initial manifestations of the disease in this child, "
This suggests, but does not prove, that there may be parturnover in these areas; the notably abnormal turnover, which appears to be even higher than in atrophic gastritis, could be an important factor linking intestinal metanlasia and gastric cancer. St. Thomas’s Hospital, D. N. CROFT. London S.E.1.
MAMMARY LIPOMA AND CARCINOMA SIR,-Dr. Behan (Oct. 1, p. 753) asks whether the findings in 4 women of a lipoma and an underlying cancer are fortuitous. Korner 15 described the association of a fatty tumour, which he called a pseudolipoma, with cancer of the breast in 4 patients in 1956. We 16 were not aware of Korner’s paper when we independently used the term pseudolipoma for the fatty tumour found in association with cancer of the breast in our description of 18 examples of pseudolipoma in 1965; we considered that such a fatty tumour was not a true lipoma, for there is no overgrowth of adipose tissue. The pseudolipoma is nearly always produced by compression of the fat lying superficial to the breast disc when the fibrous septa, which connect the breast disc to the skin, are drawn together by the contraction of underlying cancer or cancers; less commonly the pseudolipoma is produced by oedema of the fatty tissue overlying a cancer surrounded by an inflammatory reaction. The finding of a pseudolipoma is fairly common, and we now have 8 further examples, making a total of 26 (13%) in 201 consecutive breast cancers; of these 26, 11 (42%) contained multiple cancers. Duct ectasia produced a pseudolipoma by contraction of postinflammatory periductal collagenous tissue in 3 further examples, and a true lipoma was found with no underlying breast lesion in 2 other patients. The fact that a pseudolipoma is commonly associated with an underlying cancer is of particular importance, for the 12. 13. 14. 15. 16.
Montagu, K. Nature, Lond. 1966, 209, 1206. Anderson, W., Soman, P. D. ibid. 1966, 18, 58. Croft, D. N., Pollock, D. J., Coghill, N. F. Gut, 1966, 7, 333. Korner, W. Zentbl. Chir. 1956, 81, 31. Shucksmith, H. S., Dossett, J. A. Br. med. J. 1965, ii, 1495.
Fig. 1-Electron micrograph of biopsy material from child with lymphoblastic lymphoma showing nuclear projection in lymphoblast ( x 41,800). appearances of the tumour, and the fine strucof the tumour cells were typical of Burkitt’s lymphoma. Subsequently, however, the child developed hepatosplenomegaly, mediastinal lymphadenopathy, pleural effusions, involvement of the small bowel, symptoms referable to the central nervous system, and terminally lymphoblastic leuksemia. Lymphoblasts in peripheral blood and bone-marrow smears contained cytoplasmic vacuoles identical to those described by Wright in tumour cells of the Burkitt lymphoma.4 Necropsy confirmed widespread involvement of the central nervous system, associated with perivascular demyelination in the white matter of the brain. A detailed report of the fine structure of this tumour has been submitted elsewhere for publication. Two weeks ago a four-year-old girl died in St. Louis Children’s Hospital after a stormy and fulminating disease lasting ten weeks from onset of symptoms to death. There was
Lancet, 1966, i, 965. Achong, B. G., Epstein, M. A. J. natn. Cancer Inst. 1966, 36, 877. 3. Stanley, N. F. Lancet, 1966, i, 961. 4. Wright, D. Br. J. Cancer, 1963, 17, 50.