Abstract
Purpose
Elective ACLR is indicated for symptomatic instability of the knee. Despite being a common procedure, there are numerous surgical techniques, graft and fixation choices. Many have been directly compared in randomized trials and meta-analyses. The typical operation is arthroscopic-assisted, uses autograft tendon and screw fixation. Research in elective joint replacement surgery has demonstrated an inverse relationship between surgeon volume and revision and complication rates. How patient demographics, provider characteristics and graft/fixation choices influence ACLR revision rates has not been reported on a population level. We hypothesized that ACLR using tendon autograft and screw fixation performed by high volume surgeons will have the lowest rate of revision. In contrast, the risk of contralateral ACLR in the same cohort will be influenced only by patient factors.
Method
All ACLR performed in Ontario from July 2003 to March 2008 on Ontario residents aged 14 to 60 were identified using physician billing, procedural and diagnostic codes from administrative databases. Data was accessed through the Institute for Clinical Evaluative Sciences. The main outcomes were revision and contralateral ACLR sought from inception until end of 2009. Patient factors (age, gender, income, co-morbidity), surgical choices (allograft or autograft tendon; screw, biodegradable or endobutton/staple fixation) and associated procedures (meniscal repair, collateral ligament surgery) were entered as covariates in a cox proportional hazards survivorship model. Mean cohort patient characteristics were chosen as reference groups. Surgical options and associated procedures were analyzed in a binary fashion (yes/no).
Results
We identified 13,997 primary ACLR with a mean follow up of 3.2 years. The rate of revision ACLR was 1.8% (mean 1.9 years), and primary contralateral ACLR 2.0% (mean 2.0 years). In the cox model, younger age (14–19 yrs; HR 2.9, p<0.001), teaching hospital (HR 2.1, p<0.001) and the use of endobutton/staple fixation (HR 4.4, p=0.01) conferred a higher risk of revision. No effect of graft type or surgeon volume was found. Only younger age (14–19 yrs; HR 1.9, p=0.0005) and not any provider or surgical covariates conferred a significant risk of contralateral ACLR.
Conclusion
Our results confirm that young age confers a higher risk of both revision and contralateral ACLR and these patients should be counseled accordingly pre-operatively. The use of endobutton or a staple for fixation was an independent risk for revision ACLR. This finding needs to be explored further in a direct fashion. Finally, we report that the mean time to revision ACLR was almost two years a fact that should impact future randomized controlled trial design and prompt re-evaluation of those already published which typically use only 1–2 years as the endpoint of data collection.