2%). 'Other' was included in the diagnosis field in 18% of OGDs.(NB: Multiple indications and diagnoses can be entered for a procedure, hence sum of percentages is greater than 100% for these categories) Conclusion The majority of procedures were performed in the ! 50 age group, peaking between ages 70 to 79, although a quarter of all procedures were performed in people younger than 50. Close to half of OGDs are performed without sedation.Procedure uploads to the NED continue to increase exponentially year on year. The volume of data and high proportion of sites uploading allows unparalleled insights into OGD practice in the UK.

PEG and 15% RIG. The mean insertion time from diagnosis was 11 months (+/-14 months), with 30 day mortality 1.9%. Overall mortality was 87% with mean survival from diagnosis 23 months (+/-16 months).The majority of PEG insertions were arranged on dedicated lists with anaesthetic cover (83% vs 17%) compared with only a minority of RIG insertions (18% vs 82%). There was no significant difference between time to insertion (P = 0.78) and survival from diagnosis (P = 0.61) between the 2 cohorts. Conclusions Gastrostomy use in the South West is safe, with mortality rates below quoted literature. Gastrostomy practice differs between the 2 hospitals in the approach used but time to insertion and mean survival was not different. The authors intend to pursue a joint referral pathway for gastrostomy assessment. This would improve data collection quality, allowing future analysis of standardised variables to ascertain the most effective use of gastrostomies in these patients.

Serum urea level >6.5 mmol/L and change in urea concentration >1.3 fold from baseline predicted the need for endotherapy with a sensitivity/specificity of 85%/32% and 85%/ 46%, respectively; the latter predicted severe AUGIB (X 2 =10.2, p=0.001). The corresponding area under the receiver operating curve (AUROC) were 0.59 (95% CI 0.49-0.69) and 0.71 (95% CI 0.61-0.81) (figure 1). In a subgroup analysis of patients with chronic kidney disease (n=46), AUROC for urea trend was 0.65 compared to that of absolute urea levels (0.53). There were no significant associations between urea concentrations, acute and chronic kidney disease with rebleeding rates at 72 hours, inpatient mortality and readmissions for AUGIB. Conclusions We demonstrate that an increase in urea concentration of >1.3 fold from baseline is superior at predicting severe AUGIB requiring endoscopic intervention and is a useful discriminator in patients with chronic kidney disease.

No abstract

No abstract