Severe Gastrointestinal Haemorrhage: Summary of a National Quality of Care Study with Focus on Radiological Services

To identify the remediable factors in the quality of care provided to patients with severe gastrointestinal (GI) bleeding.

All hospital admissions in the first four months of 2013 with ICD10 coding for GI bleeding who received a transfusion of 4 units or more of blood. Up to five cases/hospital randomly selected for structured case note peer review. National availability of GI bleeding services data derived from organisational questionnaire completed by all hospitals.

4563/29,796 (15.3%) of GI bleeds received 4 or more units of blood with a mortality rate of 20.2% compared to 7.3% without blood transfusion. 30.8% of GI bleeds received a blood transfusion. 32% (60/185) of hospitals admitting acute GI bleeds lacked 24/7 endoscopy. 26% (48/185) had on-site embolisation 24/7 with a further 34% (64/185) accessing embolisation by transfer within a validated formal network. Blood product use was inappropriate in 20% (84/426). Improved management, principally earlier senior gastroenterologist review and/or endoscopy, would have reduced blood product use in 25% (113/457). 14.5% (90/618) had a CT scan which identified the site of bleeding in 32% (29/90). 7.8% (36/459) underwent an Interventional Radiology (IR) procedure but a further 6.3% (21/33) should have had IR. 6% (36/586) underwent surgery with 21/36 for uncontrolled bleeding. In 20/35 IR was not considered despite the majority being suitable for IR. Overall 44% (210/476) received an acceptable standard of care according to peer review.

26 recommendations were made to improve the quality of care in GI bleeding, with six principle recommendations.