OBJECTIVE: To examine elevations in levels of serum glutamic oxaloacetic transaminase (SGOT) in patients with rheumatoid arthritis (RA) and osteoarthritis (OA) takingplacebo, aspirin, or diclofenac, and to seek possible explanations for the occurrence of these elevations. METHODS: We conducted a meta-analysis of individual case reports from 3 RA protocols and 5 OA protocols, encompassing 814 diclofenac-treated patients, 443 aspirin-treated patients, and 359 placebo-treated patients. All of the RA protocols had nearly identical inclusion and exclusion criteria, as well as safety studies and followup; the same was true for the OA protocols. Analysis included correlation analysis and multiple linear and logistic regression, accounting for numerous potential confounding variables, with the SGOT as the dependent variable. F tests were used for hypothesis testing. RESULTS: By several analytic approaches, the principal determinants of SGOT concentrations were found to be baseline SGOT value, the use of aspirin in RA patients, and the use of diclofenac in OA patients. Other significant factors contributing to an increase in SGOT concentrations were duration of therapy and, perhaps, daily dosage (mg/lb). Hypothesis testing supported these results. Given a statistically average patient, we predicted a 1-2% chance of a mildly elevated SGOT level occurring among placebo-treated patients, a 6-7% chance among diclofenac- or aspirin-treated patients with RA, a 12% chance among diclofenac-treated patients with OA, and a 2% chance among aspirin-treated patients with OA. CONCLUSION: This study demonstrates a powerful method for performing meta-analysis, using available individual patient data to examine numerous factors that may affect an outcome of interest. In this case, mild elevations of SGOT were examined and found to be related to baseline SGOT levels, diclofenac use (in OA), and aspirin use (in RA). Of numerous other potential factors examined, including age, sex, alcohol use, concomitant medications, and concomitant diagnoses, only duration of therapy and, to a small extent, daily dosage, were also consistent determinants of SGOT elevation. The SGOT elevations were minimal and were not related to the occurrence of clinical hepatitis: No clinical hepatitis occurred.
RCT Entities:
OBJECTIVE: To examine elevations in levels of serum glutamic oxaloacetic transaminase (SGOT) in patients with rheumatoid arthritis (RA) and osteoarthritis (OA) taking placebo, aspirin, or diclofenac, and to seek possible explanations for the occurrence of these elevations. METHODS: We conducted a meta-analysis of individual case reports from 3 RA protocols and 5 OA protocols, encompassing 814 diclofenac-treated patients, 443 aspirin-treated patients, and 359 placebo-treated patients. All of the RA protocols had nearly identical inclusion and exclusion criteria, as well as safety studies and followup; the same was true for the OA protocols. Analysis included correlation analysis and multiple linear and logistic regression, accounting for numerous potential confounding variables, with the SGOT as the dependent variable. F tests were used for hypothesis testing. RESULTS: By several analytic approaches, the principal determinants of SGOT concentrations were found to be baseline SGOT value, the use of aspirin in RA patients, and the use of diclofenac in OA patients. Other significant factors contributing to an increase in SGOT concentrations were duration of therapy and, perhaps, daily dosage (mg/lb). Hypothesis testing supported these results. Given a statistically average patient, we predicted a 1-2% chance of a mildly elevated SGOT level occurring among placebo-treated patients, a 6-7% chance among diclofenac- or aspirin-treated patients with RA, a 12% chance among diclofenac-treated patients with OA, and a 2% chance among aspirin-treated patients with OA. CONCLUSION: This study demonstrates a powerful method for performing meta-analysis, using available individual patient data to examine numerous factors that may affect an outcome of interest. In this case, mild elevations of SGOT were examined and found to be related to baseline SGOT levels, diclofenac use (in OA), and aspirin use (in RA). Of numerous other potential factors examined, including age, sex, alcohol use, concomitant medications, and concomitant diagnoses, only duration of therapy and, to a small extent, daily dosage, were also consistent determinants of SGOT elevation. The SGOT elevations were minimal and were not related to the occurrence of clinical hepatitis: No clinical hepatitis occurred.