Genetic liability to osteoarthritis may be greater in women than menBMJ 1996; 313 doi: https://doi.org/10.1136/bmj.313.7051.232 (Published 27 July 1996) Cite this as: BMJ 1996;313:232
- Jaakko Kaprio,
- Urho M Kujala,
- Leena Peltonen,
- Markku Koskenvuo
- Senior research fellow Department of Public Health, University of Helsinki, FIN-00290 Helsinki, Finland
- Chief physician Unit for Sports and Exercise Medicine, Institute of Biomedicine, University of Helsinki
- Professor Department of Human Molecular Genetics, National Public Health Institute, Helsinki
- Professor Department of Public Health, University of Turku, Turku, Finland
EDITOR,—Tim D Spector and colleagues report a study of genetic influences on osteoarthritis in monozygotic and dizygotic female twins.1 They conclude that their results show a clear genetic effect for radiographic osteoarthritis of the hand and knee in women (ranging from 39% to 65%) independent of known environmental or demographic confounders. In his commentary on the paper John Hopper states that these figures must be considered to be an upper bound of the genetic influence.1
In a questionnaire sent to the Finnish twin cohort in 1990 we asked about the prevalence of osteoarthritis (in any joint) diagnosed by a physician.2 Self reports tend to be specific but not very sensitive. In the age group 33-60, 1757 male twin pairs (577 monozygotic and 1180 dizygotic) and 2338 female twin pairs (836 monozygotic and 1502 dizygotic) responded. Both twins reported osteoarthritis in eight (1.4%) monozygotic and 24 (2.0%) dizygotic male pairs and in 26 (3.1%) monozygotic and 43 (2.9%) dizygotic female pairs. Only one twin reported osteoarthritis in 66 (11.4%) monozygotic and 155 (13.1%) dizygotic male pairs and in 85 (10.2%) monozygotic and 219 (14.6%)dizygotic female pairs.
In male pairs the tetrachoric correlation coefficient for liability to osteoarthritis was 0.34 (SE 0.12) in monozygotic pairs and 0.38 (0.07) in dizygotic pairs. Model fitting indicated that the data were adequately accounted for (χ2 = 2.09, P = 0.72) by a model with shared (non-genetic) familial effects accounting for 37% of the total variance in liability to the disease and unshared familial effects accounting for 63%. Models that included genetic effects fitted significantly worse.
In female pairs the tetrachoric correlation coefficient for liability to osteoarthritis was 0.62 (0.07) in monozygotic pairs and 0.42 (0.06) in dizygotic pairs. Model fitting indicated that the data were best accounted for (χ2 = 3.82, P = 0.28) by a model with additive gene effects explaining 44% of the variance in liability to the disease, shared familial effects accounting for 20%, and unshared familial effects accounting for 36%. Models that did not include genetic effects fitted significantly worse.
Our data for women are consistent with those of Spector and colleagues,1 who report that 41-46% of the variance in the narrowing trait score for all joints in their subjects was accounted for by additive genetic effects. In contrast, our results for men indicate that genetic effects may be modulated by sex or by environmental factors distributed differently between men and women. We are currently investigating our affected pairs. Until now, however, investigators have held genetic defects responsible for only a small fraction of cases of osteoarthritis.