Non-steroidal anti-inflammatory induced upper gastrointestinal event rates in patients awaiting joint replacement in the United Kingdom. An epidemiologically-based burden of disease model

Objective: To quantify the likely morbidity associated with upper gastrointestinal (GI) ulceration associated with the use of non-steroidal anti-inflammatory drugs (NSAIDs) in patients awaiting hip and knee replacement surgery. Method: A burden of disease model was constructed, using Hospital Episode Statistics from the Department of Health in England, selected patient cohorts drawn from a pooled anonymised general practice database (MediPlus) and NSAID risk assessments from the published literature. Based on mean derived waiting times from the point of referral to admission for procedure and estimates of both patient exposure to NSAIDs, and the age-specific risk associated with this treatment, the excess risk associated with NSAIDs was then calculated. Results: In 2001-2, there were approximately 109000 hip and knee replacements carried out in the UK. The mean waiting time for hip and knee replacement, measured from the point of first referral by the GP, was found to be 443 days (95% CI 419-467). 73% (95% CI 72-74) of patients were found to be taking NSAIDs, equating to a total risk exposure of similar to96000 patient years. By applying known age-specific bleeding risks it was estimated that there were around 637 upper GI bleeds directly attributable to NSAID treatment in this cohort, with between 51 and 89 deaths resulting. Conclusions: In order to improve outcomes in this high-risk patient group, action must continue to reduce overall waiting times, while simultaneously adopting treatment regimes that are inherently less likely to cause upper gastrointestinal bleeding.