811 during the previous week. The likelihood of this finding being due to chance alone was less than one in a thousand. These observations support other evidence' 1'5 that aspirin causes acute bleeding from gastric erosions. In this series at least one-third of the acute bleeds appeared to have been initiated by aspirin. Is it possible to define a population at risk ? Croft' 1016 has suggested that resistance to the effect of aspirin in inducing occult bleeding may be due to a higher than

more »

... rate of turnover or replacement of the surface epithelial cells of the stomach. There is no clear answer, however, for acute episodes of bleeding. Parry and Wood investigated the problem by studying 39 of their patients in whom there was strong circumstantial evidence that aspirin had precipitated the haemorrhage. Once the patients had recovered they were given aspirin continuously for a few days and occult blood loss was measured. The subjects were found to be no more susceptible to aspirin, as measured by this test, than was a control group. This suggests that aspirin initiates acute bleeding only at times and under circumstances when the gastric mucosa is temporarily susceptible. It "takes the mucosa by surprise." The factors responsible for enhanced susceptibility require investigation, and may include alterations in the turnover or healing power of atrophic areas of gastric mucosa. What should the clinician's attitude be to aspirin ? When given continuously it is an invaluable treatment for rheumatoid arthritis and other chronic rheumatic conditions. Under these circumstances iron deficiency anaemia occurs occasionally and patients may need treatment with iron. But there is at present no method of predicting the circumstances in which a patient may be susceptible to the risk of acute bleeding. Those who have had an episode of bleeding should be warned to avoid aspirin. Paracetamol, which does not cause bleeding,17 is a satisfactory analgesic for occasional use and in intermittent doses it is unlikely to have any other deleterious effects. Surely the time has come to advise patients to keep this drug, rather than aspirin, in the medicine cupboard. Operations on the Gall Bladder Modern biliary surgery started 100 years ago when J. S. BQbbs' performed the first cholecystotomy, though it was a completely unplanned procedure-the preoperative diagnosis was an ovarian cyst. Despite this century of experience there is still controversy about some aspects of surgery in this region, particularly at the two ends of the spectrum of disease of the gall bladder-symptomless gall stones and acute cholecystitis. Recent careful studies have provided factual information24 to add to the mass of clinical impressions on these topics. Gall stones are extremely common and their frequency has been well shown by several necropsy studies. Thus G. Horn2 reviewed some 5,000 necropsies on adults in Birmingham. The overall incidence of biliary stones was found to be 5.8% in males and 13.2% in females. The incidence increased as age advanced so that about 20% of subjects over the age of 70 had calculi. The sex difference was much less definite in the older age groups. M. W. Comfort and his colleagues3 studied 112 patients in whom stones were discovered incidentally in the course of x-ray investigations and found that over a 20-year period half the patients had developed symp-toms due to the stones. Recently A. Wenckert and B. Robertson' described a similar study in Malmo, Sweden. There were 781 patients with gall stones demonstrated cholecystographically who had neither operation nor complications within the first year of the x-ray examination. They were then followed up for a period of 11 years. Half the patients remained symptom-free or had had only mild symptoms. One-third suffered attacks of biliary colic, and many of these were submitted to elective operation. In the remaining one-sixth serious complications occurred, including acute cholecystitis, obstructive jaundice, pancreatitis, and, in three cases, malignant change in the gall bladder. The incidence of complications was higher in patients who were 60 years of age or more, but there was no sex difference. These complications of gall stones are serious, painful conditions, and the conclusion to be drawn is that when gall stones are discovered incidentally in otherwise healthy patients elective surgery should be advised. Acute cholecystitis is usually treated conservatively in Britain, emergency operation being reserved for patients in whom the diagnosis is in doubt or where the signs indicate progression of the inflammatory process. Elective cholecystectomy is usually advised some three months after the acute episode. The risk of perforation of the inflamed gall bladder is small. H. Ellis and K. Cronin,5 for example, reviewed 795 patients admitted to the Radcliffe Infirmary, Oxford, with acute cholecystitis. Eleven (1.4%) had bile peritonitis, but only one of these had developed the condition while in hospital, on the seventh day of an undiagnosed illness. The mortality and morbidity rates for operations performed in the acute phase of cholecystitis are usually higher than those in elective surgery. K. A. Meyer and his colleagues,7 surveying a large series of cases in the department of surgery of the North Western University Medical School, Chicago, found the overall death rate in their series with acute cholecystitis to be 4.5 % as compared with 1. 1% in those who underwent surgery for chronic cholecystitis. Indeed the mortality rate for cholecystectomy alone when carried out for acute cholecystitis was higher than that for those patients operated upon for chronic cholecystitis even when the latter group had a common duct exploration performed at the same time. Similar results have been obtained by other workers. C. G. McEachern and R. E. Sullivan7 reported a mortality rate of 5.8% in 155 patients who had emergency cholecystectomy performed for acute cholecystitis, though it should be noted that five of their nine patients who died had perforated gall bladders, and A. W. Hargreaves' has recently reported 55 cases treated by emergency surgery at the Salford Royal Hospital ; there were three deaths and nine cases of wound infection. The advocates of early surgery in the acute phase of cholecystitis argue that the operation is easily performed (owing to the presence of inflammatory oedema), the risk of bile duct or vascular damage is small in experienced hands, and readmission to hospital is avoided. Moreover, the later elective operation is often made difficult by the development of dense adhesions around the gall bladder and its ducts.