Does interaction occur between risk factors for revision total knee arthroplasty?

Georges J Bounajem; Josh DeClercq; Garen Collett; Gregory D Ayers; Nitin Jain

doi:10.1007/s00402-023-05107-2

Does interaction occur between risk factors for revision total knee arthroplasty?

Arch Orthop Trauma Surg. 2023 Oct 30. doi: 10.1007/s00402-023-05107-2. Online ahead of print.

Authors

Georges J Bounajem^{1

2}, Josh DeClercq³, Garen Collett⁴, Gregory D Ayers³, Nitin Jain⁵

Affiliations

¹ Department of Orthopaedic Surgery, UT Southwestern Medical Center, Dallas, TX, USA. gbounajem@gmail.com.
² UT Southwestern Medical Center at Frisco, 12500 Dallas Parkway, 3rd Floor, Orthopaedic Surgery, Frisco, TX, 75033-9071, USA. gbounajem@gmail.com.
³ Department of Biostatistics, Vanderbilt University, Nashville, TN, USA.
⁴ Department of Orthopaedic Surgery, UT Southwestern Medical Center, Dallas, TX, USA.
⁵ Department of Physical Medicine and Rehabilitation, UT Southwestern Medical Center, Dallas, TX, USA.

PMID: 37902892
DOI: 10.1007/s00402-023-05107-2

Abstract

Introduction: Several risk factors for revision TKA have previously been identified, but interactions between risk factors may occur and affect risk of revision. To our knowledge, such interactions have not been previously studied. As patients often exhibit multiple risk factors for revision, knowledge of these interactions can help improve risk stratification and patient education prior to TKA.

Materials and methods: The State Inpatient Databases (SID), part of the Healthcare Cost and Utilization Project (HCUP), were queried to identify patients who underwent TKA between January 1, 2006 and December 31, 2015. Risk factors for revision TKA were identified, and interactions between indication for TKA and other risk factors were analyzed.

Results: Of 958,944 patients who underwent TKA, 33,550 (3.5%) underwent revision. Age, sex, race, length of stay, Elixhauser readmission score, urban/rural designation, and indication for TKA were significantly associated with revision (p < 0.05). Age was the strongest predictor (p < 0.0001), with younger patients exhibiting higher revision risk. Risks associated with age were modified by an interaction with indication for TKA (p < 0.0001). There was no significant interaction between sex and indication for TKA (p = 0.535) or race and indication for TKA (p = 0.187).

Conclusions: Age, sex, race, length of stay, Elixhauser readmission score, urban/rural designation, and indication for TKA are significantly associated with revision TKA. Interaction occurs between age and indication.

Keywords: Database; Revision; Risk factor; TKA.