Risk of Revision Following Total Knee Arthroplasty or High Tibial Osteotomy: A Nationwide Propensity-Score-Matched Study

J Bone Joint Surg Am. 2019 May 1;101(9):771-778. doi: 10.2106/JBJS.18.00980.


Background: High tibial osteotomy (HTO) is often performed to postpone or avoid the need for subsequent total knee arthroplasty (TKA). We designed the present study to investigate the incidence rate and risk factors for subsequent revision in patients treated with HTO compared with those managed with TKA.

Methods: In this retrospective nationwide cohort study, we reviewed the South Korean National Health Insurance claims database from January 1, 2009, to August 31, 2017. We evaluated patients ≥41 years old who had undergone TKA or HTO as the primary surgical procedure without a history of having undergone either procedure during the preceding 2 years. By including only new interventions without such prior surgery, we could eliminate the influence of previous TKA and HTO treatments. Multivariable logistic regression models were used to compare the risk of revision between the groups after propensity score matching with inverse probability of treatment weighting (IPTW). Revision was defined as conversion to revision TKA after primary TKA and conversion to TKA after HTO.

Results: After applying the IPTW, a total of 436,538 patients with TKA and 452,724 patients with HTO were identified. The risk of revision during the entire study period was higher for patients with HTO than for patients with TKA (adjusted hazard ratio [HR], 2.47). The Kaplan-Meier 8-year survival was 97.8% in the TKA group and 91.5% in the HTO group. Compared with patients with TKA, patients with HTO had an increased risk of revision in cases of advanced age (HR of 1.85 for patients who were ≥61 to 69 years old and HR of 4.17 for those who were ≥70 years old), female sex (HR, 2.90), recipients of Medical Aid program benefits (HR, 4.77), the presence of hyperlipidemia (HR, 3.70), the presence of diabetes (HR, 4.86), and the presence of osteoporosis (HR, 3.53). However, younger patients with HTO (≤60 years old) had a lower risk of subsequent revision (HR, 0.64).

Conclusions: The risk of revision was higher for patients with HTO than for patients with TKA. The risk factors for subsequent revision in patients with HTO in our cohort of patients were advanced age (>60 years), female sex, receipt of Medical Aid, and the presence of comorbidities, such as diabetes, osteoporosis, and hyperlipidemia.

Level of evidence: Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.

Publication types

  • Comparative Study

MeSH terms

  • Adult
  • Aged
  • Arthroplasty, Replacement, Knee / adverse effects*
  • Female
  • Humans
  • Incidence
  • Male
  • Middle Aged
  • Osteoarthritis, Knee / complications
  • Osteoarthritis, Knee / surgery*
  • Osteotomy / adverse effects*
  • Postoperative Complications / epidemiology*
  • Postoperative Complications / surgery
  • Propensity Score
  • Reoperation*
  • Republic of Korea
  • Retrospective Studies
  • Risk Factors
  • Tibia / surgery*
  • Treatment Outcome