Predictions of the prevalence of multiply revised knee arthroplasties to 2050: results of a multistate model based on national Danish registries

AIM: To estimate the prevalence of multiply revised knee arthroplasties in Denmark from 1998 to 2021 and to project the prevalence to 2050. METHODS: Three national Danish registries were queried for primary knee arthroplasties in the period of 1998-2021. A multi-state model was created using sex- and age-stratified proportional hazard functions to estimate the probabilities of receiving a primary arthroplasty, revision or dying. These probabilities were then applied to historical and projected population numbers from Statistics Denmark. Prevalence was defined as the number of revision knees per primary knees. RESULTS: In 2020, the number of knees alive with a primary knee arthroplasty was 101,956 knees. The prevalence of 1st revisions was 53.1 knees per 1000 primary knees, the prevalence of 2nd revisions was 9.5 per 1000 primary knees, the prevalence of 3rd revisions was 2.6 per 1000 primary knees, and the prevalence of 4th revisions was 1.4 per 1000 primary knees. In 2050, the number of knees alive with a primary knee arthroplasty was projected to increase 48%, while the prevalence of 1st revisions, 2nd revisions, 3rd revisions and ≥4th revisions was projected to increase 30.1%, 30.1%, 54%, and 114%. CONCLUSION: For knees that have been revised, we project large increases in the prevalence especially for knees that have been revised at least three times. A high proportion of knees with a revision knee arthroplasty in 2050 will be in patients above 80 years of age. Our model can be used in other populations to make future projections.