Oesophageal varices (OV) are a common sequelae of liver cirrhosis often leading to significant morbidity and mortality. Traditionally, patients with liver cirrhosis have undergone variceal surveillance by means of oesophago-gastroduodenoscopy (OGD). The Baveno VI guidance proposes that those with a platelet count of greater than 150×109/L and transient elastography (Fibroscan) reading of <20 kPa have a very low risk of having OV requiring treatment and can subsequently avoid screening OGD. We sought to apply these criteria to the annual screening workload of a regional liver unit to assess what proportion of screening OGDs could be safely avoided.Methods
A retrospective analysis was carried out of all OGDs performed for assessment of oesophageal or gastric varices by the hepatology department in a 12 month period (2016) in the Regional Liver Unit, Royal Victoria Hospital, Belfast. Data was retrieved from the endoscopy unit database (Unisoft) and patient information was obtained using the regional Electronic Care Record. Exclusion criteria included pre-hepatic or pre-sinusoidal portal hypertension (n=14), previous banding or glue therapy (n=166), TIPS (n=6), emergency endoscopy for acute bleeds (n=51) and Childs C liver disease (n=7). Transient elastography (TE) and platelet count (performed within a year of endoscopy) were assessed alongside OGD Result.Results
Of the 509 OGDs carried out in 2016, 244 were excluded due to the above criteria, leaving 265 who had OGD for varices screening. 183 (69%) of the 265 screened patients had not undergone TE due to being diagnosed with cirrhosis radiologically or histologically. This left 82 valid subjects who could be assessed by Baveno VI criteria. 24 (29%) of the valid subjects fulfilled the Baveno VI criteria to avoid screening, 20 of whom had no OV whilst the remaining 4 had 1 column of ‘barely noticeable’ or ‘possible’ OV. None of the patients required a therapeutic intervention. 59 subjects had either platelet count of <150 or TE scores of >20 kPa and therefore, by Baveno VI guidance should undergo screening endoscopy. Of these, 35 did not have OV and 24 (40.1%) had OV including 11 who required a drug intervention with beta blocker.Conclusions
At least 10% (24 of 244) of those undergoing OGD screening for varices in a regional liver unit could safely avoid OGD if Baveno VI criteria were applied. This number could be significantly higher if TE was used to assess every patient before screening OGD – 69% of those who had screening OGD in our unit did not have TE and therefore could not have Baveno VI criteria applied.