Introduction The prevention of upper gastrointestinal bleeding (UGIB) can be facilitated by understanding the changes in environmental and socio-pathological factors; these might not become obvious in short-term studies. We, therefore, aimed to study the changes in comorbidity and risk scores and their influence on the outcomes of UGIB over a 14-year period.
Methods We analysed the clinical characteristics of all patients presenting with UGIB to a single institution, 1996–2010. The Charlson’s comorbidity and the complete Rockall scores were analysed, together with patients’ drug use and 30-day mortality. Trends with time were assessed using logistic regression analysis with year of presentation as a continuous predictor variable. Regression coefficients were expressed as odds ratios (OR), representing the relative change in odds of death or other binary dependent variables over a time interval of one year.
Results A total of 2669 patients were included. The Charlson score increased significantly with time (P < 0.001), the odds of a high (3+) score increasing at a relative rate of 4.4% a year (OR = 1.044, 95% CI 1.022–1.065). No significant trend with time was noted for age (p = 0.09), haemoglobin level (P = 0.47) or Rockall score (P = 0.94). The overall 30-day mortality was 4.9% and this showed no relationship with time (P = 0.28). However, when adjusted for the increasing comorbidity, the odds of death within 30 days decreased significantly at a relative rate of 4.5% per year [OR = 0.955 (0.914–0.997); P = 0.038]. Trends in the prevalence of taking potentially damaging and protective drugs are shown in Table-1, below. The rise in use of aspirin, other anti-thrombotic drugs and SSRIs [with pro-UGIB activity] was paralleled by a rise in the use of PPIs [protective activity] and beta-blockers, ACE inhibitors, and statins [being able to affect mortality.
Conclusion Patients with UGIB presented with progressively increasing comorbidity over the 14-year period 1996–2010. The use of both potentially damaging and protective drugs also increased. Raw 30-day mortality was unchanged over the same period, but decreased significantly with time when adjusted for comorbidity. These observations might have clinical implications.
Disclosure of Interest A. Taha Consultant for: Horizon Pharma USA; Vifor Pharma UK, E. Saffouri: None Declared, C. McCloskey: None Declared, T. Craigen: None Declared, W. Angerson: None Declared