The fact that nonsteroidal antiinflammatory drugs (NSAIDs) damage the gastroduodenal mucosa is now accepted as NSAID use has been associated with a disproportionately high frequency of upper gastrointestinal bleeding and perforation of gastric and duodenal ulcers. More than 10% of patients receiving NSAIDs chronically will have a gastric ulcer on any given day, a point prevalence of ulcer disease 5-10 times higher than in patients who are not taking NSAIDs. Endoscopic studies comparing the effect of acute administration of NSAIDs on the gastroduodenal mucosa in normal volunteers have failed to predict which NSAIDs would be safest when administered chronically. All of the newer NSAIDs appear to be similar in their propensity to cause chronic mucosal damage, including peptic ulceration. Recent studies have suggested that in those starting NSAID therapy, prophylactic cotreatment with H2-receptor antagonists or sucralfate has minimal or no effect on preventing the development of NSAID-induced gastric ulcers, although duodenal ulcers may be reduced. Nonsteroidal antiinflammatory drug-induced gastric ulcers are also not prevented by drug formulations that prevent or markedly reduce the amount of active NSAID in the stomach. Cotreatment with the synthetic prostaglandin misoprostol was associated with a marked reduction in gastric ulcer development in patients with osteoarthritis receiving NSAIDs chronically, suggesting that prevention of prostaglandin generation in the gastric mucosa may play a pivotal role in NSAID-induced gastric ulcers.

Endoscopic distinction between ulcers and erosions is difficult. Consequently, existing literature, which must be taken at face value, may be misleading. Nevertheless, from published studies most gastric and duodenal ulcers associated with nonsteroidal antiinflammatory drugs appear to heal on antacids or H2-antagonists. Sucralfate appears useful for duodenal but not gastric ulcers. Continuing nonsteroidal antiinflammatory drugs does not prevent or delay healing of duodenal or small gastric ulcers; their effects on large gastric ulcers remain uncertain. Thus far, only full doses of H2-antagonists, or their combinations with antacids, have been shown to heal ulcers and prevent recurrences. Ulcer recurrences and complications have occurred in small numbers of patients on maintenance doses of H2-antagonists. Available antiulcer drugs (antacids, H2-antagonists, sucralfate) reduce severe acute injury when taken before or with nonsteroidal antiinflammatory drugs. They also reduce ulcerlike symptoms due to nonsteroidal antiinflammatory drugs. Inexplicably, chronic prophylaxis with H2-antagonists for 4 wk or more appears ineffective in preventing gastric ulcers, although duodenal injury is reduced. As the efficacy of available prophylactic therapy (H2-antagonists, sucralfate, and antacids) has not been established, routine use in all cases seems unjustified at present.

Adverse effects associated with nonsteroidal antiinflammatory drug treatment are reported more commonly to regulatory authorities than the adverse effects of any other form of treatment. Epidemiologic evidence in general suggests a doubling or quadrupling of the risk of ulcer complications or death in recipients of such treatment. The risk appears to be related to increasing age, but no other associated factor has yet been identified. It is uncertain whether individuals with preexisting ulcer are at special risk or whether treatment predisposed equally to ulcer and ulcer complications, and there is no clear evidence that the nature of the disability leading to nonsteroidal antiinflammatory drug treatment influences the risk of gastrointestinal complications.

Review of the four studies investigating the effect of initial therapy on ulcer relapse indicates that there is an increased rate of relapse after treatment with H2-receptor antagonists compared with placebo or other classes of drugs when combined. When individual drugs are considered, this appears to hold true for colloidal bismuth alone. There are several possible reasons for these differences. The most likely candidate mechanism to account for these differences is an alteration in gastric secretion, which might arise from alterations in gastrin, altered receptor regulation, parietal cell sensitivity, or peptic activity. However, the evidence for such an alteration in secretion is far from conclusive. Other factors that may be important include mucosal defense factors, smoking, Campylobacter pylori, and the statistical methodology.

Peptic ulcer hemorrhage or perforation occur commonly in patients with hitherto silent ulceration, particularly in the elderly or in those taking nonsteroidal antiinflammatory drugs. The majority of patients dying from peptic ulceration have no symptoms of ulcer disease until the presentation of their final, fatal illness. There is a need for more studies investigating the early symptoms of the fatal ulcer disease, the effects of therapies that may prevent fatalities, and the management of potentially lethal ulcer disease upon clinical presentation. Silent ulceration also occurs in patients after successful healing with medical treatment, and the apparent point prevalence of silent ulceration (in the framework of a clinical trial) is critically dependent on the frequency of repeat endoscopic examination and the treatment modality. Maintenance treatment with H2-receptor antagonists appears to some extent to decrease the frequency of asymptomatic ulcers, but more interestingly to halt the progression of asymptomatic ulcers toward symptomatic or complicated ulceration, although more careful studies in comparison with other modalities are needed.

Campylobacter pylori is now known to be the most common and important cause of gastritis, and C. pylori infections have been associated with duodenal ulcer, gastric ulcer, nonulcer dyspepsia, and gastric cancer. Although it has been only possible to culture C. pylori for about 5 yr, there are already sufficient data available to allow us to develop the basic framework that relates C. pylori gastritis to the causation of peptic ulcer disease. We review the data and propose mechanisms that implicate C. pylori as an important factor in the pathogenesis of peptic ulcer disease and consider the therapeutic implications. What we now know about C. pylori begins to unravel some of the mysteries surrounding peptic ulcer disease.

Active immunization of rabbits with the principal, endogenous prostaglandins in the gastrointestinal mucosa induces gastrointestinal mucosal ulceration. Development of ulceration in prostaglandin-immunized rabbits appears to be a direct consequence of production of specific prostaglandin antibodies, as prostaglandin antibodies per se induce gastric ulceration within 9 days when administered intravenously to unimmunized rabbits. These studies suggest that endogenous prostaglandin E2, F2 alpha, D2, and I2 in the gastrointestinal tract play an important role in preventing mucosal ulceration. The mechanism of ulcer formation is not completely understood, but most evidence points toward prostaglandin antibodies inducing mucosal ulceration by binding to endogenous prostaglandins within the mucosa and thereby negating their mucosal protective effects. Gastric acid hypersecretion and complement fixation by prostaglandin-antiprostaglandin complexes are not likely involved in the development of mucosal ulceration in this model. Use of antibodies to interfere with prostaglandin action may be an alternative approach to investigate (a) the importance of endogenous prostaglandins in mediating mucosal protective mechanisms and (b) the role of prostaglandins in acute and chronic erosive/ulcerative diseases of the gastrointestinal tract.

Peptic activity has long been recognized as an essential factor in the pathogenesis of peptic ulcer and related diseases, but only recently has it become clear that this activity is derived from a remarkable diversity of enzymes, all of which belong to the aspartic proteinase family of enzymes. These include two types of pepsinogens and two types of cathepsins. In recent years, considerable progress has been made in characterizing these proteinases and in applying this information to the study of a number of gastrointestinal disorders. The intent of this article is to update recent basic and clinical information on these topics and to suggest several areas that merit further investigation.

A proposed framework for integrated research in ulcer epidemiology is presented. This framework entails the integration of ulcer epidemiology information from national data sets and regional or specialized data sources to produce an organized sequence of studies. National data sets are used to determine the descriptive characteristics of ulcer epidemiology in terms of person, place, and time. Once subpopulations with high or low rates of ulcer disease are identified, hypotheses are generated to explain subpopulation differences. These etiologic hypotheses are then tested in regional or specialized study populations such as the Adventist Health Study, the Rand Health Insurance Study, or the American Rheumatism Association Medical Information System.

This overview is intended to provide perspectives drawn from the proceedings of the Ulcer-Epidemiology Symposium reported in this supplement. This meeting was designed to review the epidemiology and risk factors underlying the development of acid/peptic diseases and to highlight the natural history and therapy of these disorders. These topics served as a focus for the meeting. A second goal was to consider the topics in the context of ulcer complications resulting from use of nonsteroidal antiinflammatory drugs. Although nonsteroidal antiinflammatory drugs are well known to cause gastric damage and ulcers and to lead to complications of preexisting peptic ulcers, the frequency and severity of the clinical problems resulting from their use and the efficacy of different modalities for prevention and treatment have only recently begun to be the subjects of careful analysis.

This review of the mechanisms by which aspirin causes gastric mucosal damage points to the involvement of two potential mechanisms. Aspirin, which inhibits cyclooxygenase, is rapidly deacetylated to salicylate. Salicylate is toxic to cells and affects mucosal barrier function, reduces cytosolic adenosine triphosphate, stimulates sodium transport, and increases proton dissipation from surface epithelial cells. Cyclooxygenase inhibition makes the gastric mucosa more susceptible to injury, inhibits mucus and bicarbonate secretion, alters the physicochemical nature of mucus, stimulates fundic but not antral [3H]thymidine incorporation, and reduces epithelial surface hydrophobicity. No single mechanism seems to be involved. It is likely, instead, that the toxic effects of salicylate and the effect of cyclooxygenase inhibition work in concert to render the mucosa more susceptible to injury, resulting in mucosal damage.

With advances in our understanding of the pathophysiology of alcoholic liver disease, pharmacological treatments of some of the basic disease processes are now in sight. The most notable development has been the introduction of propylthiouracil for the treatment of alcoholic hepatitis. In a recent trial the mortality rate of patients treated with this drug was 62% lower than that of a control group. Its beneficial effects may stem not from its anti-thyroid properties but rather from other actions such as free radical scavenging. Corticosteroids now appear to have no place in the treatment of alcoholic liver disease. Anabolic steroids, however, show promise, though longer term trials are required before this can be confirmed. Colchicine, too, has been reported to improve survival in patients with established cirrhosis. More experience is required with this and other anti-inflammatory and anti-fibrogenic drugs. beta Adrenergic blocking drugs, such as propranolol, reduce portal venous pressure. In a trial among patients with alcoholic cirrhosis who had oesophageal varices, 39% of those receiving propranolol had not experienced a haemorrhage by 2 years compared with 74% in the control group. The mortality rates at this time were 28% and 49% respectively. Results of treatment once the first haemorrhage has occurred are less impressive. Treatment of the alcohol withdrawal syndrome in patients with liver disease is often problematic. The dose of any sedative should be reduced to 25-50% of the usual dose and sedatives should be avoided in patients who are encephalopathic. Once the patient has recovered from the acute illness, abstinence from alcohol remains the single most important factor that determines long term survival.

The 1-year survival rates of around 70% that are now being achieved have resulted in the acceptance of liver transplantation as a treatment for end-stage liver disease. The number of patients undergoing transplantation is increasing rapidly and the indications are widening. More patients are being transplanted for acute liver failure following the recent encouraging reports of successful grafting in this condition. The proportion of patients transplanted for liver cancer is falling as it becomes apparent that 80% of patients will die from recurrent disease. The selection of candidates and timing of transplantation continue to pose difficult clinical problems. Although the surgical and anaesthetic aspects of liver transplantation have been greatly improved, the 30-day mortality remains high at around 30% and postoperative complications, especially infection and rejection, continue to be major problems. However, rehabilitation is excellent for most patients and liver transplantation should no longer be considered an experimental procedure.

Autoimmune CAH is important to recognize, since it is highly responsive to treatment which undoubtedly prolongs life. Autoimmune CAH can rarely be cured; complete withdrawal of treatment leads to relapse in over 80% of patients. Prednisolone and azathioprine are the major drugs of choice, the former inducing remission while the latter maintains remission, either alone or in combination with prednisolone. Since both drugs are associated with substantial side-effects which tend to be dose-related, the object of treatment must be to induce and maintain remission with the minimum risk of relapse together with an acceptably low incidence of complications. Although PBC shares many features in common with autoimmune CAH, treatment of the underlying disease is generally unsuccessful. To date no drug has been shown to induce remission or to prolong survival. The main aim of treatment should be directed towards the complications of PBC, of which pruritus and osteoporosis are the two major complaints. Cholestyramine and antihistamines are the drugs of choice, but when these fail a variety of other therapies are also available, although many have only been shown to be effective on an anecdotal basis. No treatment has yet been shown to reverse the bone demineralization which occurs in PBC, but early calcium supplementation is recommended in this disorder. Osteomalacia is uncommon and can be prevented by prophylactic calcium and vitamin D supplementation in jaundiced patients. Liver transplantation is effective in treating PBC, and when successful leads to complete restoration of health with the prospects of increasingly long survival. Recurrence of PBC does not appear to be a significant problem.

Bacterial infection is a serious and often fatal complication of patients with liver disease and can prove fatal either directly or by precipitation of gastrointestinal bleeding, renal failure, or hepatic encephalopathy. At greatest risk are patients with alcoholic cirrhosis or decompensated chronic liver disease, or cases of acute liver disease who progress to fulminant hepatic failure or subacute hepatic necrosis. Infection appears to be unusual in patients with primary biliary cirrhosis. The site and type of infection is unrelated to the aetiology of the liver disease. Bacteraemia, pneumonia, urinary tract infection and spontaneous bacterial peritonitis are most common but infective endocarditis and meningitis, especially with pneumococci, are easily overlooked. Clinical suspicion of infection must be high as the only indication may be a general deterioration in the patients' clinical state, increasing encephalopathy or renal impairment. In the case of patients with fulminant hepatic failure, infection may precipitate the initial or recurrent encephalopathy and contributes to death in 10% of fatal cases. Spontaneous bacterial peritonitis is now recognized to occur in the absence of clinical features of peritonitis. The PMN content of the ascitic fluid may provide the only indication of infection and is the most readily available screening test. The most common types of organism responsible for all types of infection are Gram-negative enteric and streptococci, especially pneumococci, while infection with anaerobes is rare. Risk factors for infection include decompensated alcoholic liver disease, fulminant hepatic failure, gastrointestinal bleeding, invasive practical procedures and impaired host defence mechanisms against infection. Of the host defence mechanisms, impaired function of the reticuloendothelial system, complement, and PMNs represent the most common and serious defects. Defects of humoral immunity are present in ascitic fluid from patients with cirrhosis and are probably a major reason for development of spontaneous bacterial peritonitis. Diuresis improves these functions and reduces the risk of peritonitis. Treatment of infections even with the appropriate antibiotic is still associated with a high mortality but the use of adjuvant gut sterilization is promising, particularly in cases infected with Gram-negative enteric organisms. Infusions of fresh frozen plasma, blood and cryoprecipitate improve some systemic host defences and may be beneficial in the treatment and reduction of risk of infection.

Ascites is a frequent complication in patients with liver cirrhosis. The accumulation of fluid in the abdominal cavity is associated with disturbances of systemic and splanchnic haemodynamics and of kidney function, which contribute to the poor prognosis of these patients. Classically, the treatment of ascites in patients with cirrhosis has been based on the combination of a sodium-restricted diet and the administration of diuretics. However, this treatment is not entirely satisfactory, since it is associated with a relatively high incidence of side-effects, and about 20% of patients hospitalized for the treatment of an episode of ascites do not respond to such therapy. In the last two decades, alternative therapies to diuretics have been introduced. PVS is an effective method of treating ascites. The high incidence of complications observed in early studies may be reduced by adequate perioperative management and careful selection of patients. The role of the PVS in the treatment of cirrhotics with ascites, however, still remains to be established. Recently, paracentesis has emerged as an alternative method of treating ascites in patients with cirrhosis. Several studies have shown that therapeutic paracentesis plus i.v. albumin infusion is more effective than conventional diuretic therapy and is associated with a lower incidence of complications. It has also been demonstrated that therapeutic paracentesis without the i.v. administration of albumin is associated with a marked increase in plasma renin activity, suggesting an impairment of effective blood volume, and with the development of hyponatraemia and/or renal failure in 20% of cases. Therefore, the i.v. administration of albumin is an essential measure in preventing the impairment of systemic haemodynamics and renal function that frequently follows the mobilization of ascites by paracentesis.