Acute variceal bleeding (AVB) has historically accounted for up to 10% of all GI bleeds necessitating emergency out of hours endoscopy. These patients have a significantly poorer prognosis and higher re-bleed rate than non-variceal GI haemorrhage. In the last 3 years we have anecdotally noticed a significant reduction in AVB necessitating emergency endoscopy. This has coincided with the employment of dedicated hepatologists. We sought to qualify and quantify this reduction with the employment of one and then a second hepatologist at a university teaching hospital 18 months apart.Methods
This was a retrospective review identifying all AVB patients who underwent emergency endoscopy over a 3 year period between January 2015 and December 2017. Data was collected from the electronic database and the GI reporting tool. This included endoscopic findings, therapy performed and whether there was a previous history of AVB requiring endoscopy. A dedicated hepatologist was employed in January 2015 (period 1; 18 months) and a second hepatologist June 2016 (period 2; 18 months).Results
Prior to a dedicated hepatologist all patients were followed up by general gastroenterologists. There was no dedicated variceal banding programme. There were up to 300 acute GI bleed endoscopies a year with approximately 10% due to AVB. Both hepatologists began performing weekly dedicated oesophageal variceal screening and treatment endoscopy lists (between 1–2/week). During period 1, there were 30 AVB; 27/30 (90%) received therapy, in the remaining 3, banding could not be applied due to poor views and injection therapy or sengstaken tube placement was performed. Of those presenting; 21/30 (70%) had previous OGD and banding but only 8/21 (38%) had previously been on banding program.Results
During period 2, there were 20 AVB; 19/20 (95%) received therapy. 12/20 (60%) had previous OGD and banding, and only 3/20 (15%) were on a dedicated banding programme.Conclusions
Prior to a dedicated hepatologist the vast majority of those presenting with AVB have a history of previous AVB and are potentially avoidable. With the advent of dedicated banding lists (as well as closer follow up with more robust secondary prophylaxis) there has been a major reduction of AVB presenting. There has been a shift of ‘new’ AVB unknown to the system rather than previous existing patients having undergone prior therapy. This has equated to a significant reduction in AVB of 30% during period 1% and 56% during period 2. Dedicated ‘surveillance’ lists such as for Barrett’s have shown to reduce the incidence of late presentation of disease and we propose that dedicated varices surveillance and banding lists can reduce acute admissions.