Z. Kinderheilk. 113, 23'7--256 (1972) 9 by Springer-Verlag ][972

Meeting of the International European Society for Pediatric Haematology ,Abstracts o f t h e T h i r d M e e t i n g S e p t e m b e r 9 - - 1 0 , 1972 a t H a m b u r g immunology and the Gut. J. R. Hobbs, London The Normal Immune I~eactions o] the Gastrointestinal Tract. The first line of humoral reaction resides mainly in the secretory IgA system, with a second line in IgS[. IgG and IgD seem relatively unimportant to the gut. IgE is more often a nuisance than a help. Secretory IgA shows marked regional localisation and prolonged memory, with a backwash of informational 7s IgA into the body. Immune Deficiencies. Pure cellular (T-cell) deficiency has no obvious gastrointestinal symptomatology but combined cellular and humoral deficiency always shows intractable malabsorpticn indicating that the T-cells must have a very important cooperative role. Pure IgA deficiency can be compensated for by IgM but where both are deficient the gut usually shows the strain, and nodular lymphoid hyperplasia probably represents attempted T-cell compensation. Pure IgG or IgD deficiency have no gut problems. Aberrant Immunity and the Gut 1. Pathogenetic Roles Established. a) Pernicious anaemia. Glass and Tanaka (1970)

have shown how parietal cell antibodies alone can actually cause gastric atrophy and Taylor et al. (1962) how antibodies to intrinsic factor can block B12 absorption. Fixa et al. (1970) have demonstrated cellular attack on gastric antigens. b) Primary biliary cirrhosis. Complement-fixing IgM antibodies, Paronetto et al. (1964-1970). Cellular reaction to gall-bladder-wall (Saarma, 1970). 2. Probable Roles. a) Coeliae disease. Impaired IgA (Beale et al., 1971), use of IgM (Mietens, 1967; Douglas et al., 1969) to produce jejunal (Katz et al., 1968) antibodies to gluten. Local Arthus reaction and antibodies co-operating with lymphocytes (Hobbs and Fakhri, 1972). b) Ulcerative colitis. Cellular reaction to colon (Bendixen, 1969) and antibodies cooperating with lymphocytes (Shorter et al., 1970). 3. Possible Roles ( ? immune reaction added to provocative agents). a) Crohn's disease, b) Chronic active hepatitis. Desiderata

I. Post-hoc immunological events (IgM deficiency, raised IgA, catabolic IgG deficiency, T-cell impairment etc.) must be clearly distinguished from propter-hoc (parietal cell antibodies etc.) which should be demonstrable in nearly all affected patients, not just 20%. The same patients should be studied both in relapse and remission. IL Regional localisation of disease must be satisfactorily explained. If milk allergy damages the colon or faecal agent causes Crohn's why is the rest of the exposed gut normal ?

238

meeting of the International European Society for Pediatric Haematelogy

III. Why others with similar reactions (e.g. antibodies to gluten) do not have lesions must be explained. IV. What is the predisposition of the patient to the disease ? (e.g. IgA deficiency, genetics etc.). V. The timing of the postulated mechanism(s) should fit clinical observations of relapses. Immunological Reactions in Coeliae Disease. Adrian P. Douglas, Newcastle Upon Tyne The University of Newcastle Upon Tyne, Newcastle Immunological reactions can be measured at the circulating level and at the local level. The circulating level may not truly represent what is happening at the local level in the upper small intestine. Furthermore immunological abnormalities in coeliac disease may be secondary rather than primary phenomena. At the present time there are many fragmentary observations: clinical, related to circulating immunoglobulin measurements, measurements of local immunoglobulins, measurements of complement locally and systemically, and measurements of cell-mediated immunity and tissue antigens. Which all suggest that immunological mechanisms are important in this disease. We have previously suggested that coeliae disease may represent a failure of the masking action of IgA antibodies against dietary proteins perhaps duo to a qualitative defect in IgA [1]. Damage might then result by means of an Arthus-type reaction and experimental data has been published suggesting that such a reaction may well have occurred in coeliae mueosa [2, 3]. The significance of these observations remains unclear. The specific idiosyncratic effect of gluten remains unexplained. Although immunological reactions may well be important in the pathogenesis of the intestinal lesion the aetiology of coeliac disease is still uncertain. 1. Bearle, A. J., Parish, W. E., Douglas, A. P., Hobbs, J. R. : Lancet 1971 I, 1198-1200. 2. Doe, W., Henry, K., Holt, L., Booth, C. C.: Gut 13, 324 (1972). 3. Shiner, M., Ballard, J. : Lancet 1972 I, 1202--1205. Populations of Small-Intestinal Lymphoeytes. A. Fergnson and D. M. V. Parrott, Glasgow From studies of the development of intestinal lymphoid tissues of mice, we have been able to define thymus-dependent (T) and thymus-independent (B) intestinal lymphocytes. Our results show that the majority, but not all, of Peyer's patch and villous intraepithelial lymphocytes are thymus-independent. The effects of antigen-Deprivation were observed in hcterotopically implanted grafts of foetal mouse intestine. In these grafts, Peyer's patches were small and lacked germinal centres; intraepithelial lymphocytes and lamina propria plasma cells were present but their numbers were markedly reduced. These experimental observations, made in mice, correlate well with the morphological findings in children with primary immunodeficieney syndrome. Antiretieulin Antibodies in Childhood Coeliac Disease. G. Della Cella, M. Governa, and 1). Durand, Genoa Various immunological abnormalities have been shown in patients with eoeliae disease. Among them are an increase of antighiten antibodies in small intestinal

Abstracts of the Third Meeting

239

cells shown by immunofluoreseence (Malek et al., 1964), an increased number of immunoglobulin-G containing cells in the small bowel mucosa of treated or untreated coeliacs and alteration of the levels of circulating immunoglobulins. More recently autoantibodies reacting with reticulin or basement have been described. Sera of a group of twenty-three children with coeliac disease have been studied in older to verify whether there were antibodies to reticulin. The sera were examined by standard indirect immunofluorescent methods, using unifixed cryostatic sections of rat liver and stomach. Moreover antibodies to the same dietary antigens - - i.e. milk, bovine and egg proteins - - were detected. Antibodies to reticulin were found in about a quarter of the sera. The antibodies were not present in the sera of normal children. There was a :positive correlation of retieulin antibodies to other dietary antigen antibodies. In addition, we have shown that such antibodies were absent in an ever higher proportion of children with other type of chronic diarrhoea. The reticnlin antibody was not as specific for celiac disease as in the studies of Seah et al. (1971). The role of reticulin antibodies is uncertain. They may be a direct consequence of intestinal tissue damage. They may reflect absorption of retieulin antigens from the diet. Amman and Hong (1971) suggested that they might be similar to antibodies to basement membrane in rat liver and kidney which they demonstrated in patients with celiac disease and selective IgA deficiency. All these findings suggest that immune mechanism are involved in the pathogenesis of coeliae disease, but that it is not an autoimmune disease. Some changes suggest that the effect of gluten is not directly on the epithelium. I t could be via a cellular-mediated immune mechanism, but further studies are certainly needed, particularly in young newly diagnosed cases, before this controversial theorie can be finally setled.

The Value of Serum Antibodies and Lymphoblast Transformation Studies as Diagnostic Tools in Gluten Sensitive Coeliae Disease and in Cows Milk Intolerance. J. J. Baudon, J. F. Mougenot, and N. Desquilbet, Paris The serum antibodies to purified a-lactalbumine, fi-lactoglobuline, easeine, and pepsinized and trypsinized gliadin were determined in children with coeliae disease and cows milk intolerance. The method used was passive haemagghitination with benzidine treated red c,.~lls. Simultaneously lymphoblast stimulation studies were performed using 0.5~o a-lactalbumine or /~-lactoglobnline as stimulating agents. The patients were predominantly children below two years of age with coeliae disease or cows milk intolerance. a) Serum Antibodies We observed in 31 children with cows milk intolerance and in 36 children with gluten sensitive cocliac disease a similar percentage of positive results for serum antibodies to a-lactalbumine, to/~-lactoglobuline or to both substances (only titers above 1/128 for the milk antibodies and over 1/8 for gliadin antibodies were regarded as abnormal). Gliadin antibodies were found only in one child with milk intolerance, but antigliadin antibodies higher than 1/16 were found in 56~/o and antibodies higher than 1/8 in 680/o of the patients with coeliac disease. We have also obsel~ced numerous pathological results of serum antibodies in children with non-specific diarrhoeas. This phenomenon is thought to be due to a higher permeability of the small intestine. The appearance of these antibodies is therefore not regarded as a disease-specific phenomenon.

240

Meeting of the International European Society for Pediatric Haematology

In a few cases kopro-antibodies were also investigated and the increase of these substances in the stools following reintroduction of milk or gluten into the diet appears to be an useful diagnostic sign.

b) Lymphoblast Stimulation Lymphoblast stimulation induced by milk antigens appears to be a more specific test and we have found lymphoblast stimulation of more than 4% in twelve out of 18 children with cows milk intolerance and more than 6~o in 10 out of the 18 eases. By contrast none of the patients with coeliac disease has shown lymphoblast transformation induced by milk antigens. But 4 of 8 patients with coellae disease had positive serum antibodies to milk proteins at the same time when the specific lymphobIast stimulation test was negative. In 14 of 17 infants the stimulation was induceable with fl-lactoglobuline and in 4 of these patients the test was also positive with lactalbumine. One infant showed a reaction of the anaphylactie type when on feeding with soiaprotein. He had a positive lymphoblast stimulation test with a-laetalbumine alone. After omission of the milk from his diet the test became negative but showed again a positive result to lactalbumine and fl-lactoglobuline after reintroduction of cows milk in his diet.

c) Conclusions The lymphoblast transformation test appears to be a more specific diagnostic aid than the determination of serum antibodies in coeliac disease and in cows milk intolerance. I f it should hold true that the test using fl-laetoglobulines as stimulants is always negative in coeliac disease this test would represent a very useful diagnostic aid in differentiating between coehac disease and cows milk intolerance.

Immunohistoehemieal Study of Jejunal Biopsies ot Children with Coeliac Disease. E. Savilahti, P. Kuitunen, and J. K. Visakorpi, Helsinki Although the aetiological role of gluten in coeliac disease has been known over two decades, the pathogcnesis by which it induces the intestinal damage has remained unsolved, there is ample evidence that altered or defective immunological mechanism of the gut may be involved, as indicated by elevation of serum IgA in eoeliae disease and the association of IgA deficiency and coeliac disease. In recent studies changes in the quantities of immunoglobulin-containing cells in the jejunal mucosa of adult coeliac patients has been noted. In this study the numbers of immunoglobulin-containing cells were studied by direct immunofluorescence in the small intestinal mucosa of 21 coeliac children with total vilous atrophy of the jejunum. 17 of the specimens were taken prior to dietary treatment and 4 after gluten provocation. 27 age matched children without intestinal diseases served as controls. IgA-containing cells were predominant in all except one specimen, where more IgM-containing cells were present. The number of IgA-eontaining cells was, on the average, twofold compared with the controls. The excess of IgM-containing cells was even more marked, three times more of these cells were seen in coeliac patients. Also the number of IgG-containing cells was twice that in controls. Clearly more IgE-eontaining cells were present in 13 specimens studied, their number was equal to that of IgG-containing cells in the controls. In 9 controls studied the number of IgE-eontaining cells was always less than 1/4 of the number of IgG-eontaining cells.

Abstracts of the Third Meeting

241

In 7 cases a new biopsy specimen was studied after a short period of gluten free diet (1--3 months). The number of IgA-containing cells fell almost to the level observed in the controls. There was some decrease in the quantity of IgM- and IgGcontaining cells, but their number was still significantly higher than in the controls. The study shows that in children with coeliae disease there is no quantitative defect in the IgA-producing system of the gut as has been suggested in studies concerning eoeliac disease adults. On the contrary producing cells falls fairly fast to normal levels. The strong increase of IgM producing cells in the gut could be due to exhaustion of the natural local IgA response or some qualitative, yet unknown, defect in this response, and this, together with elevation of IgG and IgE producing cells, could lead to the presence of antibodies of either complement fixing or reaginie type in the lamina propria of the jejunum with their injurious effect on the structures of the small intestinal mucosa.

Reduced Concentrations of Plasma fllG-fllA Globulines in Patients with Coeliac Disease. F. Carswell, R. W. Logan, and A. Ferguson, Glasgow The plasma fllG-filA Globulins (the C3 fraction of complement), IgA, IgG, and IgM concentrations were estimated by immunodiffusion in fifty-five children with untreated coeliac disease and in thirty-three other children. There was a significantly lower concentration of fllG-fllA Globulins and a significantly higher concentration of IgA in the plasma of the children with coeliac disease. There was no difference in the incidence of ini~ction of infestation with Giardia lamblia in the two groups. The eoeliac patients with tow fllG-fllA Globulins had normal titres of immunoeonglutinin. The relationship of plasma and duodenal juice food antibodies to the reduced plasma concentrations of the fllG-/~IA Globulins was examined. This reduction in ~IG-filA Globulins concentrations is consistent with either reduced syntesis or increased consumption of templet in eoeliac disease.

Immunological AspeclLs on Ulcerative Colitis and Crohn's Disease. R. Lagercrantz, P. Perhnann, and S. ttammarstrSm, Stockholm Etiology and pathogenesis of ulcerative colitis and Crohn's disease are unknown. Many clinical data, in, vitro studies and animal experiments indicate that immunological mechanisms play a role in both diseases. Most, if not all patients with ulcerative colitis haw~ humeral antibodies to an antigen which is present in the mucosal cells of the colonic crypts and which is also secreted into the mucus: The purified colon antigen is a mucopolysaccharide, chemically closely related to the blood group substances of the human AB0-system. However, the antigenic determinants with which the patient's antibodies react are distinctly different from the known determinants of the blood group system. Antibodies to the ulcerative colitis antigen are not found in healthy individuals, nor are they found in patients with other diseases, including those of the gastrointestinal tract. However, a notable exception are patients with Crohn's disease, even those without colon involvement. Detailed studies of the human anticolon antibodies indicate a relationship of the ulcerative colitis antigen to the cross reacting antigen of E. coli. ("Common antigen"), present in most strains of bacteria belonging to the large family Enterobacteriacae (E. eoli, Salmonella and Shigella). All individuals get early in contact with this antigen. Susceptible :individuals can thereby loose their tolerance to the similar but not identical colon-antigen and develop auto-immunity to colon. Female close 16 Z. Kinderheilk., Bd. 113

242

Meeting of the International European Society for Pediatric Haematology

relatives to patients with ulcerative colitis also have a significantly higher antibodytiter to colon-antigen. This may indicate a genetic predisposition to autoimmunity and possibly also to the disease or - - possibly - - a common environmental factor. Humoral antibodies to colon are not cyto-toxic in vitro to colon-tissue. However, patient's lymphocytes (from peripheral blood and regional lymphnodes) are. Although their relative role for the appearance of tissue damage in vlvo is largely unknown, this data support the assumption that auto-immunity in ulcerative colitis is potentially harmful. In Crohn's disease some results indicate a depressed function of the thymns-dependent lymphocytes. For references see Proceedings of International Congress of Allergy, Florence, Oct. 1970, in press, and Perlmann, P., Lagercrantz, P~., and Hammarstr6m, S.: Textbook of Immunopathology, Sec. Ed., Vol. II. New York: Grune and Stratton (1972, in press).

Acquired, Non-Leueemie Aplastie Anemia in Childhood. M. Seip, Oslo A review is given of the etiology of acquired aplastic anemia (drug exposure, chemicals, infections, irradiation and "idiopathic" acquired aplastie anemia (about 50~/o). There are different theories of pathogenesis as toxic damages to stem cells, damage to stroma or mierocirculation or immunological mechanisms but little exact knowledge is available. The management extends on removal of offending agents if known, prophylactic antibiotic therapy, hormon therapy (androgens and eorticosteroids), transfusions of whole blood, packed cells, platelets. If repeated platelet transfusions are necessary it is preferable to use a donor of the same AB0 and HL-A type as the patient. Spleneetomy may be indicated in a few selected cases with certain signs of splenic destruction. Bone marrow transplantation is still an experimental approach with great hazards.

Evolution and Prognosis of Pancytol)enia in Childhood. W. Schr6ter, U. Dellbriigge, and G. Landbeck, Hamburg In a period of ten years 33 cases of idiopathic pancytopenia were treated in the Department of Pediatrics, University of Hamburg. The bone marrow index leukopoiesis/erythropoiesis makes prediction possible of the course and the evolution of the disease. I n patients whose index is five or less (----hyperplastie erythropoiesis) a chronic course predominates according to the definition of Bernard and Najean. The prognosis of these eases is better than that of patients whose index is higher than five. This latter group dies witlfin a few months. The paneytopenia with hyperplastic erythropoiesis is three times more frequent than that with hypoplastic erythropoiesis. Anemia, relative reticulocytosis and elevation of the hemoglobin F concentration are m u c h more pronounced in this group, whereas granulocytopenia and thrombocytopenia are less prominent. Etiological differences between the two groups could not be demonstrated. 4 of the 33 cases developed leukemia. In about half of the cases with hyperplastic erythropoiesis a relative pyruvate kinase and glutathione reductase deficiency of the red cells was found. These defects are obviously not genetically determined but are acquired and of transient nature. Their cause and pathogenetic significanceis not yet clearlyunderstood.

Abstracts of the Third ~eeting

243

Studies on the Pathogenesis and Treatment of Chronic Y[yelogenous Leukemia. R. A. Gatti, W. A. Robinson, A. S. Deinard, M. Nesbit, 1~. Ballow, and 1~. A. Good, ~Minneapolisand Denver Long-term studies were performed on a 12-year-old girl, an identical twin, with cyclic leukocytosis and Ph 1-positive CML in an attempt to analyze the relationship of these cycles to her disease and to utilize the cycles as landmarks in the progression of her disease. She remained untreated for 2 ~ years. Colony Stimulating Factor levels showed an inverse relationship to the peripheral leukocyte count. Two distinct phases in cyclic patterns of peripheral leukocyte counts could be appreciated in retrospect: Phase I, a 15-month period during which leukocyte counts peaked every ten weeks and then returned to base-fine levels and Phase II, an 18-month period of insidious deterioration during which the low-points of successive cycles became progressively higher and leukocytes were accumulating in the peripheral blood at a rate of 400 • 106/day. It appeared that the cyclic increase in production of myeloid cells in this patient did not represent a life-threatening situation until remora| of myeloid cells from the peripheral circulation began to fall behind production rates. Treatment was, therefore, aimed at assisting cell removal by cell separa~bor leukophoresis instead of by conventional chemotherapy. The patient has remained asymptomatie since leukophoresis 11 months ago.

Leukaemia-Like, Lethal Myelopathy Presenting in Two Newborn Sisters. G. Sansone, A. Rasore-Quartino, and A. Braito, Genoa A hitherto undescribed, possibly genetical disease, affecting two siblings is described. The onset was characterized by pallor, anorexia, vomiting, failure to thrive. The liver was enlarged, spleen and lymph nodes were normal. The infants had severe normochromic anemia, normal white blood cells, neutropenla, thrombocytopenia. A highly cellular bone marrow showed undifferentiated multinucleated cells, early myelocytes and atypical myeloblasts with distorted nuclei and scarce granulations, large polymorphonuclear cells with polysegmentcd nuclei. Megaloblasts and macroblasts were evident. Atypical mytoses were occasionally present. Chromosome breaks were visible in a blood marrow culture. Postmortem examination showed the absence of cellular infiltration in the liver, spleen and lymph nodes. The diagnosis of congenital leukaemia did not fit with clinical and laboratory findings. It is possible that an unknown enzyme defect has deeply altered the normal development of the haemopoietic elements in the bone marrow.

Leukaemia with Panhypoplastie Marrow. H. J. Pliiss and W. H. Hitzig, Ziirich The peripheral pancy~openla displayed by 1/.2 of the children with acute leukaemia is linked with increased bone marrow cellularity. Panhypoplasia of the marrow in the initial sample (!i.e. before any treatment) is rare: we observed it in only 4 out of 150 children with acute leukaemia diagnosed from Jan. 1964 to June 72. Two of them, in addition, had no increase in the blast cell count, and the diagnosis of leukaemia could only be made after 1 and 3 months resp. of observation. The percentage of blasts at diagnosis was less than 50 in all 4 patients. The cells were classified in one patient as stem cells, in 2 as acute lymphoblastie (ALL) and in one as acute myeloid (AML). Only this last child failed to respond to therapy, whereas the other 3 went into complete remission with standard cytostatic treatment. 16"

244

Meeting of the International European Society for Pediatric Haematology

A comparison of the initial blood and bone marrow pictures with those of 5 children with idiopathic pancytopenia, admitted during the same period, showed no difference in 2 of the 4 leukaemie children, but observation without treatment showed an increase in blasts and/or a rapid normalization of the peripheral blood not observed in idiopathic pancytopenia. As all 4 patients had a history of drugs, we assume that these substances have caused a short phase of agranulocytesis superimposed on the early signs of leukaemia.

Morphological Aspects el Human Lymphoeytes. H. K. Miiller-Hermclink, Kiel Lymphocytes are disseminated throughout the mammalian organism partly staying within the structural organization of the lymphatic organs, partly migrating through blood and tissues. New experimental approaches concerning lymphocyte function lead to the concept that 1. lymphocytes represent a heterogenous population and 2. that the immunological homoiostasis of the body is achieved by the recirculating capacity of lymphocytes. Having in mind the important progress of modern immunology, morphologists, at the present time, are not capable to demonstrate the exact cytological equivalents of the functional status of lymphocytes. Moreover the situation is complicated by the fact that severe alteration of lymphocyte morphology occur during any lymphocyte stimulation. Nevertheless by this presentation we intend to demonstrate what we know about lymphocyte morphology and what are the parameters that may lead to a morphological differentiation of distinct lymphocyte subpopulations. For this purpose we investigated random mixtures of lymphocytes from human blood, tonsils, thymus and lymph nodes in order to reduce during evaluation morphology on pure cytological characteristics. The methods employed have been: light microscopy of fixed smears, light-microscopical cytochemistry (PAS, acid phosphatase, non-specific esterases), electron microscopy, electron microscopical cytochemistry (acid phosphatase, ATPase, 5'nucleotidase, staining of glycocalyx) and morphometry. Each of the applied methods reveals a heterogeneity of lymphocytes, expressed by structural differences, content of lysosomes, differences in enzyme activity or content and organization of the lymphocyte membrane. Thus we may differentiate at least two lymphocyte populations within the blood differing in the applied parameters. As far as possible we will give a mutual correlation of the results with different methods. However, this is not complete and the value of the techniques varies from one to another, as e.g. differences of enzyme activities occur during lymphocyte stimulation. Our investigation leads to the following implications" 1. Exact morphology reveals reproducable cytological differences among lymphoeytes, 2. correlation to immunological functions is not achieved and thus 3. lymphocyte morphology in terms of function, today, is as valuable as the exact immunological circumstances are known.

Morphological Differentiation of Lymphoeytes in the Peripheral Blood. E. Arabs and R. Oronemeyer, Wiirzhurg Nucleoli of lymphocytes can be demonstrated by methylenblue staining according to the method of Pischinger modified by Stockinger and Keliner. This allows their classification in macro- and polynueleolar lymphocytes (depending on their loealization in lymph nodes Grundmann distinguished follicular and sinus lymphoey~es). The application of tuloidinblne staining according to Plenert made it possible to differentiate peripheral lymphoeytes populations of children without any haematological diseases, treated for longer periods with eorticosteroids or eytostatic drugs

Abstracts of the Third Meeting

245

(immunsuppressive therapy): compared with control groups (normal persons, children with bacterial infections treated with antibiotics, partly for a short period with cortieosteroids) the percentage of macronucleolar lymphocytes increased while polynucleolar types diminuishing. An interpretation is not yet possible.

Histoehemieal Studies on Thymoeytes. K. Kouvalainen and O. lCuuskanen, Turku Gninea-pig cortical thymocytes show regularly a very high alkaline phosphatase (AP) activity. AP seems to be present in the cortical thymoeytes from the very beginning of the development of the thymus. Very few medullary thymocytes show AP-activity. About one per cent of the lymphocytes in blood are AP-positive. Several months after thymectomy there is a trend towards lower numbers of APpositive lymphocytes in blood. During the intrauterine life highest numbers of AP-positive lymphocytes in blood are seen a few days before and at birth. The levels drop to adult values in a few days. There are a few AP-positive lymphocytcs in the spleen, and about one per cent of the lymphocytes in lymph nodes are AP-positive. There is no correlation between the numbers of AP-positive lymphocytes and the cells showing the so-called Pompidou-phenomenon(Lancet 1971 II, 245). Our studies indicate that cortical thymocytes are not histochemicaily identical with the main population of lymphocytes in the thymic medulla and the peripheral lymphoid organs. Neither are they identical with the so-called thymus-derived cells. The function of cortical thymoeytes and AP-positive lymphocytes outside the thymus remains to be clarified.

Surface Bound Immunoglobulins on Lymphocytes with Special Reference to Immunodeficiency Diseases. ~{. Seligmann and J. L. Preud'Homme, Paris The presence of membrane bound Ig on the surface of normal lymphocytes is well documented in different species including man. Although thymus derived lymphocytes apparently carry on their surface small amounts of Ig which are directly detectable only by very sensitive methods, those lymphocytes which bear surface Ig in sufficient amounts to be evidenced by rather unsensitive technics such as direct immunofluorescence have been shown to be bone marrow or bursal derived (B) lymphocytes. Igl~ is the main Ig class on peripheral blood lymphocytes of normal individuals. The value of surface Ig as B cell markers is evidenced by the results of the study of patients with lymphoproliferative diseases. The surface Ig are receptors for antigens. For instance, the monoclonal surface IgM was shown in several patients with ]ymphoproliferative diseases to possess an anti-IgG antibody activity. We have studied by direct immunofluorescence surface Ig on the living blood lymphocytes of 50 patients with primary immunodefieiencies. Ig bearing lymphocytes were practically not detectable in sex linked agammaglobulinemia and in several cases of variable immunodeficiencies with humoral predominance, irrespective of the degree of infection. The absence of surface Ig bearing lymphocytes may be due either to ~ feilure of B lymphocytes to incorporate Ig into their membrane or to an actual absence of B lymphocytes which could itself be the consequence of the absence of precursor cells or of a block in the maturation process of B lymphocyte at an early stage. In contrast, normal or increased numbers of lymphoeytes carrying Ig in the absence of the corresponding plasma cells were found in several cases of variable immunodeficieneies, particularly in those with lymph node follicular hyperplasia and during infections. Similar results were obtained in

246

Meeting of the International European Society for Pediatric Haematology

selective IgA deficiency and ataxia telangiectasia. This situation is strongly suggestive of a defect in the maturation process involving the conversion of a B lymphocyte into an antibody producing plasma cell. In Wiskott-Aldrich syndrome, the cellular findings reflected the serum Ig pattern. Very high numbers of Ig bearing lymphoeytes with a fair percentage of double producers for y and ~z-chains, were a striking finding in a neonate with severe combined immunodeficiency.

Immunologic Capacities of an Infant Born- and Nursed for Four Weeks under Germfree Conditions. L.J. Dooren, J.M. Vossen, and D. van der Waay, Leiden and Rijswijk An infant with a family history suggestive for combined immunodeficlency was delivered by caesarean section and nursed during a period of six weeks using laminar flow technique for protective isolation. Extensive microbiological examination showed the infant to be germfree during a period of four weeks. Several parameters of humoral and cellular immune capacities of the infant, examined during the germfree period, were normal. Contamination was performed at the age of four weeks with a faecal flora consisting of only two potentially pathogenic microbial species, with low pathogenicity and a favourablc sensitivity pattern to antimicrobial drugs. The immunologic parameters were followed after the contamination and after discharge at the age of six weeks, and showed a normal immunologic development. Follow up of the physical and psychomotor development until the age of one year did not show any adverse effect of the short-duration period of gnotobiotic life.

Sterile Care in Pediatrics. U. DShmann, H. J. Pl/iss, E. Loher, and W. H. Hitzig, Zfirich Patients with congenital or acquired immunodcficienciesare especially susceptible to infection. Environmental microorganisms easily lead to exogenous infections; these are through the usual isolation methods partly, through strict isolation completely avoidable. On the other hand, endogenous infections, especially from the patient's G-I tract, might only be prevented by partial or total elimination of his own microflora (decontamination). We nursed five patients in a m~it with sterile laminar airflow" two newborns with a family history of immunodeficiency were delivered by caesarian section and sterile conditions maintained after delivery. Decontamination was relatively easy with two of the three other children. Nursing period was between twelve days to eight weeks. - - The considerable costs and personnel efforts are justified in those cases where prevention of infections would substantially contribute to a better course of the illness. The expected psychological problem due to isolation was surprisingly little.

Experimental Studies of the Circulation and Migration o~ the T and B Lymphocytes and Their Potential Application in Immunodeticiencies of Man. D. Guy-Grand, P. Vassalli, and C. Griscelli, Paris Chronic drainings of thoracic duet lymph and transfer of thoracic duct lymphocytes (TDL) were performed in rats and (or) mice. B and T cells were detected in TDL and in recipients after transfer by a combinatlon of isotopic and fluorescent tracers using rabbit anti-MBLA (anti B); antiL Ig chains, anti-aIg chains and anti-MSLA (anti T) sera.

Abstracts of the Third Meeting

247

Proportion of T and B cells of the lymph and the blood were determined in normal and stimulated animals 4, 24 hrs and several days after starting the lymph collection. Patterns o!~TDL migration into lymphoid organs were observed. Our experiments have shown: 1. The larger proportion of T cells in the lymphatic circulating pool (65 to 75%). 2. Increase o f B cells proportion in the remaining blood and lymphatic lymphocytes after a long term draining. This results are in correlation with other studies which have shown a better circulation of the T c~lls than the B cells. According to this observation we have explored the immunological picture in intestinal lymphangectasia in which hypogammaglobulinemia is due to the protein loss without appreciable B cells decrease and, lymphopenia, the consequence of the T cells deperdition.

Drugs and Lymphocyte Blastogenesis. G.R. Burgio, A. Astaldi, G. Biscatti, and A. Ugazio, Pavia The effect of different not antiblastic drugs and of a new antiblastic on the lymphocyte PiIA-blastogenesis was investigated. Butazolidin (]3t.), Thiamphenicol (Th.), Kelfizin (K1.)and the new antiblastic platinum (II) diamino dichloride (PDD.) were employed. To 106 lymphocyte PHA-cultures different amounts of the drugs were added. Bt., Th. and K1. in the amounts of 6.25, 12.5, 25 ?g/ml respectively, reduced the blastogenesis to about 40--50% in confront to the control PHAcultures; 2 ~g/ml of the PDD. were sufficient to produce the same effect. Similarly, a reduction of the blastogenesis rate was observed in PHA-cultures of lymphocytes obtained from 25 patients treated with Bt., Th., orK1. (7, 8,10 children respectively). The blastogenesis we~sreduced to 40--60% in confront to that obtained before the beginning of the treatment; the differences resulted statistically significant. A significant reduction of blastogenesis was observed also when Bt., Th. or KI. were added to the mixed lymphocyte cultures (6.25 t~g/ml); the same result was obtained with the addition of 2--3 i~g/ml of PDD.

In Vitro Reactions by Cord Blood Lymphoeytes. A. R. Hayward, London Spontaneous thymidine uptake by cord blood lymphocytes (CBL) differs from that of adult lymphocytes in being greater and reaching a peak value at 3 days. This uptake was abolished by ALS but not by anti-kappa serum. The cells responsible were shown to have the morphology of CBL by autoradiography. These cells will spontaneously lyse chicken red cells (CRC) at a greater rate than adult lymphocytes, sho~dng that the uptake activity has some relation to the efferent limb of immune response. Similar activity is detected in the CBL of infants delivered by caesarian section done before the onset of labour, so transfusion of maternal lymphocytes during labour is not the main cause of this phenomenon. Thymidine uptake by CBL is increased by PHA and MLC using pooled stimulating lymphocytes, though less so with maternal lymphocytes. Lysis of CRC is stimulated by PHA and by coating the target cells with aggregated IgG or immune complexes. I n spite of the high normal thymidine uptake of pro-labour CBL, the relatively low response in MLC with maternal cells suggests that the latter may provide the stimulus for spontaneous activity. Possibly transfer factor is liberated from lysed maternal cells, passively transferring cell mediated immunity. The observation of a sm~ller increase of spontaneous activity as early as 33 weeks of gestation suggests that this may happen for a considerable proportion of pregnancy.

248

Meeting of the International European Society for Pediatric Haematology

Transfer Factor in the Treatment of Candidiasis. W. H. ttitzig, S. Paul, L. E. Spitler, and H. H. Fudenberg, Zfirich and San Francisco Five patients with impaired cellular immune reactions were treated with a transfer factor preparation of known specificity. Their age varied between 13/12 and 176/12 years. The common clinical sign was an extensive candidiasis. The basic disease was unknown in 3 cases ("idiopathic mucocutaneous candidiasis", in one case combined with endocrinopathy). One patient had Wiskott-Aldrich syndrome, and one combined immunodeficiency syndrome. Skin reactivity against candidine was negative in all 5 patients. In vitro reactivity of lymphocytes against the same antigen was, however, positive in 4 of the patients. After application of transfer factor normalization for different periods of time was possible in all cases. Transfer factor seems of particular interest to the clinician, since - - consisting of a small molecule - - it does not elicit antibody formation. One possible source of error, the presence of blocking antibodies, could conceivably interfere in two cases only. In the others a transfer of immune reactivity by the preparation used appears to be certain.

Is Transfer Factor A Therapy in Immune Deficiencies ? C. Criscelli, J. P. Revillard, C. Herzog, J. L. Touraine, and H. Betuel, Paris and Lyon Dialysable Transfer Factor was prepared from a large pool of normal adult donor's leucocytes following Lawrence's technique. One unit is defined as the final product of 6.109 leucocytes. This homogenous preparation was injected into 3 normal recipients and 14 patients with immune deficiencies: 5 ataxia telangieetasia, 4 Wiskott-Aldrich syndromes, 2 humoral predominant and 2 combined immune deficiencies, 1 chronic muco-cutaneous candidiasis. After injection of 1 to 3 units, the transfer of delayed hypersensitivity to several antigenic markers was achieved in 12 cases. Streptokinase-Streptodornase and Candidine gave the highest incidence of positive tests among adult donors and recipients. 3H-thymidine incorporation into recipient's lymphocytes stimulated with the appropriate antigens was increased in some cases following transfer. When performed, leucocyte migration tests did not show conclusive changes. In a few cases PttA and ALS responses of the recipients lymphocytes where also increased, and 4 recipients developped a positive reaction to DNCB, a chemical to which none of the donors were sensitized. Clinical improvement was observed in 2 cases of Wiskott-Aldrich (diminution of infections and eczema). Therapeutic effects were absent or uncertain in the other cases. If confirmed, possible dissociation between skin reactions, blast transformation and clinical improvement would suggest that transfer factor may Mso act by inducing the peripheral manifestation of an already acquired but non expressed cellular DH.

Influence of Transfer Factor (TF) on Immunologic and Clinical State in Chronic Mueocutaneous Candidiasis~ F. Bl~tker and H.H. Hellwege, Hamburg, P. Grob, Zfirich, K. II. Schulz, Hamburg Chronic mucocutaneons eandidiasis is most frequently combined with partial cellular immune defects which result especially in a general cutaneous anergy. This typical lesion was found in 2 of 3 children with chronic mucocutaneons candidiasis. In one

Abstracts of the Third Meeting

249

case with most severe candidiasis, progressive cellular immune defect but intact humoral immunity we made an attempt to reconstitute the cellular immune system by administration of TF. One year ago we started an intermittent therapy with low doses. A reconstitution of cutaneous hypersensitivity against candida albicans could be demonstrated 7--14 days after each injection, but disappeared between 3rd and 4th week after the application of TF. Also the clinical and mycological findings converted only for this short period. Due to these findings we consequently started a therapy with high doses of TF in shorter intervals. In this way we could influence more effectively the course of candidiasis and the state of the cellular immune system.

Eleetronmicroscopical and Immunofluorescent Evidence of Epstein Barr Virus in Sections of a Lymphoblastic Lymphoma. S. Thunold, J.V. Johannessen, A.O. Y[yking, J. N. Wiig, and P. J. Moe, Bergen A Burkitt-like tumour of the small intestine was removed from an 8-year-old Norwegian girl. The abdominal lymph nodes were affected, but the tumor was confined to the abdom~n. The bone marrow and peripheral blood were normal. Histology showed an undifferentiated lymphoblastie lymphoma with distinct starry sky pattern. The tumor cell eytoplasma contained fat vacuoles. Bone marrow smears examined 2 months later were hypocellular with infiltration of blast cells, and the peripheral blood smears contained scattered blast cells. Despite local radiation and intensive cytostatie therapy the patient died 3 months after admission. Autopsy revealed large intestinal and abdominal tumor masses and lymph node, ovarian and bone marrow involvement. Electronmicroscopicat studies of tumour sections obtained from the biopsy specimen revealed herpes-like particles in & small number of turnout cells. In addition, 5--10% of tumour cells in frozen sections or acetone fixed tumour imprints gave positive nuclear immunofluorescent staining when incubated with infectious mononucleosis sera followed by a labelled anti-human immunoglobulinserum. The mononucleosis sera contained anti-viral eapsid antigen (VCA) antibodies and no anti-envelope antigen (EA) antibodies when tested against established Epstein Burr virus-carryinglymphoblastic cell fines. The patient's serum and selected normal human sera which were anti-VCA and anti-EA negative did not react. The results are compatible with the presence of Epstein Barr virus in the tumour tissue.

The Development of Fetal Red Cell Receptors. K. Fischer, A. Poschmann, I. Schedel, and K. H. Winterstein, Hamburg We have previously demonstrated varying degrees of development of A, B, and H receptors of fetal red cells utilizing a special immunofluorescent technique. In contrast to A, B, and let receptors the Rh-f&ctor D is already present in the very young fetus. These findings adequately explain clinical differences between haemolyric disease due to AB0- and to Rh-incompatibility. Further receptors were recently detected in our laboratory utilizing antibody-like substances of plants and animals, for example Attp (Heli[x pomatia) and BHEe (Evonymus europaea). Studies will be presented applying i~eterologous antibodies against red cell membrane enzymes in an attempt to demonstrate the enzyme substrate by immunofluorescenttechnique.

250

Meeting of the International European Society for Pediatric Haematology

Haematological and Cytogenetie Effects Induced in the Newborn by Anticonvulsant Treatment o~ the Mother during Pregnancy. L. Massimo, I~L F. Pantarotto, and M. G. Vianello, Geneva Eighteen infants, born from epileptic mothers treated during pregnancy with anticonvnlsant drugs (15 diphenyl-hydantoin, 3 primidone and barbiturates) were examined for the following features: haematological conditions at birth, chromosomal research, lymphoblastic transformation. Nine of the 15 babies suffered from bone marrow aplasia or hypoplasia; only six showed a lymphoblastic transformation within normal Hmits at birth. Babies born from mothers treated with hydantoin showed a high percentage of aneuploid cells, amitosis, chromosomal pulverization, conglutination, few fragmentations. All the findings were not present in five infants born from epileptic mothers not treated during pregnancy and in the three whose mothers did not take hydantoin. Surface immunoglobulius were normally present. Two of the 15 babies had congenital malformations: one cleft lip and palate and the other one polydaetlly. Ten epileptic children in treatment with diphenyl-hydantoin, also examined, showed a low lymphoblastic transformation index and high aneuploidy. Studies of Maternofetal Red Cell Passage by Means ot Fluorescent Anti-HbAs-Antibodies. Th. Sanguansermsri and K. Betke, ~finchen Haemoglobin As ( = as~2), a small component of human haemoglobin, is found only in minimal amounts in newborn infants. In adults it is evenly distributed among the red cells. By injection of chromatographically pure Hb As into rabbits anti-Hb A~-antibodies can be obtained. Labelled with ftuorescein isothiocyanate they are suitable to mark tIb A2 containing red cells on smears. In smears of adult blood all cells give bright fluorescence, in smears of newborn blood no or only very weak fluorescence is seen. Thus the method is a sensitive means to detect maternal red cells in the blood of a newborn. Positive findings were obtained in 49 out of 152 newborns, mostly less than 1 maternal cell per i000 fetal cells. Two infants showed massive materno-fetal transfusion, amounting to 5 and 26% respectively of the cells.

Free a-Chains in Cord Blood Haemolysates. E. Kohne and E. Kleihauer, Ulm A minor haemoglobin component regularly present in haemolysates of human cord blood and adult blood has been detected, and identified to consist of free a-chain. The isolation was achieved by chromatography on DEAE-Sephadex As0 in TrisHC1-KCN-buffers at pH 8.65. In fullterm newborns the amount of free a-chains is in the range between 0.042 and 0.101% (mean 0.075~o) of the total haemoglobin. This observation is not in agreement with the supposed "physiolocal" deficiency of a-chains in normal newborn infants, which was thought to be indicated by the presence of Hb Bart's (Fa).

The Regulation of Erythropoiesis in the Fetus and Newborn. K. Halvorsen and S. Halvorsen, Oslo Previous studies indicate that the erythropoiesis regulatory system through erythropoietin (ESF) is functioning in the human fetus during the last trimester of pregnancy. Up to recently no data have been available for the first part of pregnancy. Human fetal liver cells in culture responded to ESF with increased heine production.

Abstracts of the Third Meeting

251

Data on the erythropoiesis regulatory system in the newborn period are conflitting. Studies in rats and mice may indicate that ESF plays a minor role. This was tested in rabbits. Newborn rabbits have Ht about 50~o and keep this level up to the 10th day. Then Ht starts to fall. The lowest levels are found at the end of the suckling period (20--25th day). During this period no ESF was detectable in blood, but following hypoxia the levels were increased. Total RCV increased only slightly following 10 days of hypoxia and erythropoiesis did not increase following bleeding. When parenteral iron was given, the rabbits were able to keep Ht-Hb within adult levels, and they increased their total RCV following hypoxia. The erythropoiesis regulatory system in the newborn rabbit is thus functioning as in adult animals.

Erythrocyte glutathione Peroxidase in Infants with Iron Deficient Anemia. R. Braeei, I. Giani, V. Micheli, I~. Berardi, and F. Malandrini, Siena I n an attempt to detect connections between iron deficiency and erythrocyte glutathione peroxidase (GSH-Px), the enzyme activity has been tested on 30 infants with iron deficient anemia 2--7 months old. As a control the same test has been carried out on 40 subjects of various ages. No statistically significant differences between enzyme activities in infants with anemia and control subjects have been found (3.6 q- 1.8 I.U. - - 5.4 • I. U. respectively). On the other hand statistically significant differences can be seen between the values of GSH-Px from low birth weight anemic infants and those from normal subjects (2.7 ~ 1.8 I.U. - - 5.4 ~ 1.2; P < 0.001). The low birth weight infants treated with iron showed considerable increase of erythroeyte GSH-Px (4.03 =]= 1.3 I.U.). Determination of the er~hrocyte age showed slight increase of the young erythrocyte population in the anemic low birth weight infant, after iron therapy.

Congenital Toxoplasmosis and its Repercussions on the Immnnoglobulin-Forming System. D. Frommel, G. Desmont, B. Geny, C. Griscelli, and P. Mozzieonacci, Paris The survey of 27 newborns and babies affected by the severe and generalized form of congenital toxoplasmosis revealed in 11 cases the presence of a serum M-component of the IgC class. These immuaoglobulins of limited heterogeneity appeared to be synthetized by the fetus as they could be detected up to the 75th day postpartum and were absent in maternal serum. No definable antibody activities could be assigned to the M-components. During physiological immaturity of the immune system, the expression of antigenic stimulation is limited by the pool of reactive cells and can lead to proliferation and immunoglobu]Jn synthesis of a limited number of clones. Paralleling in this respect the description of restriction in heterogeneity of the immunoglobulins in some genetically determined or acquired states of immunedeficiency, our observations in congenital toxoplasmosis may reflect a comparable immunological situation.

Neonatal Hepatitis Assoeiated with Australia Antigen. C. Kattamis, D. Demetriou, V. Syriopoulou, and N. Matsaniotis, Athens Very few eases of neonatal hepatitis associated with the presence of Australia antigen have been reported so far. It is the purpose of this report to describe five infants

252

Meeting of the International European Society for Pediatric ttaematology

aged 2 ~ - - 3 ~ months old with the clinical and biochemical findings of neonatal hepatitis in which the Australia antigen was detected in the acute phase. In one patient the disappearance of the antigen was followed by the production of Australia antibody. The mothers of two infants suffered from hepatitis at the 4th and 8th month of gestation; In the third infant both parents had Au (1), but without clinical or laboratory evidence of liver impairment. The fourth infant has been transfused at the 3rd day of life because of haemorrhagic disease of the newborn, while the fifth had at the same age an exchange transfusion. These data exclude the possibility of the intrauterine transmission of serum hepatitis and point to the dangers of transfusions and exchange transfusions in the neonatal period.

Inhibition of Lysis by Calcium Ion in Normal and Fetal Red Cells*. F. Schettini and A. Mautone, Bari Inhibition of hemolysis by Ca-ion in hypotonie solutions of NaC1 was investigated in red cells from normal children, from newborns, from thalassemia major and trait and from iron-deficiency anemia. The Ca-ion were added to red cell suspension in concentration of 0.1 mM and of 0.01 mM/ml. The hemolysis was assayed by an automated procedure (Fragiligraph rood. D-2). Ca-ion markedly inhibited hemolysis in red cells o f normal, of newborns, of thalassemia trait and of iron-deficiency anemia. The reduction of rate of lysis was from 44.1% to 53.1% on 0.1 mM/ml and from 13.1% to 25.3% on 0.01 mM calcium/ ml. In thalassemia major red cells the inhibition of rate of lysis was of 38.5% on 0.1 mM/ml and of 1.6% on 0.01 mM calcinm/ml. In conclusion thalassemic major red cells are less sensitive to the inhibition of rate of lysis by Ca-ion in hypotonic solutions of NaC1 and differ for this property from fetal red cells.

Normal Folio Acid Levels in Plasma and Red Cells, and Folie Acid Requirement during the First Months of Life. J. Ek and E. Magnus, Oslo The folio acid levels in plasma and red blood cells were followed in infants on different diets to study the normal concentrations and the daffy requirement of the vitamin during the first months of life. In newborn infants the concentrations of the vitamin in plasma and red cells were considerably higher than in their mothers or the normal adult population. These high levels were sustained in breast fed infants. In infants fed on commercial available milk formulas, the levels of folic acid decreased considerably during the first months of age. If the folic acid levels found in breast fed infants are considered as optimal, relic acid supplementation to infants fed on milk formulas is necessary. When the vitamin is given as enrichment to the milk formula 50 microgram folic acid per day to a three months old infant is too little. The daily requirement during this age is probably 50% higher. The hemoglobin concentration, the number of red cells etc. were also studied. No differences between the groups were observed. * Supported by a grant from C.N.R., Italia, Contract n. 71.0089204.

Abstracts of the Third Meeting

253

Heinz Body Formation in Erythrocytes of Newborn Infants: The Role of Hemoglobin Content in Red Cell Membranes. W. Tillmann and W. SchrSter, Hamburg The cell membranes (ghosts) of human erythrocytes of newborn infants produce Heinz bodies faster after an exposure to acetylphenylhydrazine than the ghosts of adults. Ghosts prepared from crythrocytes of newborn infants, incubated for 20 firs in their own plasma or in the plasma of adults, form Heinz bodies faster than the ghosts of incubated red cells of adults. A direct relation can be seen between the hemoglobin content of the ghosts and the formation of Heinz bodies. We suggest that the ability of the membranes of erythrocytes of newborn infants to retaia more hemoglobin durhlg the ghost preparation both before and after the in vitro incubations, explains the enhanced susceptibility of the ghosts to form Heinz bodies and thus also of the intact cells of newborn infants.

Red Cell Values at Birth as a Function oI Gestational Age. R. Zaizov and Y. Matoth, Petah-Tikva, Israel The red cell count, Hb concentration, hematoerit and reticulocyte count were obtained on the first day of life in 72 prematurely born hffants of known gestational age and 19 infants born at term. From week 23 to term the hemoglobin concentration remained constant, around 19.5 g%, while the red cell count increased steadily from 4.63, S.D. • 0.44 to 5.15, S.D. 4- 0.7. The MCV and MCH showed a steady decline, while the MCHC did not change. The reticulocyte count decreased steadily from 8.3 to 3.2. The calculated total number of circulating red cells increased by a factor of 3.7 during the last 13 weeks of gestation, while the corresponding factor for the term hemoglobin mass was 3.3. The last trimester of pregnancy is therefore characterized by rapid, growth of the red cell mass and by gradual replacement of fetal macrocytes by smaller cells. The constancy of the hemoglobin concentration and the actual decline in the reticulocyte count are not in keeping with the widely held view that during this stage of gestation the fetus develops polycythemia, presumably in response to a liypoxic environment.

Neonatal Copper Deficiency Associated with Sideroblastic Anemia, Neutropenia, Bone Changes, and Retarded Development. A. Ashkenazi, S. Levin, D. Benvenisti, and M. Djaldeti, Rehovot A 2 months old premie (B. Wt., 1140 g) was seen because of anemia, neutropenia, hepatosplenomegaly and slight motor retardation. Hb was 7.2 g%, retie. 1%, leuc. 6000/ram 3 with 3% neut. Serum iron was 68 rag%, and I.B.C. 306 rag%. Bone marrow was cellular with mild hypoplasia of red and white cells, and with numerous ring sideroblasts, and marked vaeuolisation of erythroid and myeloid cells. Ferrokinetics showed non-v.tilisation of iron. Metaphyseal bone changes were similar to those seen in scurvy and rickets. There was no hematological or clinical response to accepted therapies and the infant required numerous blood transfusions. Copper deficiency was suspected, and serum copper and ceruloplasmin levels were found to be very low. 2 mg copper/day was given, and the blood abnormalities reverted rapidly to normal. Retie. rose to 10.5% in 5 days, neutrophils returned to normal within 18 days, and the marrow abnormalities disappeared. Ferrokinetics returned to normal.

254

1VIeetingof the International European Society for Pediatric Haematology

This case shows the importance of copper for the proper utilisation of iron by the erythrocytes, for normal maturation of neutrophils and probably for normal motor development in infants.

Phytohemagglutinin (PHA) Skin Test in Children. A. I. Berkel, Ankara A total of 286 children from birth to 15 years were checked for the in vitro response to PHA. 10 ~g of the antigen in a 0.1 ml volume was given intradermally, erythema and induration was measured at 24 and 48 hrs. Induration more than 5 mm at the end of 48 hrs was considered to be a positive response. From 90 newborn, only 8 (8.8%) had a positive reaction at 48 hrs. In 75 patients from 1 month to 1 year of age, 31 (41.3%) had a positive response. In 121 children between 2 and 15 years the test was positive in 78 (64.4%). Induration less than 5 mm at 24 hrs was 37.7% and 5.2% in the newborn and the children over one month respectively. 58.7% of the newborns showed significant induration at 24 hrs while this figure was 80.3% in children older than 1 month. Two patients with Bruton's disease and sporadic agammaglobulinemia had positive responses. Tumors and immtmosuppressive therapy seem to be the most common causes for the negative response in children over 1 year.

Immunological Studies in Children with Xeroderma Pigmentosum. A . I . Berkel, 0. Kiran, and V. Sezer, Ankara Eleven children (9 males and 2 females) from 7 families, ages ranging from 2 to t2 years and four of the parents were investigated. 5 families had more than one affected children. IgG, IgM and IgA levels were normal in all. Delayed hypersensitivity was detected by skin testing with DNCB, Candida, SKSD and P.P.D. antigens. Skin tests were negative in 50--90% of the patients in comparison to age matched healthy controls. Homograft survival was slightly prolonged (15 and 17 days) in two patients. I n vitro response of the lymphocytes to phytohemagglutinln (PHA) was detected by measuring the incorporation of tritiated thymidine into DNA and was expressed as CPM per 10~ cells. Comparison is made with the lymphocyte response of normal healthy subjects. 4 of 11 patients and 2 of 4 parents had a low response to stimulation with PHA. Mixed leucocyte response was poor in 3 patients studied. The plasma of the patients with low P1LIA response inhibited the normal in vitro P H A response of healthy subjects ranging from 13.5 to 26.8%.

Influence of Exogenous Human Leukemic DNA on Recipient Cells. U. C. Wulff, L. S. Desai, J. L. Cohen, and G. E. Foley, Boston and Liibeck Human cells in vitro incorporate intact labeled DNA derived from human leukemic cells in suspension culture. Radiographic studies indicate that the exogenous DNA migrates to the nucleus of the recipient cell. Leukemic DNA induced a specific dose-dependent inhibition of nucleic acid synthesis by recipient cells which decreased as incubation time increased, with concomitant stimulation of protein synthesis - - an effect similar to that induced with certain viruses, as described by others. In general, leukemic cells were more sensitive to such inhibition, due perhaps to structural similarities between the endogenous and exogenous DNA.

Abstracts of the Third Meeting

255

Nolunal human fibroblasts (WI-38) exhibited phenotypic changes following incorporation of leukemic DNA which persisted in serial sub-culture. Such cells lost their fibroblast-like morphology, nueleoli, and no longer attached to glass surfaces. Search for evidence of accompanying genotypic changes is in progress.

Hemostasis and Fibrinolysis at Different Temperatures. A. H. Sutor, Freiburg We studied the hdiuence of temperature on plasmatic, thrombocytic, and vascular components of hemostasis. Plasma factors are most effective between 38 and 40~C; at these temperatures platelet functions are impaired, they are best at room temperature. Vascular hemostasis increases with cold. The overall picture of hemostasis was studied with o u r new qnantitative bleeding time test. There is no difference between body and room temperature, however, below 20~ both, blood loss and bleeding time increase. At temperatures below i i ~ hemostasis does not take place. At 16~ only an intact thrombocytic and plasmatic system brings about hemostasis. This fact enables us to diagnose bleeding disorders, even of mild degree, with our ,,Cold-Tolerance Test", using minimal amounts of blood. Fibrinolysis is most active at febrile temperatures and significantly reduced at room temperature, probably due to impaired plasmin action on fibrin, whereas the plasmin generation seems to be almost unaffected by cold.

@oagulation Defects in Purulent Meningitis. D. Zoumboulakis, C. Kattamis, T. Mandalaki, D. Philippatou, 0. Panayotopoulou, and N. Matsaniotis, Athens Studies on coagulation were performed on 21 patients (aged 4 months to 13 years) with purulent meningitis due to meningococeus, Staphylococcus or Pnemnocoecus. To this end platelets' count, fibrinogen, factor XIII, plasma fibrinolytie activity, and fibrin degradation products were performed in 14 patients. Of 21 patients 7 (33%) presented coagulation defects compatible with that observed in diffuse intravaseular coagulation syndrome. Namely a reduction of factor X I I (Bellow 35%) of fibrinogcn (< 200 mg~ and an increase of fibrin degradation products as determined by the inhibition haemagglutination immunoassay of Merskey (titer above 1 : 16) were noted.

Globin Chain Synthesis in Families with a Thalassemia. Y. Matoth and R. Zaizov, Petah-Tikva, Israel The incorporation of C14 leucine into globin chains in vitro was examined in children of Yemenite and Iraqi origin known to have had Hb ]~art's at birth, in their families and in families of patients with Hb H disease. All children who had Hb Bart's at birth showed a low a/fi chain ratio. The degree of severity of the defect in a chain synthesis paralleled the concentration of li b Bert's at birth. At least one parent and some of the siblings of children with Hb Bert's at birth showed a low a/fl ratio. A defect of a chain synthesis was also found in all the parents of patients with Hb I~I disease. In some of these families the defect was equally mild or severe in both parents, while in others the degree of severity differed between parents. A normal distribution of values for a/fi ratio was found in a population of some 50 heterozygotes for a-thalasscmia, composed of children with Hb Bert's at birth,

256

Meeting of the Laternational European Society for Pediatric Haematology

parents and siblings of these children and of patients with Hb tt disease. Our observations in Yemenite and Iraqi Jews are consistent with the transmission of a-thalassemia in these populations through one gene with variable expressivity. Hb ~ is the most severe expression of a-thalassemia observed in these populations and may well represent the homozygous state for this gene.

Further Studies on the Relationship between G-6-PD and Erythrocytc Acid Phosphatase Polymorphism in the Sardinian Population. R. Agostino, G. Antegnoni, E. Bottini, L. Bnsinco, M. Di Mino, P. Lucarelli, and R. Palmarino, Rome We have priviously reported a positive association between favism and pa and pc alleles for acid phosphatase. 50 Sardinian males with a positive history of favism and 1640 school boys and girls with an age between 7 and i4 years from 17 Sardinian villages were examined in the present investigation. The data show that in G-6-PD deficient subjects who had had hemolytic favism, thalassemia trait is negatively associated with pa allele for acid phosphatase. This new observation suggests a strong interference of thalassemia on the interaction between G-6-PD and acid phosphatase polymorphisms. The present data also show that in the villages studied the frequency of pc allele correlates positively with the altitude and negatively with the past malarial morbidity and GdMed prevalence. No similar relationship was observed for pa and pb alleles. This pattern of gene frequencies distribution is in agreement with the observations in subjects with favism suggesting a relatively simple relationship only between GdMed prevalence and pc, while pa frequency should be a function of both GdMed and thalassemia prevalence.

Variability o3 Tumour Specific Membrane Antigens in the Course o~ Childhood Leukacmia. F. Zintl, G. Aurich, and W. Plenert, Jena The existence of tumour specific antigens on the surface membranes of human leukaemic cells could be demonstrated already two years ago by the same authors. In the meantime these studies could be extended and substantiated so far as to give some possibilities to formulate conclusions concerning further research and immunotherapy. Using the technique of indirect immunofluoresccnce in the autologous system it was possible to demonstrate turnout specific membrane antigens on the parablasts of childhood leukaemia. These antigens could not be found in the initial phase but only in the second relapse and later on. There was no cross-reactivity with cells from Burkitt-Lymphoma. Changes of lymphocyte transformation (in vitro) and macrophage-migration-inhibitiontest in the course of childhood leukaemia corresponded to the results of the mentioned immunofluorescence studies.