Reduction in mortality from upper gastrointestinal haemorrhage

Evaluation of the patient with gastrointestinal bleeding: an evidence based approach

Gastrointestinal bleeding is a common and potentially life-threatening problem. The incidence of upper gastrointestinal bleeding (UGIB) is 40 to 150 episodes per 100,000 population per year. Mortality in large series is 6% to 10% for UGIB and has remained unchanged over the past 30 to 60 years. The incidence of hospitalization for acute lower gastrointestinal bleeding is about 20 to 27 episodes per 100,000 population per year, with a 200-fold increase with advancing age from the third to ninth decades. The mortality rate is 4% to 10% or higher. The evaluation of overt or acute gastrointestinal bleeding in the ED is reviewed here from the perspective of evidence-based medicine.

Prediction of recurrent bleeding after endoscopic haemostasis in non-variceal upper gastrointestinal haemorrhage

Endoscopic haemostasis by injection of adrenaline was attempted in 135 consecutive patients with active upper gastrointestinal bleeding. Initial haemostasis was obtained in 127 patients following injection of 5-15 ml 1:10,000 adrenaline; eight patients in whom haemostasis was not achieved underwent immediate laparotomy. There was further haemorrhage in 25 patients, which was successfully treated by further injection of adrenaline in ten. Fifteen patients had major rebleeding requiring emergency surgery. Stepwise logistic regression analysis identified three factors that, taken together, were highly predictive of the need for surgery: pulse rate on admission, the position of the ulcer and whether the patient was obese. A scoring system was derived from the logistic analysis equation that was found to predict correctly the need for emergency surgery in 84 per cent of patients. In patients with a high probability of rebleeding surgery should be considered after initial endoscopic haemostasis and stabilization. In the majority of patients endoscopic treatment alone is sufficient for permanent haemostasis.

A review of 1338 patients with acute upper gastrointestinal bleeding at Chulalongkorn University Hospital, Bangkok

Acute upper gastrointestinal bleeding (UGIB) is a major gastroenterological problem in Thailand. Peptic ulcer has remained the leading cause of UGIB, followed by acute mucosal erosions, and variceal bleeding. In the present study, covering the period between 1980 and 1984, early endoscopic diagnosis was obtained in 82% of cases, while conservative treatment was effective in 83.6%. Factors associated with increased mortality were advanced age, shock, sepsis nonsteroidal anti-inflammatory drug (NSAID) ingestion, severe blood loss, associated diseases and surgical treatment. The overall mortality was 6.7 percent.

Factors influencing mortality from bleeding peptic ulcers

The mortality, causes of death, and the factors that are influencing deaths from bleeding acute and chronic peptic ulcers were evaluated retrospectively. During a 2-year period (1986-87) 272 patients were endoscoped for acute gastrointestinal hemorrhage and were found to have bled from a peptic ulcer (chronic gastric ulcers, 90; chronic duodenal ulcers, 114; acute gastroduodenal ulcers, 66; stomal ulcers, 2). The overall mortality was 9.6% (n = 26) (gastric ulcers, 6.7%; duodenal ulcers, 11.4%; acute ulcers, 10.6%). Statistically significant factors adversely affecting prognosis for gastric ulcer were re-bleeding, need for operation and serious intercurrent illness; for duodenal ulcer were units of transfused blood, re-bleeding, signs of recent hemorrhage at endoscopy, need for operation, and serious intercurrent illness; and for acute ulcer were increasing age, shock, units of transfused blood, re-bleeding, and serious intercurrent illness. Multivariate analysis was only attempted for duodenal ulcers because of sample sizes; it suggested that active bleeding or visible vessel at endoscopy, re-bleeding, and serious intercurrent illness were independent factors for mortality. From this study it is apparent that the major determinants of a fatal outcome in bleeding peptic ulcer diseases were serious intercurrent illness and rebleeding. As it must be anticipated that patients with these particular problems are at high risk of a poor outcome, it follows that it is important they have access to skilled treatment as provided by a specialist team in an intensive care ward. It also follows that every effort should be made to keep ulcers in remission to free the ulcer patient of potentially lethal complications.

Audit of mortality in upper gastrointestinal bleeding

The outcome in 1017 patients with haematemesis and malaena referred to two major hospitals in Nottingham within a 2-year period has been prospectively evaluated. Ninety one (9%) patients died during the time period under consideration and all but four were found to have been over 60 years of age. There were 13 (14%) deaths following rebleeding, of whom 5 (5%) could have been potentially avoided by alterations in management. Some improvement of mortality might result from intensive-care facilities with better management of transfusions and earlier detection of rebleeding allowing earlier endoscopy or surgery. The majority of patients (81%), however, died from concomitant disease which was exacerbated by gastrointestinal haemorrhage or bleeding developed in patients with an already existing end-stage disease. These results show that a reduction of mortality in patients with gastrointestinal bleeding is hampered by the high number of poor-risk patients. The rise in the proportion of elderly patients with this disorder seems to continue.

Endoscopic haemostasis for non-variceal upper gastrointestinal haemorrhage

Endoscopic haemostasis can be effective in non-variceal upper gastrointestinal haemorrhage, and should be regarded as potential front-line treatment. Diverse methods are available, and although no single technique has become firmly established, current evidence favours thermal coagulation and injection therapy.

Surgical management of bleeding gastric ulceration

In a nine-year period from 1977 to 1985 sixty-one patients underwent surgery for bleeding gastric ulceration in the Western Infirmary, Glasgow. Nineteen patients were treated by partial gastrectomy, twenty-two had undersewing of the ulcer plus vagotomy and drainage and twenty had undersewing alone. Mortality in the three groups was 26, 45 and 10 per cent respectively. All groups of patients were similar in terms of age, severity of haemorrhage, delay before surgery and grade of surgeon performing the procedure. Out-patient follow-up (mean: 37 months) of patients treated by undersewing alone revealed that 73 per cent were symptom free. Treatment of bleeding gastric ulceration by undersewing alone is effective and should be considered in patients who require surgery.

Tranexamic acid as an aid to reducing blood transfusion requirements in gastric and duodenal bleeding

A prospective randomised double blind study examined the effect of the antifibrinolytic drug tranexamic acid compared with placebo in 154 patients bleeding from verified benign lesions in the stomach or duodenum or both. Three out of 72 patients receiving tranexamic acid underwent emergency surgery compared with 15 out of 82 given placebo (p = 0.010). Nineteen patients receiving placebo rebled during their admission as compared with 10 in the active treatment group (p = 0.097). Blood transfusion requirements were significantly reduced by tranexamic acid (p = 0.018). Side effects occurred in six patients, of which an uncomplicated deep venous thrombosis was the most severe. Tranexamic acid reduces the blood transfusion requirement and need for emergency surgery in patients bleeding from a benign gastric or duodenal lesion.

Clinical factors in the prediction of further haemorrhage or mortality in acute upper gastrointestinal haemorrhage

Data collected prospectively from 326 admissions for acute upper gastrointestinal haemorrhage were examined to identify factors predicting further haemorrhage or mortality. Seven predictive factors were identified by univariate analysis for both further haemorrhage and/or mortality, but only age over 60 years, an admission haemoglobin less than 8 g/dl and the presence of endoscopic stigmata of recent haemorrhage were shown by stepwise logistic regression to have independent significance for further haemorrhage. An age over 60 years, the presence of clinical shock on admission and an episode of further haemorrhage following admission emerged as independently significant in the prediction of mortality.

Controlled trial of small bipolar probe in bleeding peptic ulcers

204 of 460 patients with upper gastrointestinal bleeding admitted to a busy district hospital were found to be bleeding from peptic ulcers or to have signs of recent haemorrhage at endoscopy within 24 h of admission. To determine if the small bipolar probe could stop bleeding or rebleeding, patients were allocated to electrocoagulation (101) or not (103); other aspects of treatment were identical. Groups were stratified by ulcer site to give similar numbers in each. To allow for differences in sex, age, initial haemoglobin, presence of other diseases, and shock, data were analysed by logistic regression. Fewer patients in the treated group (17) continued to bleed or rebled compared with controls (34). Rebleeding and mortality rates in the treated group were higher early in the trial, suggesting the need for experience in application of the probe. Further improvements in technology and technique may result in significant reductions in mortality.

A ten year prospective experience with semi-elective shunt in selected patients for bleeding oesophageal varices

In a prospective study from 1972 to 1982, 38 patients with bleeding oesophageal varices and relatively good liver function were treated by semi-elective or emergent shunt 2-4 weeks after admission. There were no deaths within 3 months of surgery. This constituted 30% of the total 125 patients admitted with bleeding varices during that period. No referred case for treatment of portal hypertension was included in this study. The procedures performed were end-to-side portacaval shunt in 30 cases, lieno renal shunt in seven cases and interposition mesenterico-caval shunt in one case. All patients were admitted to a special unit with a two stage policy of management. Immediate endoscopic diagnosis and balloon tamponade in those with continued bleeding was followed by shunt in selected cases. Follow-up of the 38 patients showed a cumulative survival at 1 year of 89%, at 3 years of 75% and 5 years of 65%. In four cases (13%) a clinical diagnosis of portal systemic encephalopathy was made, all were controlled by medical management. In four of ten follow-up deaths, liver failure was the cause, in none of these cases was encephalopathy a problem. It is concluded that with a policy of early diagnosis and control of haemorrhage, 30% of a typical series of prospectively studied patients admitted with bleeding varices can be treated without mortality by definitive surgery. There was low incidence of encephalopathy and no cases of incapacitating mental confusion. Centralization of treatment and prospective study is essential for the implementation of such a policy of management.