A G Smith1, G J Dovey, R A Cartwright. 1. Leukaemia Research Fund Centre for Clinical Epidemiology, University of Leeds, Institute of Epidemiology, Margaret Smith Building, Leeds, UK.
Abstract
PURPOSE: To investigate the possible role of topical chloramphenicol in the development of adult acute leukaemia. METHODS: The design of the study was a population-based age- and sex-matched case-control study, which collected cases of adult acute leukaemia between 1991 and 1996. Caucasian cases (807) and 1593 Caucasian controls were interviewed in person using a highly structured questionnaire. General practitioner medical records were abstracted for previous topical chloramphenicol use. RESULTS: 797 cases and 1570 controls were included in the analysis. No association was observed for topical chloramphenicol use and acute leukaemia (adjusted odds ratio, 1-year lag period (OR) 0.91 95% confidence interval (CI) 0.70-1.17). Similar results were observed when the analysis was repeated by diagnostic subgroup and sex. For all the data, a small, non-significant increased risk was observed (OR=1.21, 95% CI 0.65-2.25) if chloramphenicol had been prescribed three, or more times, but there was no statistically significant dose-response relationship (chi(2)=1.40, two-sided p=0.24). CONCLUSIONS: The results, based on a robust study design, show no evidence of an increased risk of developing adult acute leukaemia after topical chloramphenicol use. Copyright (c) 2000 John Wiley & Sons, Ltd.
PURPOSE: To investigate the possible role of topical chloramphenicol in the development of adult acute leukaemia. METHODS: The design of the study was a population-based age- and sex-matched case-control study, which collected cases of adult acute leukaemia between 1991 and 1996. Caucasian cases (807) and 1593 Caucasian controls were interviewed in person using a highly structured questionnaire. General practitioner medical records were abstracted for previous topical chloramphenicol use. RESULTS: 797 cases and 1570 controls were included in the analysis. No association was observed for topical chloramphenicol use and acute leukaemia (adjusted odds ratio, 1-year lag period (OR) 0.91 95% confidence interval (CI) 0.70-1.17). Similar results were observed when the analysis was repeated by diagnostic subgroup and sex. For all the data, a small, non-significant increased risk was observed (OR=1.21, 95% CI 0.65-2.25) if chloramphenicol had been prescribed three, or more times, but there was no statistically significant dose-response relationship (chi(2)=1.40, two-sided p=0.24). CONCLUSIONS: The results, based on a robust study design, show no evidence of an increased risk of developing adult acute leukaemia after topical chloramphenicol use. Copyright (c) 2000 John Wiley & Sons, Ltd.