Abstract
Objectives:
Postoperative motion loss is the most common complication following multiple ligament knee injuries (MLKIs). MLKIs are often associated with concomitant musculoskeletal and neurovascular injuries and complex surgical treatment; however, factors that increase the risk of postoperative motion loss remain largely unknown. The purpose of this study was to identify predictors of motion loss following MLKI. We hypothesized that injury classification (Knee Dislocation [KD] grade), and shorter time between injury and surgery would be independent predictors of postoperative motion loss.
Methods:
This was a multicenter retrospective review of 773 MLKIs that occurred between 2011 and 2015. MLKI was defined as a complete tear of two or more ligaments where at least 1 ligament was repaired or constructed. Individuals with a MLKI defined as a complete tear (grade III tear) of two or more ligaments that underwent surgical repair or reconstruction of at least one ligament were included. Cases with loss of motion were identified from review of follow-up documentation or the need for subsequent manipulation under anesthesia and/or lysis of adhesions. A multivariable logistic regression with forward variable entry (p<0.05) was performed to determine if any of the following variables were predictive of motion loss after MLKI: biological sex, KD grade, number of days between injury and surgery, graft type, use of external fixation, if staged procedures were performed, and associated vascular, nerve or tendon injury.
Results:
Of the 773 patients identified in the multicenter review, 562 (72.7%) had complete data for inclusion in the current analyses with 65 (11.6%) having postoperative motion loss (Table 1). Of the 562 patients (76.0% male, age=30.7±12.9 y, BMI=29±6.8 kg/m2), the median number of days between injury and surgery was 59 days (IQR=22-192 days). KD I injuries were the most common (59.6%) with 3.6% having KD II, 14.4% KD III-M, 13.9% KD III-L, 5.3% KD IV, and 3.2% KD V injuries. External fixation was used in 8.2% and 6.8% of the patients underwent staged procedures. Multiple graft combinations were used (all allograft = 37.5%, all autograft = 11.4%, combination = 48.9%, all repairs = 2.1%). Associated nerve injuries occurred in 19.8%, tendon injuries in 15.8%, and vascular injuries in 4.1%. Regression analysis identified decreased time between injury and surgery (OR=0.998, 95%CI[0.997-1.00]) and increased injury severity (KDIII OR=2.31, 95%CI[1.19-4.48], KDV OR=5.42, 95%CI[1.84-15.98]) as independent predictors of motion loss after MLKI (p < 0.001, Nagelkerke R2=0.08).
Conclusions:
While KD grade and time between injury and surgery are predictive of motion loss after MLKI, these variables only explain 8% of the variability in motion loss. This confirms that clinicians should be cognizant that decreased time between injury and surgery and more severe injuries may increase this risk of motion loss; however, postoperative factors not measured in this secondary analysis might more heavily impact motion loss after surgical treatment for MLKI. Future studies are necessary to identify other factors such as pre-operative ROM, the postoperative inflammatory state of the knee, utilization of physical therapy and/or psychosocial factors that might influence motion loss after MLKI reconstruction to better inform postoperative care.
