Predictors of mortality after total knee replacement: a ten-year survivorship analysis

N D Clement, P J Jenkins, I J Brenkel, P Walmsley

Research output: Contribution to journalArticlepeer-review

25 Citations (Scopus)


We report the general mortality rate after total knee replacement and identify independent predictors of survival. We studied 2428 patients: there were 1127 men (46%) and 1301 (54%) women with a mean age of 69.3 years (28 to 94). Patients were allocated a predicted life expectancy based on their age and gender. There were 223 deaths during the study period. This represented an overall survivorship of 99% (95% confidence interval (CI) 98 to 99) at one year, 90% (95% CI 89 to 92) at five years, and 84% (95% CI 82 to 86) at ten years. There was no difference in survival by gender. A greater mortality rate was associated with increasing age (p < 0.001), American Society of Anesthesiologists (ASA) grade (p < 0.001), smoking (p < 0.001), body mass index (BMI) < 20 kg/m(2) (p < 0.001) and rheumatoid arthritis (p < 0.001). Multivariate modelling confirmed the independent effect of age, ASA grade, BMI, and rheumatoid disease on mortality. Based on the predicted average mortality, 114 patients were predicted to have died, whereas 217 actually died. This resulted in an overall excess standardised mortality ratio of 1.90. Patient mortality after TKR is predicted by their demographics: these could be used to assign an individual mortality risk after surgery.

Original languageEnglish
Pages (from-to)200-4
Number of pages5
JournalJournal of Bone and Joint Surgery
Issue number2
Publication statusPublished - Feb 2012


  • Adult
  • Age Distribution
  • Age Factors
  • Aged
  • Aged, 80 and over
  • Arthritis, Rheumatoid
  • Arthroplasty, Replacement, Knee
  • Body Mass Index
  • Female
  • Humans
  • Male
  • Middle Aged
  • Prognosis
  • Prospective Studies
  • Risk Factors
  • Scotland
  • Smoking
  • Survival Analysis


Dive into the research topics of 'Predictors of mortality after total knee replacement: a ten-year survivorship analysis'. Together they form a unique fingerprint.

Cite this