Abstract
Category:
Ankle Arthritis; Ankle
Introduction/Purpose:
Over the past decade, the incidence of total ankle arthroplasty (TAA) has markedly risen – particularly in relation to the incidence of total knee arthroplasty (TKA). Relatedly, individual patients have increasingly opted to undergo both procedures. Previous studies have elucidated differences in ankle biomechanics and symptomatology after TKA, however, there is a dearth of evidence in the existing literature suggesting whether sequencing of TAA or TKA may impact arthroplasty failure rate. This study examines how variable sequences of same-patient TAA and TKA impact arthroplasty failure rates. We hypothesized that rates of TAA and TKA failure would be higher in patients undergoing both procedures, and that the sequence of procedures would impact arthroplasty failure rates.
Methods:
The TriNetX Global Collaborative Network was queried for patients undergoing TAA only, TKA only, TAA followed by TKA, and TKA followed by TAA. All database queries were constructed using corresponding CPT and ICD codes in the TriNetX Query Builder interface. Rates of failure were compared between cohorts as defined by the following parameters: (1) TKA failures – revision TKA only; (2) TAA failures – revision TAA and/or ankle arthrodesis and fusion. Rates of failure were compared using chi-squared statistical testing. Cohort balancing was performed using propensity score matching according to age, race, and sex.
Results:
Non-sequenced comparison between those undergoing TAA and TKA, and those undergoing TKA only showed a significant increase in TKA failure in both unmatched (2.66% vs. 1.53%, p<.01) and matched (2.66% vs. 1.47%, p = .05) cohorts. Non-sequenced comparison showed no significant difference in TAA failure in both unmatched (2.93% vs. 2.67%, p=.60) and matched (2.93% vs. 3.02%, p=.89) cohorts. Sequenced comparison using unmatched cohorts showed a significant increase in TKA failure among patients who had TAA, then TKA, (2.81% vs. 1.57%, p=.032), but not among patients who underwent TKA then TAA (1.61% vs. 1.57%, p=.53). (Table 1). Using matched cohorts, there was no significant difference in TKA or TAA failure, irrespective of sequencing, however, balancing reduced cohort sizes below a database-defined floor of 10.
Conclusion:
We observed a higher rate of TKA failure in patients who also had TAA – specifically those who underwent TAA before TKA. Indications for TAA may alter knee biomechanics, predisposing subsequent TKA to failure. Previous studies have suggested identification of ankle center during TKA may be complicated by TAA. Gait disturbances found post-TAA may play a role. Some knee complaints may result from referred pain or mechanical disturbances at the ankle. Procedure laterality was not included in this study due to database constraints, thus, we suspect the effect of more biomechanically-consequential ipsilateral procedures may be larger than demonstrated here.
≤10 represents a database limitation whereby values between 1-10 all are represented as 10 p-values represent comparison against baseline values; significant <.05 Ø - neither TKA nor TAA
Abstract Details Continued