L. HONG1, S. LEVY2, J.J. WARREN3, and B. BROFFITT3, 1University of Missouri-Kansas City, USA, 2University of Iowa, Coralville, USA, 3University of Iowa, Iowa City, USA |
Objectives: Amoxicillin use was reported to be associated with fluorosis on both primary and early-erupting permanent teeth. This analysis was to assess the association between amoxicillin use and fluorosis on late-erupting permanent teeth. Methods: As part of the Iowa Fluoride Study, subjects were followed from birth to 32 months with questionnaires every 3-4 months to gather information on fluoride intake and amoxicillin use. 307 subjects (52% female) were used for this analysis. Late-erupting permanent tooth fluorosis (canines, premolars and second molars) was assessed by three trained dentists using the Fluorosis Risk Index (FRI) at approximately age 13. A case was defined as fluorosis if two or more teeth had FRI zone II scores of 2 or 3. Results: 25.7% had fluorosis on at least two teeth. In bivariate analyses, amoxicillin use from 20-24 months significantly increased the risk of fluorosis on FRI classification II zones (37% vs 23%, RR=1.62, p=0.02), but other individual time periods were not. RR for amoxicillin use cumulatively from 20-32 months was 1.56 (32% vs 21%, p=0.056). Other factors (sex, birth weight, family income, parents' education, and use of other antibiotics) were not statistically significant. Multivariable logistic regression confirmed the increased risk of fluorosis for amoxicillin use from 20-24 months (OR=2.52, p=0.017) after controlling for otitis media, breast-feeding, and fluoride intake during the same time period. A similar trend was found when fluorosis prevalence was compared between amoxicillin users during 20-24 months and those who never used amoxicillin from birth to 32 months (37% vs 24%, RR=1.59, p=0.293), but it was not statistically significant. Conclusions: The results indicate that amoxicillin use during infancy could be a risk factor in the etiology of fluorosis on late-erupting permanent teeth, but further research is needed. (Supported in part by NIH grants 2R01-DE09551, R01-DE12101, 2P30-DE10126 and CRC-MO1RR00059) |