Article Text

SAT0457 A Markedly Higher Proportion of The Variance in Hip Arthroplasty than in Knee Arthroplasty Can Be Explained by Heritability
  1. K. Magnusson1,
  2. E. Ystrøm2,
  3. R. Ørstavik2,
  4. T. Nilsen2,
  5. C. Nielsen2,
  6. Ó.A. Steingrímsdόttir2,
  7. P. Ferreira3,
  8. K. Scurrah3,
  9. A.M. Fenstad4,
  10. O. Furnes4,
  11. K.B. Hagen1
  1. 1Rheumatology, Diakonhjemmet Hospital
  2. 2Norwegian Institute of Public Health, Oslo, Norway
  3. 3Australian Centre of Excellence in Twin Research, University of Melbourne, Melbourne, Australia
  4. 4Orthopaedic Surgery, Haukeland University Hospital, Bergen, Norway


Background Osteoarthritis (OA) may lead to costly total joint replacements (TJR) of the knee and hip joints. To what extent TJR due to OA at the different joint sites is caused by hereditary or environmental factors is unknown. This information may aid in determining whether hip or knee arthroplasties are possible to prevent.

Objectives To explore the genetic contribution to TJR due to hip and knee OA for women, men, young and elderly using a twin registry design.

Methods The study is a prospective cohort study based on a linkage of the Norwegian Twin Registry and the Norwegian Arthroplasty Registry using the National ID number in 2014. The presence of a TJR due to hip and knee OA has been registered for >95% of the population from 1987 (hip) and 1994 (knee). We first calculated prevalence estimates across twin groups (same sex twin pairs born 1915–1960), and subsequently examined the relative contribution of genetic factors (heritability), shared and non-shared environmental factors, applying the classical twin model.

Results The sample comprised 9029 complete pairs of which 3803 monozygotic (MZ) and 5226 dizygotic (DZ) pairs (54% females and mean (SD) age 71.5 (12.7) for both zygosity groups). The prevalence for having a TJR was similar across zygosity with 3.6% (n=614) for the hip and 1.8% (n=317) for the knee joint. In total 50 MZ/30 DZ and 10 MZ/5 DZ twin pairs were concordant for hip and knee arthroplasty, respectively. Concordances and polychoric correlations were higher for MZ twins than DZ twins both for hip and knee artrhoplasty in all age and sex strata (Table). A model including only genetic and environmental factors unique to each twin in a pair provided the best fit. The heritability for TJR largely depended on OA site. A markedly higher proportion of the variance in hip arthroplasty than in knee arthroplasty could be explained by heritable factors (Table). The variance in TJR due to knee OA was to a greater extent explained by environmental factors unique to each twin (Table). A significant gene-age interaction could be observed for the knee joint with a high observed heritability for twins below median age (68 years) changing to a low heritability after median age. Heritability estimates for TJR due to hip OA remained high and stable across age and sex groups, although a slightly higher heritability could be observed for men (Table).

Conclusions The heritability of TJR due to OA depends on joint site. Preventive interventions may have a greater potential for knee arthroplasty than hip arthroplasty, particularly for knee arthroplasty at old age. The difference in heritability across age and sex groups for TJR due to knee OA should be further studied due to a low number of concordant pairs.

Disclosure of Interest None declared

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.