Article Text

Download PDFPDF

Gait mechanics after ACL reconstruction: implications for the early onset of knee osteoarthritis
  1. R J Butler1,
  2. K I Minick1,
  3. R Ferber2,
  4. F Underwood1
  1. 1
    Department of Physical Therapy, University of Evansville, Evansville, Indiana, USA
  2. 2
    Faculties of Kinesiology and Nursing, University of Calgary, Calgary, Alberta, Canada
  1. Dr R J Butler, Department of Physical Therapy, 1800 Lincoln Avenue, Evansville, Indiana 47722, USA; rb83{at}


Background: Individuals who sustain a rupture of the anterior cruciate ligament (ACL) are at an increased risk for developing early-onset knee osteoarthritis (OA). The mechanism behind the early onset of the disease is still unknown. Knee OA progression has been previously examined by calculating the internal knee-abduction moment during gait. However, knee-joint moments have not been examined in individuals after ACL reconstruction as a potential mechanism for disease progression in early knee OA.

Objective: To determine if individuals who have undergone ACL reconstruction exhibit altered gait mechanics that may be associated with knee OA progression.

Methods: In total, 17 people who had previously undergone ACL reconstruction were enrolled in the study. A matched control group was recruited for comparison. All participants underwent gait analysis at an intentional walking speed to examine variables previously associated with knee OA progression, primarily the internal peak knee-abduction moment, during gait. One way ANOVAs were performed to examine differences in gait mechanics between the two groups. All joint moments were calculated as internal moments.

Results: The peak knee-abduction moment was increased by 21% in the ACL compared with the control group (p = 0.04). No other differences were seen in frontal plane knee or hip mechanics.

Conclusion: It seems that individuals who have undergone ACL reconstruction exhibit an increased peak knee-abduction moment that may establish a potential mechanism of the earlier onset of knee OA in this population.

Statistics from

The Healthy People 2010 Initiative, introduced by the US government, encourages exercise to maintain healthy lifestyles.1 However, one hurdle to this initiative is the greater risk for injuries that are sustained during exercise activities, which can lead to long-term pain and disability.2 One of the most traumatic musculoskeletal injuries that an individual can incur during exercise is a tear of the anterior cruciate knee ligament (ACL) in the knee, which is associated with long-term joint degeneration in addition to the initial trauma.3 4 It is reported that 1 in 3500 individuals (estimated total of 95 000) in the general population sustain an ACL rupture each year.5 To reduce the risk of this injury, a large number of research studies are currently being conducted to identify ways to reduce the rate of ACL injuries and to improve rehabilitation after the injury.617 However, no current research has focused on the long-term effects of the injury, such as premature degenerative joint disease.18 19 It has been reported that approximately 45% of individuals who sustain an ACL rupture experience premature knee osteoarthritis (OA) within 10 years of the injury.20 Thus, it seems that although continued research on treatment of the initial injury is important, there is also a need for a greater understanding for the mechanism of the early onset and progression of knee OA in people who have undergone ACL reconstruction.

Knee OA is typically considered a disease of “wear and tear”, which is seen most frequently in people aged >65 years, and is diagnosed using radiography and MRI.21 However, current research suggests that an increased number of individuals will exhibit premature knee OA in their 30s and 40s as a result of a knee-joint injury.4 22 It has been estimated that an ACL rupture ages the knee by 30 years.3 Epidemiological studies have reported that individuals who suffer a knee injury are at a >5-fold greater risk of developing knee OA.23 This risk is much greater than the relative risk of 1.7 for increasing body weight, which has been previously strongly correlated with the development of knee OA.23 Thus, it is important to examine the mechanism behind the early onset of knee OA in individuals who have sustained an ACL injury.

Previous researchers have successfully associated variables measured during gait analysis with knee OA progression, and individuals with knee OA have been reported to have altered gait mechanics compared with an asymptomatic population.2429 These mechanics are thought to be related to dynamic frontal plane knee malalignment, which can cause increased displacement between the resultant ground reaction force and the knee joint centre, causing greater frontal plane moments about the knee joint.30 31 The increased frontal plane moment at the knee has been suggested to promote degradation of the medial tibiofemoral compartment of the knee, and has been seen to be a significant predictor of knee OA disease progression.2428 32 It has also been reported that for every 1% increase in the internal abduction moment, the risk of knee OA progression increases 6.46 times, as measured by Kellgren–Lawrence grade and joint-space narrowing.25 33 Although changes in gait mechanics in individuals with diagnosed medial knee OA have established a potential mechanism of disease progression, no research has examined changes in the frontal-plane moments in individual who have undergone ACL reconstruction.

Individuals who have sustained ACL reconstruction are at increased risk for sustaining early onset knee OA; however, the mechanism behind the early development of knee OA has not been established. The use of gait analysis has been established as a successful tool to assess the progression of knee OA in individuals with a diagnosis of the disease, and may be a tool to examine the mechanism of earl-onset knee OA after an ACL rupture. Therefore, the aim of this study was to examine the internal knee moments and knee joint angles in individuals who have undergone ACL reconstruction compared with a group of matched controls with no history of knee injury. In addition, because movements at the knee may be affected by movement further up the kinematic chain at the hip, we wanted to examine if the frontal plane abduction moment and adduction angle at the hip were altered in individuals who have sustained ACL reconstruction, in order to examine more proximal causes for the early onset of knee OA. We hypothesized that the individuals who have undergone ACL reconstruction would have an increased peak knee and hip frontal plane moments compared with the control group.


The study was given ethics approval by the institutional review board. Before entering the study, all subjects read and signed an approved informed consent document.

An a priori power analysis based upon data using similar methods in individuals with knee OA and a meaningful change of 10% revealed that 15 subjects were necessary to examine differences in the peak knee-abduction moment.34 In order to fulfil this criterion, 17 subjects (13 women, 4 men) with a previous ACL reconstruction were recruited for the study (ACL). To be included in the ACL group, it was necessary for all potential subjects to have returned to their previous activity level and to have no current musculoskeletal or neurological conditions that limited their activities of daily living. All of the subjects had undergone a surgical repair at least 1 year before entering the study. For comparison, a control group of individuals who had never sustained a knee injury were recruited, and matched for gender, age (+/− 2 years), activity level (Tegner scale) and body mass index (BMI) (+/−2.5 kg/m2), All individuals in the control group also had no current musculoskeletal or neurological conditions that limited their activities of daily living.

Upon completing the informed consent, each subject was scheduled for gait analysis. Analysis of each individual’s gait was collected unilaterally using a previously published and established marker set (fig 1).35 For the ACL group, the gait data were collected on the injured limb, while gait data on the control group was collected on their dominant leg (defined as the leg they would use to kick a ball the farthest).

Figure 1

Marker set used for gait analysis.

Clusters of four retroreflective markers on thermoplastic shells were attached to the thigh and shank of the involved leg. Pelvic markers were placed on the anterior superior iliac spine, iliac crest and L5-S1 juncture. Tracking markers for the rear foot were placed in the midline of the heel counter of the shoe and just lateral to the midline of the heel counter. To define and establish segmental coordinate systems and inertial parameters, anatomical markers were placed bilaterally on the greater trochanter, and on the medial and lateral epicondyles of the knee, medial and lateral malleoli, first metatarsal head, fifth metatarsal head, and the distal end of the foot. The anatomical markers were removed after a standing calibration trial to allow for a more normal gait pattern that would not be inhibited by the anatomical markers placed on the medial aspect of the leg. All subjects walked in a standardized laboratory shoe that was a neutral running shoe (New Balance 1022, Boston, MA, USA).

Segmental position data, as defined by the retroreflective markers, were recorded as the subject walked at an individualized intentional walking speed (defined as the “speed that they would use when getting mail from their mailbox”) along a walkway where the subject’s speed was monitored by tracking the average velocity of the L5–S1 marker. The subject’s intentional walking speed was maintained within +/− 5% to be considered as an acceptable trial. An eight-camera motion analysis system (VIcontrol, Oxford, UK) was used to record the kinematic data at 60 Hz. A force platform (Bertec Corporation, Columbus, Ohio, USA), placed in the centre of the walkway and flush with the floor, was used to collect ground reaction force data at 1080 Hz as subjects traversed the walkway.

All gait analysis data were analysed from heel strike until toe lift of the gait cycle. This period was normalized to 100% of stance. Joint kinematics and kinetics were calculated after the gait analysis using Visual 3D software (C-Motion, Bethesda, Maryland, USA). Marker trajectories were low-pass filtered at 6 Hz using a second-order recursive Butterworth filter. The force-platform data were filtered using a low-pass recursive 50 Hz Butterworth filter. All kinetic data were normalized to body mass. Joint- moment data were calculated as internal joint moments and were normalized to body mass and height. The calculation of internal joint moments has a different interpretation to that of external joint moments that are commonly reported in the knee OA literature, but fundamentally are equal and have the opposite sign.

The variables of interest for the study focused on frontal-plane motion and moments at the knee and hip along with the vertical ground-reaction forces. Peak adduction and adduction excursion (peak position minus initial position) were examined at the knee and hip. The peak abduction moments at the knee and hip were assessed during the first half of stance. Statistical analyses were performed using a one-way ANOVA (ACL vs control) for all of the variables of interest. SPSS V.14 (SPSS Inc., Chicago, Illinois, USA) was used for all statistical analysis and a significant difference was identified as p<0.05.


No significant differences existed between the groups for age, activity level, body mass index or intentional gait velocity (table 1).

Table 1 Descriptive statistics for the group who had undergone ACL reconstruction and the control group

Individuals in the ACL group exhibited a 21% larger peak knee-abduction moment than the control group, which was significantly different (fig 2). No significant differences were seen for the peak frontal plane moments at the hip (table 2). Additionally, no significant differences were seen for the peak adduction or adduction excursion at the knee or hip (table 2).

Figure 2

Ensemble average traces of the internal knee-abduction moment during stance for the individuals who had sustained an ACL rupture (mean, ACL) and the matched control group (mean+/−standard error, control).

Table 2 Descriptive statistics for the variables of interest during the gait analysis between the group who had undergone ACL reconstruction and the control group.


The aim of this study was to examine the internal joint moments and joint angles at the knee and hip during walking in individuals who have undergone ACL reconstruction compared with a group of controls with no history of knee injury. Every year there are an estimated 95 000 ACL ruptures in the USA. Of this group, it is expected that 42 750 will develop knee OA within 10 years based on previous research estimates.3 However, little is known about the mechanism behind the early development of knee OA in this group. Based on the findings of this study, it can be suggested that one potential mechanism behind the early development of knee OA in this group is that individuals who have sustained ACL reconstruction exhibit an increased peak knee-abduction moment compared with a control group matched for a gender, age, activity level and body mass index. The peak knee-abduction moment has previously been correlated with knee OA progression, and thus it is possible that this increased moment is a potential mechanism causing the early onset of knee OA after ACL rupture.2429

There is relatively little research involving the potential mechanisms behind the early onset of knee OA and therefore, comparison of our data with previous data is difficult. Compared with a group of individuals with diagnosed medial knee OA, the peak knee-abduction moment values for the ACL group in our study are similar (−0.36 Nm/kg.m to −0.38 Nm/kg.m).36 The decreased values seen in our samples are potentially due to the increased activity level and lower BMI of our study group, but are within 1SD of previously published results using a similar marker set. In general, the values for the peak knee-abduction moment seem to be similar to previously published data using the same marker set and protocol.36

What is already known on this topic

  • Individuals who have undergone ACL reconstruction are at risk for the early onset of knee OA.

  • The incidence of a knee injury places an individual at a higher risk for knee OA than an increased BMI.

  • No studies have examined the potential mechanism of early onset knee OA after ACL reconstruction.

What this study adds

  • Individuals who have undergone ACL reconstruction exhibit increased frontal plane knee moments; suggesting a potential mechanism for the early progression of knee OA.

  • Future research should focus on reducing the frontal plane knee moment in individuals after ACL reconstruction to reduce the incidence of early-onset knee OA.

The results of this study suggest that it may be beneficial to address the potential for early knee-joint degeneration after ACL reconstruction during the rehabilitation process with the patient. Current research on reducing the effects of ACL injuries is primarily focused on screening protocols to reduce the incidence of injury along with establishing efficacious treatment protocols to get patients to return to their pre-injury activity level.18 19 In light of the findings of this study, it seems that interventions that can reduce the peak knee-abduction moment may need to be implemented into a patient’s plan of care after ACL rupture to reduce the risk for the early onset of knee OA.

Conservative treatments (eg, wedged foot orthoses and knee bracing) that address malalignment for people with diagnosed knee OA have been shown to reduce symptoms associated with knee OA along with reducing the peak knee-abduction moment.36 37 Although there are few long-term data using these interventions, it seems that hese interventions may offer an effective conservative treatment option to slow the progression of knee OA. Conservative interventions that alter mechanical alignment have previously been reported to be most effective in the early stages of joint degeneration. Thus individuals who do not exhibit radiographic joint degeneration and are asymptomatic but are at risk for disease progression due to additional risk factors (ie knee injury) may benefit most from such interventions.38 Although these conservative interventions that alter mechanical alignment may serve as potential treatment options for individuals who have sustained an ACL rupture, long-term studies examining effectiveness along with issues involving compliance and comfort are required to examine the overall efficacy of such an intervention.

Additionally, conservative mechanical interventions are designed to reduce loading to the compartment that is affected by joint degeneration.. In the knee, joint degeneration is more prevalent in the medial compartment, however individuals still develop knee OA in the lateral compartment. Individuals who present with lateral knee OA tend to exhibit significantly reduced internal peak knee-abduction moments at the knee compared with healthy controls.39 Based on this information, it could be suggested that two of the subjects in our ACL group may be at a greater risk for progressing to knee OA in the lateral compartment (fig 3). Therefore, it may be that an initial gait analysis after ACL surgery may be sensitive enough to detect the compartment of the tibiofemoral joint that is at greater risk for the development of knee OA. This information would be beneficial in guiding the design of any conservative mechanical interventions. Information of this sort is crucial in that if the incorrect side is unloaded then the intervention may actually increase the degenerative disease process and accelerate the early onset of knee OA. More research regarding the relationship between gait mechanics and tibiofemoral compartment knee OA development would be beneficial to assist health care professionals in the prescription of conservative mechanical interventions to reduce an individual’s risk for the early onset of knee OA after ACL reconstruction.

Figure 3

Peak knee-abduction moment values for all subjects in the ACL group (grey bars represent individuals >1 SEM below mean value for controls, white bars represent individuals within 1 SEM of mean value for controls and the hatched bars represent individuals >1 above mean value for controls).

In conclusion, individuals who have sustained ACL reconstruction exhibit an increased peak knee-abduction moment which suggests a greater risk for knee OA development and may provide a mechanism behind the early onset of knee OA seen in this population. In light of this finding, the long term maintenance of joint integrity may want to be considered in the plan of care after the surgical intervention in addition to the desire to return to their previous activity level for individuals with an ACL rupture. Long term research studies examining the efficacy of conservative treatment options aimed at slowing down knee OA progression should be examined in individuals after an ACL rupture to determine if they are effective in reducing the incidence of early knee OA onset.


We thank the Institute for Global Enterprise in Indiana for their financial support of this work. We also thank New Balance for donating the shoes used in this study.



  • Competing interests: None.

  • Patient consent: Obtained.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.