Accuracy and clinical utility of a peri-operative risk calculator for total knee arthroplasty

J Arthroplasty. 2013 Mar;28(3):445-8. doi: 10.1016/j.arth.2012.08.014. Epub 2012 Nov 9.

Abstract

Utilizing the Medicare Provider Analysis and Review dataset, a, peri-operative total knee arthroplasty (TKA) risk calculator was created based on select preoperative comorbidities. We retrospectively identified and reviewed 2284 primary TKAs at a single institution from 2000-2008. A numerical, predicted complication risk was established for each patient. Actual complications occurring within the first 14 post-operative days were recorded. Statistical analysis was performed using the C-statistic and ANOVA test. Patients with higher predicted probability of a complication did show higher complication rates. The corresponding C-statistic was 0.609. (95% Confidence Interval: 0.542-0.677). When the patients were divided into 4 groups based on their calculated complication risk (0-5%, 5-10%, 10-25%, >25%) the statistical significance of the associated ANOVA was P < .001, showing that patients with higher predicted risk experienced more complications, and those with lower predicted risk experienced fewer complications. Based on our results, the calculator has predictive value and is clinically relevant.

MeSH terms

  • Adolescent
  • Adult
  • Aged
  • Aged, 80 and over
  • Arthroplasty, Replacement, Knee* / adverse effects
  • Female
  • Humans
  • Male
  • Middle Aged
  • Perioperative Period
  • Reproducibility of Results
  • Retrospective Studies
  • Risk Assessment*
  • Risk Factors
  • Young Adult