Besides changes in behaviour and lifestyle we nowadays have the choice of specific drugs in the treatment of reflux oesophagitis. A distinction can be made in motility modulating drugs, which stimulate oesophageal peristalsis and LOS pressure, mucosa-protecting drugs, which form a protective layer on the oesophageal mucosa, acid neutralizing (antacids) and acid suppressing drugs (H2-receptor antagonists, omeprazole). So far the results of medical therapy of reflux oesophagitis are still suboptimal. Giving the H2-receptor antagonists with the evening meal would possibly be more appropriate. A valid alternative is the mucosa-protecting agent sucralfate. Monotherapy will probably be insufficient for full healing, which explains why trials of combination therapy (H2-receptor antagonists plus sucralfate or plus cisapride) are being conducted. If omeprazole becomes available, it will revolutionize the therapy of severe reflux oesophagitis. Many questions (dose, duration, maintenance, safety monitoring etc.) remain to be determined.

The number of options now open to surgeons in the treatment of carcinoma of the oesophagus is considerable. One, two or three different approaches can be used to remove tumours at any level between the hypopharynx and the cardia. The Sweet procedure involves a left thoracotomy followed by anastomosis. The Lewis Tanner operation begins with the stomach being mobilized through an abdominal approach followed by resection and anastomosis by a right thoracotomy. A triple approach--cervical, thoracic and abdominal--is selected when anastomoses are extrathoracic. In McKeown's operation, the whole stomach is used and the posterior mediastinal route selected. Akiyama tubulizes the stomach and has chosen the retrosternal route. Orringer has recently developed oesophagectomy without thoracotomy. When tumour removal is impossible or there is a local or general reason for refusing excision, the surgeon can turn to palliative surgery to give the patient the means of enjoying a normal life during the time that is left to him. The whole stomach can be used or it can be made into a tube by resecting the lesser curvature. Postlethwait made use of Lortat-Jacob's technique. Reversing the stomach has also been suggested. Colonic oesophagoplasty is possible if previous gastrectomy has been carried out. The surgical management of malignant oesophagotracheal fistulae can be limited to bipolar exclusion of the oesophagus. Ideally, a retrosternal gastric plasty should also be performed with drainage of the lower oesophagus into a Roux-en-Y loop. The choice of treatment is made on the basis of the preoperative assessment of the patient. The extent of disease spread is evaluated. The most important diagnostic tools are fibreoptic bronchoscopy and ultrasound. The type of surgery selected is contingent on the tumour site and the patient's physical condition. Oesophagectomy without thoracotomy has meant that surgery is available to patients for whom thoracotomy would have been inadvisable. Malignant oesophagotracheal fistulation must be treated by the Kirschner operation. Palliative bypass is carried out only in patients under 50 years of age. It is our opinion that surgery is too often overlooked in the treatment of oesophageal carcinoma. The survival rate at 5 years is 23% for potentially curable resection and the operative mortality 2.6%. In other cases, palliative resection (or bypass for patients under 50) allows the patient to feed himself and lead a normal life until the inevitable fatal outcome.

A variety of pathological abnormalities of the smooth muscle and myenteric plexus result in clinical syndromes of disordered small intestinal and colonic motility. These pathological abnormalities have been noted by conventional light microscopy and by utilization of Smith's technique for visualizing the myenteric plexus with silver. We have classified the neuromuscular disorders into two major categories, i.e., those affecting the myenteric plexus and those affecting the smooth muscle. The classification is further developed based on the variety of clinicopathological features of the various disorders. Although we can now identify the underlying pathology of these motor disorders and thus understand these illnesses better than we did a decade ago, we have much more to learn. With the great strides being made to understand the normal structure, function, and development of the myenteric plexus and smooth muscle, there is hope that we will be able to learn much more about the etiology and pathogenesis of these neuromuscular disorders in the decade to come.

Although the presence of gastric bacteria has been long established, the recognition and isolation of Campylobacter pylori and similar organisms has opened a new era in the understanding of inflammatory gastroduodenal conditions. Visualization or isolation of gastric Campylobacter-like organisms (GCLOs) is significantly associated with histologic evidence of gastritis, especially of the antrum. Correlation with peptic ulceration also exists but probably is due to concurrent antral gastritis. Outbreaks of hypochlorhydria with concomitant gastritis have been attributed to GCLO infection, and a human volunteer became ill after ingesting C. pylori. Despite rapid microbiologic characterization of the organisms and the epidemiology, pathology, and serology of infection, the pathogenetic significance of GCLOs remains unknown. Whether GCLOs cause, colonize, or worsen gastritis must be considered an unanswered question at present. The efficacy of antimicrobial treatment of GCLO infection on the natural history of gastritis is not presently resolved. Nevertheless, GCLOs are at the least an important marker of inflammatory gastroduodenal disease, and attempts to ascertain their clinical significance are clearly warranted.

(1) The serological diagnosis of PBC is possible in almost 100% of cases when appropriate methods and specific antigen preparations are used such as the purified ATPase fraction by ELISA for the detection of anti-M2, sonicated mitochondria by immunodiffusion for the demonstration of precipitating antibodies against M-A or M-B, and cell cultures by immunofluorescence for the detection of antibodies against nuclear dots. (2) The establishment of AMA profiles obtained by ELISA and CFT seems to be a sensitive approach to a better definition of the natural course of PBC. A distinction between a rather benign and a more progressive course seems especially possible in the presence of the AMA profiles A and B (anti-M9 and/or anti-M2-positive only by ELISA) versus D (anti-M2-, anti-M4-, anti-M8-positive in the CFT). (3) The analysis of cellular immune reactions in vitro and in vivo suggests an activation of cytotoxic T cells as well as a defect in the function of T suppressor cells. (4) Although the aetiology of PBC is unknown, the detection of MHC Class II antigens on bile duct epithelial cells in liver biopsies of patients with PBC but not of normal individuals may imply that an infectious agent being exposed in association with these MHC structures may trigger the disease. The inability of the immune system in controlling this infectious process would then lead to an ongoing inflammatory reaction which is responsible for the continuous destruction of bile ducts within portal triads.

Experimental work showing that IgA plasma cell precursors activated in gut associated lymphoid tissue (GALT) of rats and sheep migrate to the lamina propria of the gut via the regional lymphatics, mesenteric lymph node and blood, has been supported by immunohistochemical studies. In rats, immunoblasts with cytoplasmic IgA are present in the Peyer's patches in association with the high endothelial venules which is probably an important, though not the only, site of extravasation into the gut, whereas cells with cytoplasmic IgA are rarely observed in the dome regions of Peyer's patches. Immunohistochemical studies of human Peyer's patches have revealed differences between the distribution of cells with cytoplasmic IgA in man compared to rats. In man, immunoblasts with cytoplasmic IgA are not concentrated in the zone of cells containing the high endothelial venules, whereas they are present in the dome regions of the Peyer's patches. The following questions arise: Do precursors of IgA plasma cells activated in human GALT migrate to the lamina propria via the blood, but extravasate predominantly via the capillary network, rather than the high endothelial venules? or do IgA plasma cell precursors 'mature' in situ in the Peyer's patches of man and subsequently migrate laterally to seed the lamina propria? Three lines of evidence from studies of primary B cell lymphomas of GALT support the latter hypothesis: 1) Primary B cell lymphomas of the gut remain localized to GALT for long periods of time; 2) Histological studies of the lymphoid tissue in these lymphomas have shown a gradation of cell types, from the muscularis mucosae towards the mucosal epithelium, which strongly suggests that plasma cells develop in situ in the gut from the adjacent layers of cells; 3) Preliminary studies of DNA extracted from the blood-borne cells from patients with GALT-derived B cell lymphoma have failed to demonstrate the presence of clonal gene rearrangements. Normal and malignant human GALT contains a perifollicular population of B cells with centrocyte-like morphology which lack surface IgD. No direct equivalent can be detected in rodent Peyer's patches. Their quiescent nature and distribution in malignant GALT suggests that they are follicle centre cell-derived and precursors of immunoblasts and plasma cells. As such they may be memory B cells. Their association with epithelium is a consistent feature of normal and malignant GALT which is of unknown but undoubted significance. The function of intraepithelial T cells is still unknown. Malignant T cells in MHI may be derived from intraepithelial T cells.(ABSTRACT TRUNCATED AT 400 WORDS)

Coeliac disease is defined as that disorder in which there is an abnormality of the small intestinal mucosa manifested by contact with the gluten of wheat and certain other cereal grains. In the immunological theory of the pathogenesis of coeliac disease, gluten, or a component, is viewed as the antigen responsible for the immune response. The search for the gluten component responsible for 'toxicity' and, by implication, antigenicity, is described. The antigen may be presented differently to the immune system by an abnormal cell membrane, either of the enterocyte, lymphocyte or macrophage. Alternatively, increased amounts of antigen may be absorbed due to increased membrane binding or permeability, either of which could be genetically determined. As a further possibility, coeliac disease may occur because the patients are immunologically hyperresponsive and this too appears to be genetically determined. The perturbations which occur in the mucosal immune system and the systemic immune system are described. It is conceivable that the major complications described (intestinal ulceration, malignancy and splenic atrophy) result from immunological disturbances. The incidence of childhood coeliac disease is declining, which may be due to altered exposure to, or increased protection from, the antigen in infancy, or to changes in environmental factors. The immunological mystery of coeliac disease continues to excite interest and fascination, and has certainly been a stimulus to our deeper understanding of gastrointestinal immunology.

Pernicious anaemia (PA) and chronic atrophic gastritis (CAG) aggregate in families, occur more often in women, and are associated with various heritable traits such as fair skin and blue eyes. They are linked to certain HLA types. Linkages are relatively weak for A and B antigens, but somewhat stronger in the case of DR antigens. There are strong associations between PA and other organ-specific autoimmune diseases, particularly those affecting the thyroid. Discordance for PA in monozygotic twins has been reported, and it may well be that expression of the disease requires, in a genetically susceptible individual, initial injury to the gastric mucosa by some environmental agent such as a virus or some physical irritant, with perpetuation of injury then depending upon autoimmune mechanisms. Numbers of T cells are substantially increased in the gastric mucosa of patients with PA, but the ratio of T suppressor to T helper cells is normal. There is a relatively greater increase in numbers of cells not of T lineage, presumably B-cells. Gastric autoantibodies, both to different components of the parietal cell and to two sites on the IF molecule, are present in a majority of patients with PA. There is evidence that these autoantibodies, especially PCA, may be cytotoxic to parietal cells, and may also inhibit their maturation and proliferation. Antibodies to chief cells have not been described, and the parallel disappearance of these cells in atrophic gastritis is unexplained. The peripheral blood lymphocytes of some patients with autoimmune gastritis transform, or produce lymphokines, when exposed to gastric antigens, and patients with PA have been shown to have delayed type cutaneous hypersensitivity to gastric antigens. The relevance of these observations to the pathogenesis of their gastric mucosal lesion is unclear. There is a growing body of evidence to support the operation of humoral immune mechanisms in autoimmune gastritis, but this clearly does not preclude the coexistent involvement of cellular mechanisms. For example, impaired suppressor T cell function has been strongly implicated in certain other autoimmune disorders, but has received scant attention in PA. By generally accepted criteria, PA is an excellent example of an organ-specific autoimmune disease. As yet, there is no acceptable single unifying hypothesis which will account for all of the phenomena which have been described in the disease.(ABSTRACT TRUNCATED AT 400 WORDS)

In summary, there has been a dramatic increase in our understanding of food allergy as a result of research in immune mechanisms and clinical studies over the last decade. The subject has been comprehensively reviewed in a major new publication (Brostoff and Challacombe, 1986).