SAT0065 Meta-analysis of nsaid-induced gastrointestinal (gi) side effects in oa patients in japan

Abstract

Background GI intolerance is one of the major causes of failure of NSAID treatment for arthritis patients. There has been a large number of studies to examine the risk of NSAID-induced GI side effects in US and Europe, but few studies in Japan. Objectives In order to evaluate the GI tolerability of NSAIDs in OA patients in Japan, a systematic review of Japanese randomised controlled trials was performed. Methods This meta-analysis consisted of double blind randomised controlled trials with 4-weeks NSAID treatment in OA patients in Japan. The analysis was composed of 25 trials with 26 types of NSAIDs, among which 6 types were used in more than two trials. The endpoints evaluated in the meta-analysis included incidence of any side effects, any digestive side effects and symptoms indicative of upper GI symptoms such as abdominal pain, nausea and dyspepsia. The general variance-based method was used to estimate the risk of these endpoints. Additionally, to evaluate the effect of risk factors for GI discomfort, odds ratios (ORs) of side effects by sex and age were estimated for individual treatment arms and then combined. Results The analysis included a total of 4,725 patients. On average, cumulative incidences of patients who had experienced any side effects and any digestive side effects were 14.3% (95% CI: 13.3–15.3%) and 10.4% (95% CI: 9.4–11.4%), respectively. Incidence of any side effects by type of NSAID were as follows: diclofenac 75 mg/day, 19.9% (10 trials), indomethacin 75 mg/day, 16.2% (12 trials), acemetacin 90 mg/day, 12.6% (2 trials), nabumetone 800 mg/day, 9.2% (2 trials), fenbufen 600 mg/day 8.8% (2 trials), and ibuprofen 900 mg/day, 7.8% (2 trials). Incidence of any digestive side effects were as follows: diclofenac, 16.6%, indomethacin, 10.2%, nabumetone, 8.3%, fenbfen, 6.9%, and ibuprofen 900 mg/day, 6.4%. Abdominal pain, nausea/vomiting and dyspepsia were the three most common digestive side effects. The cumulative incidence for the upper GI symptoms was estimated to be approximately 10.9%. Comparing the risk of upper GI symptoms between male and female, the summary OR was 1.71 (95% CI: 1.11- 2.65). Females were at significantly higher risk than males in the incidence of upper GI symptoms. Comparing the risk of upper GI symptoms between patients under 59 years and above 60 years old, the summary OR was 1.07 (95% CI: 0.75–1.52). Conclusion Despite the incidence of side effects varying across the different types of NSAIDs being used, there was an obvious increased risk of GI symptoms. Abdominal pain, nausea/vomiting and dyspepsia were the three most common GI discomforts and this trend is similar to the evidence in US and Europe. Females show a higher risk of side effects and are approximately 50% more likely to have side effects compared to males. Older patients tend to have a higher risk of side effects, but this is not statistically significant.