National survey of UK emergency endoscopy units

Introduction

Upper gastrointestinal bleeding is a common cause of hospital admission and is accompanied by considerable mortality. For patients to survive, the timing of endoscopy can be critical. Clinical scoring systems identify high risk patients who need prompt endoscopy after appropriate resuscitation.1 Early endoscopic intervention to prevent rebleeding is effective in high risk patients.2 A recent report indicated that patients are still dying as a consequence of delayed endoscopy,3 but no data exist on the provision of emergency endoscopy services in the United Kingdom. As part of a national census of endoscopy training units, we examined the extent of out of hours endoscopy provision, including volume of work and resources used.

Participants, methods, and results

We approached endoscopy units registered with the UK Joint Advisory Group. We developed a questionnaire from the British Society of Gastroenterology working party report,4 and distributed it to lead clinicians in 2002. We sent two reminders to centres that failed to reply. We finished collecting data by August 2002. The number of endoscopy rooms was a surrogate marker for the size of the unit. The response rate was 77% (150 centres).

Overall, 35 of the 150 units that responded (23%) did not provide an emergency out of hours endoscopy service. In the 115 (77%) units that did, this was provided by a median of five consultants and featured junior endoscopists in 47 units, acting independently in 15. Forty one units reported having an ad hoc or goodwill rota rather than a formal on-call arrangement. Out of hours procedures were done in the endoscopy department in 61 units, in theatre in 43 units, and on the ward for the remainder. Trained endoscopy staff helped the endoscopist in 49 units. Theatre staff support was needed in 47 units and ward staff in 15 units. Larger units tended to do the endoscopies in the endoscopy department, but there was no variation in location or staffing for units of smaller sizes (table). A mean of 90.2 (95% confidence interval 72.0 to 108.5) emergency endoscopies per 100 000 population were done each year for upper gastrointestinal bleeding, of which 26.7 were out of hours. Although larger units (including tertiary centres) received more patients with gastrointestinal bleeding and did more out of hours procedures this was not significant.

Comment

Hospitals that admit patients with acute upper gastrointestinal haemorrhage lack emergency endoscopy provision; hospitals need to manage about 100 patients per 100 000 population with acute upper gastrointestinal haemorrhage.1 Mortality from upper gastrointestinal bleeding remains high at 14%, and this has been attributed to the ageing population.1 Our survey indicates, however, that in many hospitals patients might be dying because of a lack of an appropriately timed endoscopy, which would identify high risk patients and offer the possibility of endoscopic therapeutic intervention.

Emergency endoscopies in high risk patients were often done in unfamiliar surroundings, with staff not used to dealing with such patients, conflicting with guidance issued by the British Society of Gastroenterology.4 Mortality in hospitals with a dedicated bleeding unit is almost half the national average,5 indicating that at least 40% of the deaths associated with gastrointestinal bleeding are preventable. We believe that one reason for this is the failure of many units to ensure that out of hours emergency rotas exist for such patients. Smaller units should consider combining with larger ones to provide cross cover and rectify a shortfall in the service that is essentially manpower related. Physicians and surgeons should work together in this important area so that 24 hour cover can be provided by a hospital equipped to deal with all aspects of serious gastrointestinal haemorrhage.

This article was posted on bmj.com on 11 March 2005: http://bmj.com/cgi/doi/10.1136/bmj.38379.662616.F7