Lower Extremity Muscle Activation Onset Times During the Transition from Double-Leg Stance

Lower extremity muscle activation onset times during the transition from double-leg stance to single-leg stance in anterior cruciate ligament reconstructed subjects

Bart Dingenena (corresponding author), Luc Janssensb, Steven Claesc, Johan Bellemansd, Filip F. Staese

a KU Leuven Musculoskeletal Rehabilitation Research Group, Department of Rehabilitation Sciences, Faculty of Kinesiology and Rehabilitation Sciences, Belgium. Tervuursevest 101 b1501, 3001 Heverlee, Belgium. E-mail:

b KU Leuven Department of Electrical Engineering, Faculty of Engineering Technology Services, Andreas Vesaliusstraat 13, 3000 Leuven, Belgium. E-mail:

b KU Leuven Cardiovascular and Respiratory Rehabilitation Research Group, Department of Rehabilitation Sciences, Faculty of Kinesiology and Rehabilitation Sciences, Belgium. Tervuursevest 101 b1501, 3001 Heverlee, Belgium. E-mail:

cDepartment of Orthopedic Surgery, AZ Herentals Hospital, Herentals, Belgium. Nederrij 133 2200 Herentals, Belgium. E-mail:

dDepartment of Orthopedics, University Hospitals Leuven, Campus Pellenberg, Leuven, Belgium. Weligerveld 1, 3212 Pellenberg, Belgium. E-mail:

eKU Leuven Musculoskeletal Rehabilitation Research Group, Department of Rehabilitation Sciences, Faculty of Kinesiology and Rehabilitation Sciences, Belgium. Tervuursevest 101 b1501, 3001 Heverlee, Belgium. E-mail:

Word count abstract: 236

Word count main text: 3897

1

ABSTRACT

Background:Previous studies mainly focused on muscles at the operated knee after anterior cruciate ligament reconstruction, less on muscles around other joints of the operated and non-operated leg. The aim of this study was to investigate muscle activation onset times during the transition from double-leg stance to single-leg stance in anterior cruciate ligament reconstructed subjects.

Methods:Lower extremity muscle activation onset timesof both legs of 20 fully returned to sport anterior cruciate ligament reconstructed subjects and 20 non-injured control subjects were measured during the transition from double-leg stance to single-leg stance in eyes open and eyes closed conditions. Analysis of covariance (ANCOVA) was used to evaluate differences between groups and differences between legs within both groups, while controlling for peak center of pressure velocity.

Findings:Significantly delayed muscle activation onset timeswere found in the anterior cruciate ligament reconstructed group compared to the control group for gluteus maximus, gluteus medius, vastus medialis obliquus, medial hamstrings, lateral hamstrings and gastrocnemius in both eyes open and eyes closed conditions (P<.05). Within the anterior cruciate ligament reconstructed group, no significant differentmuscle activation onset timeswere found between the operated andnon-operated leg(P.05).

Interpretation:Despitecompletion of rehabilitation and full return to sport, the anterior cruciate ligament reconstructed group showed neuromuscular control deficits that were not limited to the operated knee joint.Clinicians should focus on relearning multi-segmental anticipatory neuromuscular control strategies after anterior cruciate ligament reconstruction.

Key Words:anterior cruciate ligament reconstruction, neuromuscular control, lower extremity, rehabilitation, injury prevention

1. INTRODUCTION

The main goal of an anterior cruciate ligament (ACL) reconstruction is to restore mechanical knee joint stability. However, the restoration of mechanical knee stability after ACL reconstructiondoes not automatically imply a return to normal neuromuscular control.1,56Developing optimal neuromuscular control strategies during rehabilitation after ACL reconstructionis considered to be essential to facilitate successfulshort- andlong-term outcomes.18

Alterations in neuromuscular control after a knee joint injury may not only occur at the involved joint, but also at proximal and distal joints.42Nevertheless, most studies measuring muscle activity after ACL reconstructiononly focused on muscles surrounding the operated knee joint.8,13,15,17,39,50,56Only a scarcity of research focused on neuromuscular control of the whole lower extremity (including hip, knee and ankle muscles) after ACL reconstruction.27,37,38Furthermore, it is difficult to draw firm conclusions due to the differences in tasks, graft selection, time after ACL reconstruction and outcome measurementsin these studies. Bilateral neuromuscular control deficits may exist after unilateral ACL reconstruction.8,37,38,56Caution is therefore warranted when conclusions are based on the comparison of the operated leg with the non-operated leg after ACL reconstruction.The rationale to investigate muscle activation patterns not only of muscles surrounding the operated knee joint, but also of the adjacent joints and the contralateral leg, is supported by the growing evidence demonstratingthe crucial role of central nervous system (CNS) adaptations after ACL injury and reconstruction.28

The transition from double-leg stance to single-leg stance has previously been used to assess neuromuscular control deficits in subjects with a variety of musculoskeletal impairments of the lower quadrant.20,31,32,34,48,51The advantage of this transition task is that the sensorimotor system can be experimentally challenged in specific ways when eliminating or altering for example the visual input, changing movement speed or decreasing the movement preparedness. In addition, this transition task allows measuring subjects across different stages of a rehabilitation process. An anticipatory muscle activity during this task can be interpreted as a strategy selected by the CNS to prepare the lower extremity for the upcoming postural perturbation created by the transition task, while slower muscle recruitment may decrease the ability to effectively stabilize the lower extremity joints.10,20

Dingenen et al20were the first to focus on muscle activation patterns of the whole lower extremity during this transition task in ACL-deficient subjects,tested prior to ACL reconstruction surgery.Delayed muscle activation onset times were not only foundat thekneebut also at thehip and ankle muscles.Furthermore, no consistent differences between the injured and non-injured leg of the ACL injured group were reported.20These bilateral and multi-segmental neuromuscular deficits support the contribution of alterations in the organization of the CNS after ACL injury.28,33 However, it remains unclear whether these altered muscle activation patterns are still present after ACL reconstruction. These insights might allow clinicians to broaden their vision on neuromuscular alterations after ACL injury and reconstruction, which can assist improving rehabilitation approaches.55

The aim of the present study was therefore to investigatemuscle activation onset times of knee, hip and ankle muscles of both legs in ACLR and non-injured control subjects. First, it was hypothesized that ACLR subjects would show delayed muscle activation onset times compared to non-injured control subjects, not only around the operated knee joint, but also around the hip and ankle. Second, it was hypothesized that no significant differences would be found between the operated and non-operated leg of the ACLR group.

2.METHODS

2.1 Subjects

The same 40 subjects of the study of Dingenen et al19 were tested.Before participating in the study, all subjects read and signed an informed consent form, which was approved by the local ethical committee. The ACLR group (n=20) included subjects with a history of oneACL reconstruction at least 9 months before the testing, who completed rehabilitation and fully returned to their pre-injury competitive sport involving pivoting, jumping and/or cutting activities.The time after ACL reconstruction was mean (SD) = 22.98(13.97) months (range: 9.60-54.70 months). Subjects reporting ankle, knee, hip or low back pain during athletic activities or previous lower extremity surgery (except the primary ACL reconstruction) on a custom-made self-report questionnaire were excluded. All ACL injuries were caused by a non-contact injury mechanism and treated with an ipsilateral hamstring autograft. From all ACLR subjects, an equal number of subjectshad undergone surgery on the dominant (n=10)or non-dominant leg (n=10). The dominant leg was defined as the preferred leg to kick a ball. The control group (n=20) included subjects with no history of ankle, knee, hip or low back injury.24 The activity level of all subjects was evaluated with the Tegner activity rating scale.11,35Subjects younger than 18 and older than 55 years old, and with the following conditions were also excluded: chronic ankle instability (subjects with a history of at least 2 ankle sprains at the same ankle in the past 2 years and reporting a subjective feeling of “giving way” of the ankle),24 Parkinson, multiple sclerosis, cerebrovascular accident, peripheral neuropathies, circulation disorders, serious joint disorders (rheuma, osteoarthritis, etc.).

2.2 Data collection

Ground reaction forces and muscle activity of 10 lower extremity muscles were measured simultaneously and synchronously during the transition from double-leg stance to single-leg stance.20 Ground reaction forces were measured by a force plate (Bertec Corporation®) at 500 Hz using a Micro 1401 data-acquisition system and Spike2 software (Cambridge Electronic Design, UK) and low pass filtered with a cut-off frequency of 5 Hz. Surface electromyography (EMG) (Noraxon Myosystem 1400®) signals were measured at 2000 Hz using MyoResearch 2.0 (Noraxon USA, Inc., Scottsdale, AZ) and Spike2 software. The gluteus maximus, gluteus medius, tensor fascia latae, vastus lateralis, vastus medialis obliquus, hamstrings medial, hamstrings lateral, tibialis anterior, peroneus longus and gastrocnemius were measured unilaterally on the upcoming single-leg stance leg.20 Placement of the electrodes was based on the instructions of Basmajian and De Luca.5 One reference electrode was put on the anteromedial side of the tibia. The silver-silver chloride, pre-gelled bipolar surface EMG electrodes (Medicotest Inc., Rolling Meadows, IL) were placed over the muscle belly and aligned with the longitudinal axis of the muscle, with a center-to-center distance of 10 mm. The minimum distance between electrode pairs was set at 3 cm to reduce the possibility of cross-talk between neighbouring muscles. The skin area where electrodes were applied was shaved and gently cleaned with 70% isopropyl alcohol to reduce the impedance. The EMG signals were stored on a PC for further analysis. The position of the electrodes was confirmed by isolated manual muscle tests.

2.3 Procedure

The procedure used in this study is based onprevious studies.19-24 Subjects were asked to stand barefoot on a force plate with the feet separated by the width of the hips and the arms hanging loosely at the side. They performed a transition task from double-leg stance (13seconds) to single-leg stance (13seconds). Both legs of both groups were tested (Fig. 1).The leg that was tested first was assigned randomly.The position of the feet during double-leg stance was indicated on a paper lying on the force plate to ensure that subjects returned to the same starting position after each trial. Subjects were instructed to lift one leg on the command of the examiner toward approximately 60° of hip flexion within 1 second, using a metronome as a reference. For all subjects, an equal number of fake trials (shifting the weight to the non-tested leg) were randomly included to avoid preparedness. The transition task from double-leg stance to single-leg stance was tested with eyes open and with eyes closed. Both conditions were repeated 3 times in an alternating order. In the eyes open tests, subjects were instructed to keep their gaze straight ahead facing a white wall. The eyes closed condition was included as one may hypothesize to find more apparent differences between groups because of the increased reliance on visual information during postural control tasks after ACL injury and reconstruction.28All subjects were allowed to familiarize with the test conditions and movement speed by performing 2 practice trials of each condition before the actual measurements. Between conditions, subjects could rest to avoid fatigue. Afterwards, body height and weight were measured.

2.4Data analysis

Force plate and EMG signals of each trial of each condition were exported from Spike2 into LabVIEW (National Instruments Corp, Austin, Tex). The raw EMG data were first rectified and then filtered by 6th order Butterworth low-pass filter with a cut-off frequency of 45 Hz to generate the envelope of the EMG. A fixed window of 100 ms before the stance transition (during the double-leg stance phase) was compared with a moving window of the same length along the measurement.20,22,23 An increase of more than 2 standard deviations on top of the average baseline activity was identified as the onset of muscle activity in reaction to the transition.30 The onset of muscle activity determined by the algorithm was checked against the muscle activity onset identified visually.30

When moving from double-leg stance to single-leg stance, the center of pressure (COP) first moves toward the opposite direction of the final standing leg with a certain maximum deviation (contralateral push-off movement).24 This is followed by a movement toward the single standing leg to finally finish on one leg (single-leg stance phase).24Onset of muscle activity was calculated with respect to the mediolateral moment of force (Mx).20,22,23First, the average Mx was calculated for the double-leg stance phase. The time point at which the Mx crossed that average after the contralateral push-off movement during the weight transfer to the single standing leg was defined as the Mx onset. An increase of muscle activity before Mx onset is reflected by a negative value for the onset of muscle activity. A positive value indicates an increase of muscle activity after Mx onset.20,22,23

The peak COP velocity when moving to the upcoming single stance leg was calculated as the maximum first derivative in time of the COP displacement between the initiation of the contralateral push-off movement and the time needed to reach a new stability point during the single-leg stance phase.24

2.5 Statistical analysis

2.5.1 Subject characteristics

All subject characteristics, except activity level, showed a normal distribution (Shapiro-Wilk). Comparisons between groups for age, height, weight and body mass indexwere done with independent t-tests and for gender with the chi-squared test. Activity level was compared with the Mann-Whitney U test.

2.5.2 Muscle activation onset times

The average of 3 trials of all muscle activation onset times was calculated for each condition of each subject. First, the operated leg of the ACLR group (dominant/non-dominant) was matched with the same dominant/non-dominant leg of the control group.19-21Analysis of Covariance (ANCOVA) was conducted to evaluate the differences between the ACLR and control group on muscle activation onset times for eyes open and eyes closed, while controlling for peak COP velocity. Second, differences between legs within both groups (dominant versus non-dominant leg for the control group, and operated versus non-operated leg for the ACLR group) were tested with ANCOVA for eyes open and eyes closed separately, while controlling for peak COP velocity.

2.5.3 Peak COP velocity

Peak COP velocity was normally distributed for eyes open, but not for eyes closed after matching the operated leg of the ACLR group with the same dominant/non-dominant leg of the control group (Shapiro-Wilk). An independent t-test and Mann-Whitney U test were used to compare peak COP velocity between groups for respectively eyes open and eyes closed.

Within both groups, peak COP velocity was normally distributed, except for the eyes closed condition in the ACLR group. Dependent t-tests were used to compare the dominant and non-dominant legs within the control group for eyes open and eyes closed, and the operated and non-operated leg within the ACLR group for eyes open. Within the ACLR group with eyes closed, the Wilcoxon Signed Rank test was used to compare both legs. Statistical significance was set at P<.05 for all analyses. All statistical analyses were performed using IBM SPSS Statistics for Windows, Version 19.0, Armonk, NY: IBM Corp., USA.

1. RESULTS

3.1 Subject characteristics

No significant differences between groups were found for age, gender, height,weight, body mass index and activity level (P>.05) (Table 1).19

3.2Muscle activation onset times

3.2.1 Differences between groups

In the eyes open condition, significantly delayed muscle activation onset times of gluteus maximus (P=.001), gluteus medius (P=.005), vastus medialis obliquus (P=.020),medial hamstrings (P=.048), lateral hamstrings (P=.043) and gastrocnemius (P=.036) were found in the ACLR group compared to the control group (Fig.2A).

In the eyes closed condition, significantly delayed muscle activation onset times gluteus maximus (P=.001), gluteus medius (P=.043), vastus medialis obliquus (P=.027), medial hamstrings (P=.002), lateral hamstrings (P=.032) and gastrocnemius (P=.026) were found in the ACLR group compared to the control group (Fig.2B).

3.2.2 Differences between legs within each group

No significant differentmuscle activation onset times were found between the dominant and non-dominant leg in the control group (Fig. 3AB) and between the operated leg and the non-operated leg in the ACLR group for both the eyes open and eyes closed condition (P>.05) (Fig. 3CD).

3.3 Peak COP velocity

The peak COP velocity was not significantly different between groups (Table 2) nor within groups (Table 3) (P>0.05).