Adherence to guidelines in bleeding oesophageal varices and effects on outcome: comparison between a specialized unit and a community hospital.

Objectives Randomized controlled trials have shown beneficial effects of vasoactive drugs, endoscopic treatment and prophylactic antibiotics on the outcome of bleeding oesophageal varices (BOV). However, translating guidelines based on randomized controlled trials into clinical practice is difficult Our aims were to compare adherence to evidence-based guidelines in BOV between a specialized unit and a community hospital, and to investigate whether differences in adherence affected the outcome. Methods Two cohorts hospitalized during 2000-2007 with a first episode of BOV were retrospectively enrolled, one in a community hospital comprising 66 patients and one in a specialized unit comprising 111 patients. Data on treatment, rebleeding and mortality were collected from medical records according to the Baveno III/IV Criteria. Results Treatments in the specialized unit versus the community hospital were: vasoactive drugs 79 vs. 66% (P=0.06), prophylactic antibiotics 55 vs. 27% (P<0.01), endoscopic treatment 86 vs. 74% (P=0.04) and Sengstaken-Blakemore tube was used in 5 vs. 21 % (P<0.01). Secondary prophylaxis with pharmacological, endoscopic or transjugular intrahepatic portosystemic shunt therapy was initiated in 91 vs. 74% (P<0.01) (specialized vs. community). Six-week mortality was 17 vs. 24% (P=0.2 5 ) with 5-day mortality of 6 vs. 3% (P=0.34) and mortality day 6-42, 12 vs. 22% (P=0.07) (specialized vs. community). Failure to control bleeding and failure to prevent rebleeding were not statistically different. Conclusion Our study shows that patients with BOV are more likely to receive therapy according to guidelines when hospitalized in a specialized unit compared with a community hospital. This however did not affect mortality.