Article type
Year
Abstract
Introduction/Objective: Endoscopically proven gastric ulcers are present in 12% to 28% of patients taking regular NSAIDs. Recently, several low-risk gastrointestinal (GI) injury NSAIDs have become available. This study aimed at reviewing the literature to test 2 hypotheses: (1) Do low-risk NSAIDs, as a class, have lower GI toxicity (as measured by endoscopy) than other NSAIDs? (2) Is there a difference in the GI toxicity within the low-risk NSAIDs?
Methods: A Medline search was performed to identify randomized controlled endoscopic studies published from mid-1976 to December 1994 comparing the low-risk NSAIDs (etodolac, nabumetone and salsalate) with any other NSAID. All manuscripts were reviewed as well as their bibliographies. Eligibility criteria for this meta-analysis were: at least single-blind (endoscopist), endoscopy at baseline and end of the trial, and data reported about the frequency of multiple erosions and gastroduodenal ulcers. Relative risk reduction (RRR) was defined as the effectiveness measure. The number-needed-to-treat to prevent an ulcer/multiple erosions was calculated for each individual and for the low-risk class of NSAIDs.
Results: Six etodolac, 4 nabumetone and 5 salsalate studies were eligible for inclusion in the analysis. The overall number of subjects randomized in each drug comparison were 235, 245 and 152, respectively. The overall RRR (95% CI) for each of the low-risk NSAIDs were 0.86 (0.62-1.00), 0.85 (0.61-1.00) and 0.88 (0.74-0.99), respectively, and for all low risk combined was 0.87 (0.77-0.95). Seven, 8, 3 and 5 patients would need to be treated with etodolac, nabumetone, salsalate or any of these drugs, respectively, to prevent an additional ulcer/multiple erosion.
Discussion: Low-risk NSAIDs, individually or as a class, have lower GI toxicity as measured by endoscopic findings than other NSAIDs. There is no statistically significant difference within the low-risk NSAIDs.
Methods: A Medline search was performed to identify randomized controlled endoscopic studies published from mid-1976 to December 1994 comparing the low-risk NSAIDs (etodolac, nabumetone and salsalate) with any other NSAID. All manuscripts were reviewed as well as their bibliographies. Eligibility criteria for this meta-analysis were: at least single-blind (endoscopist), endoscopy at baseline and end of the trial, and data reported about the frequency of multiple erosions and gastroduodenal ulcers. Relative risk reduction (RRR) was defined as the effectiveness measure. The number-needed-to-treat to prevent an ulcer/multiple erosions was calculated for each individual and for the low-risk class of NSAIDs.
Results: Six etodolac, 4 nabumetone and 5 salsalate studies were eligible for inclusion in the analysis. The overall number of subjects randomized in each drug comparison were 235, 245 and 152, respectively. The overall RRR (95% CI) for each of the low-risk NSAIDs were 0.86 (0.62-1.00), 0.85 (0.61-1.00) and 0.88 (0.74-0.99), respectively, and for all low risk combined was 0.87 (0.77-0.95). Seven, 8, 3 and 5 patients would need to be treated with etodolac, nabumetone, salsalate or any of these drugs, respectively, to prevent an additional ulcer/multiple erosion.
Discussion: Low-risk NSAIDs, individually or as a class, have lower GI toxicity as measured by endoscopic findings than other NSAIDs. There is no statistically significant difference within the low-risk NSAIDs.