Background Circumstances in early life, including low birth weight, have been proposed as risk factors for development of rheumatoid arthritis (RA) in adulthood. The twin-control study design provides an opportunity to investigate the significance of potential prenatal determinants for adult morbidity by accounting for maternal characteristics and important early environmental and genetic factors.
Objectives To investigate whether birth weight is a risk factor for the development of RA in adulthood.
Methods We investigated the association between birth weight and RA in a sample of 14 monozygotic, 19 dizygotic same sexed and 9 opposite sexed twin pairs discordant for rheumatoid arthritis in which valid information on birth weight, length and order were available from midwife records.
Explorative analysis was performed with difference plot. Conditional logistic regression, using twin pairs as groups, was used to investigate the relationship between RA and birth weight adjusting for birth length, birth order and sex.
Results Mean birth weight did not differ between MZ and DZ twins while mean birth length of MZ twins was significantly higher than in DZ twins (p<0.003). Five (3 MZ, 1 DZss and 1 DZos) twin pairs had identical birth weight. The intra-pairwise differences, RA twin minus co-twin, ranged from -750 grams to 1100 grams with an overall mean of 78 grams (95% CI = -13 to 70) and 146 grams (95% CI = (-36 to 329) in MZ, 32 grams (95% CI = -90 to 154) in DZss and 69 grams (95% CI = -122 to 260) in DZos twin pairs. The odds ratio for birth weight was 1.00 (95% CI = 0.997 to 1.003), adjusting for birth length, birth order and sex. Restricting the analysis to ACPA positive RA did not change the results.
Conclusions Birth weight in itself is not a risk factor for the development of adult rheumatoid arthritis.
Disclosure of Interest None Declared