Skip to main content

Table 4 Characteristics of methods of included studies

From: Which assessments are used to analyze neuromuscular control by electromyography after an anterior cruciate ligament injury to determine readiness to return to sports? A systematic review

Authors and year

Tasks: number of repetitions, duration, frequency

Muscles/legs measured

EMG related outcome measure(s), variables

Direct link to RTS?

Busch et al. (2019) [13]

10 × stair descent, warm-up on treadmill with 5 km/h for 10 min to normalize EMG data, (KOOS, Tegner Activity Score, VAS for pain and general well-being)

VM, VL, BF, ST bilaterally

Normalized root mean squares for each muscle, limb and movement phase (preactivation, weight acceptance, push-off) (%subMVC)

No

Alkjaer et al. (2003) [33]

6 trails of walking across 2 force plates at a speed of 4.5 km/h

VL, VM, ST, BF of injured leg of patients and right leg of healthy controls

Mean amplitudes during weight acceptance (%maxEMG); coactivation between VL and BF (method by Rudolph et al. 2001 [45]) (%maxEMG)

(No) → copers and non-copers

Alkjaer et al. (2002) [34]

15 consecutive forward lunges with recordings from hitting a force plate (rest between trials if wanted)

VL, VM, ST, BF of injured leg of patients and right leg of healthy controls

Peak and mean values of EMG amplitudes (microvolts)

(No) → copers and non-copers

Arnason et al. (2014) [35]

3 trials of Nordic hamstring exercise, 3 trials of TRX hamstring curl exercise; order of exercises was randomized, time

MH, LH bilaterally

Peak normalized muscle activation (%MVIC)

(No) → soccer

Bryant et al. (2009) [36]

ACLD and ACLR: involved limb; healthy controls: both limbs; maximal single limb hop for distance on their involved limb from a standing position. 5 trials with 1 min rest in between trials, landing in a fixed position on the takeoff foot

VL, VM, ST, BF

Timing of the onset of muscle activity relative to IC (onset-IC; ms) and timing of the peak of muscle activity relative to IC (ms)

No

Burland et al. (2020) [37]

Single limb forward hop task, distance of their limb length (tip of the greater trochanter to the tip of the lateral malleolus) → unlimited practice trials, 3 successful trials captured consecutively for each limb (trial = successful when participants landed on force platform and balanced on injured limb for a least 1 s); task performed bilaterally, order of limb testing was randomized

VL bilaterally

Peak muscle activity of the VL: EMG signals from heel strike (defined as 10.0 N) to when PKEM was reached were used for statistical analysis. Mean peak muscle activity obtained from this period of interest across the 3 trials was used

Dynamic EMG data recorded during task were then normalized to the peak muscle activity recorded across all trials. Muscle activity onset times of VL relative to PKEM (EMG onset = time of PKEM–time of EMG “on”) were established using the Teager–Kaiser Energy Operator (EMG onset = median + 3SD)

No

Cordeiro et al. (2015) [38]

3 instep soccer kicks with dominant leg, (KOOS, TSK)

RF, VL, VM, BF, ST

Muscle activation during knee extension phase (% MVC)

(No) → soccer, instep kick

Dashti Rostami et al. (2019) [39]

Single leg vertical drop landing; 3 proper trials

GM, AL; only the injured limb of ACLR and ACLD individuals and the dominant limb of controls were tested

Preparatory and reactive muscle activity and coactivation from 100 ms prior to initial contact to 250 ms after contact; mean and peak activity (%MVIC); coactivation of GM:AL (method by Rudolph et al. 2001 [45])

No

Jordan et al. (2016) [40]

80 s repeated squat jump test (jump test) on a dual force plate system

VL, VM, BF, ST

Normalized EMG amplitudes at takeoff, at the 25-ms interval prelanding, and at postlanding for the ACLR limb (affected limb), contralateral limb, and limbs of the control subjects (control limb), (Asymmetry index, jump height of body center of mass)

(No) → fatigue, downhill skiing

Lessi et al. (2017) [41]

Single leg landing before and after fatigue (fatigue protocol: 10 squats, 2 vertical jumps, 20 steps)

VL, BF, Gmax

EMG average amplitude of activation, expressed as a %peak EMG during landing

No

Oliver et al. (2018) [42]

Single leg jump from a 25-cm tall box, with hands on hips and without gaining momentum; five times with each leg (injured/non-injured)

VM, VL, RF, ST, BF

Mean values per each patient, leg, and muscle were considered in the analysis; muscle latency time over time of each muscle was defined as the time from touchdown to peak amplitude of EMG activity (RMS) in each muscle. RMS was normalized at the maximum activity of the muscles (%MVC)

No

Ortiz et al. (2014) [43]

60-cm double legged and a 40-cm single legged drop jumps to assess bilateral and unilateral landing strategies, respectively

VM, VL, RF, MH, LH measured in the involved leg of women with ACLR and the dominant leg of the control subjects

Rectified normalized electromyographic activity of the quadriceps and hamstrings (amplitude and latency) in %maximum contraction; quadriceps/hamstrings electromyographic co-contraction ratio (values between 0 and 1); time to maximum neuromuscular activation (time-to-peak muscle activation) in seconds for hamstring and quadriceps muscle groups

No

Patras et al. (2009) [44]

10 min running at moderate intensity (20% below the lactate threshold) and 10 min running at high intensity (40% above the lactate threshold) on separate occasions separated by a time span of 48 h and completed within 10–12 days; moderate intensity = at 20% below the lactate threshold; high intensity = at 40% above the lactate threshold

VL, BF bilaterally

Values from 15 strides averaged to calculate the mean peak amplitude during stance for each recording period

No

Patras et al. (2010) [45]

10 min running at moderate intensity and 10 min running at high intensity on separate occasions separated by a time span of 48 h; moderate intensity = at 80% of the lactate threshold; high intensity = at 40% of the difference betweenVO2max and lactate threshold

VL bilaterally

EMG amplitude during stance, over time respectively in microvolts

No

Pincheira et al. (2018) [46]

2 destabilizing platforms (1 for each limb) generated a controlled perturbation at the ankle of each participant (30° of inversion, 10° plantarflexion simultaneously) in a weight bearing condition; time between the release and the stop (impact) of the mechanism was 200 ± 10 ms

VM, ST

Muscle activation onset times (ms)

No

Rudolph et al. (2001) [47]

5 trials of walking and jogging with 1-3 min rest intervals between trials

LH, VL, SO, medial head of the gastrocnemius muscles of both limbs

Peak EMG activity; onset and termination of muscular activation; duration of muscular activity; co-contraction (integrals calculated)

(No) → copers and non-copers

Rudolph et al. (2000) [48]

Single leg hops

LH, VL, SO, medial head of the gastrocnemius muscles of both limbs

Peak EMG activity over 30 ms from either the dynamic or maximum isometric trials was used to normalize the EMG data (%MVIC); muscle timing variables, muscle intensity: integrating the linear envelope of the EMG curves over a weight acceptance interval (defined as the range from 100 ms prior to initial contact to the point of peak knee flexion. Muscle co-contraction: using normalized EMG data, between the VL and LH, and VL and medial gastrocnemius

(No) → copers and non-copers

Rudolph and Snyder-Mackler (2004) [49]

Step up and over a 26 cm high step; 10 trials, 5 each with the right and left leg ascending a 26 cm step (higher than a typical step, provide a more challenging condition), EMG collected from landing limb

LH, VL, SO, medial head of the gastrocnemius muscles of both limbs

Peak EMG activity (%max); onset and termination of muscular activation; duration of muscular activity; co-contraction

(No) → copers and non-copers

Swanik et al. (2004) [50]

Landing from a hop: The subject stood on a 20-cm step, balanced momentarily on test limb, and hopped to target placed 30 cm horizontally; knee perturbation (special knee pertubation device, 100 N force on the posterior aspect of the tibia → anterior displacement of the tibia)

VL, VM, MH, LH

Muscle activity before and after landing from a hop (area of integrated EMG recordings), hamstring latency after joint perturbation (reflexive muscle activity in the hamstrings assessed by measuring the onset time after anterior translation of the tibia)

No

Briem et al. (2016) [51]

3 consecutive maximal hops (triple jump, single-limb crossover hop for distance), 2 practice trials, 1 single maximal test trial; same procedure for each limb. ACLR participants started with non-surgical limb, each matched control participant with matched limb

MH, LH

Peak activation of the normalized signal (%MVIC)

No

Lessi et al. (2018) [52]

Single leg drop vertical jump landing before and after fatigue protocol (fatigue protocol: 10 squats, 2 vertical jumps, 20 steps)

VL, GM, Gmax

Mean amplitude of activation during landing (% of the peak RMS obtained during the landing task)

No

Lustosa et al. (2011) [53]

Walking at self-selected speed on a 3 m-walkway with 2 stable platforms and 1 electromechanical balance board that could apply a sudden perturbation (20° tilt in the frontal plane (medial/lateral) → varus stress in the slightly flexed knee, leading to external rotation of the femur (= common etiology of ACL injury))

VL, BF

Co-contraction pre- and postperturbation between groups and limbs (co-contraction levels in the 250 ms before perturbation and in the 250 ms after perturbation periods), %MVIC; muscular co-contraction calculated

(No) → stratification of included patients (full RTS or limited RTS)

Nyland et al. (2010) [54]

Single leg CMJ performance

Gmax, VM, MH, GC

Mean EMG signal amplitudes (%MVIC); EMG activation duration during propulsion and landing phase (ms)

No

Nyland et al. (2013) [55]

Single leg CMJ performance

Gmax, VM, MH, GC

EMG amplitude comparison during single leg CMJ propulsion (Difference = involved—uninvolved lower extremity) (%MVIC)

No

Nyland et al. (2014) [56]

Single leg hop test for distance

Gmax, VM, MH, GM

Standardized EMG amplitudes during single leg hop for distance propulsion [%MVIC involved lower extremity − %MVIC uninvolved lower extremity); standardized EMG amplitudes during single leg hop for distance landing [%MVIC involved lower extremity − %MVIC uninvolved lower extremity]

No

Boerboom et al. (2001) [57]

Walking at normal, slower, and faster than normal speed

VM, VL, BF, ST, GC medialis, GC lateralis; of injured leg (patients)

Deviations of the normative EMG profiles (individual averaged EMG pattern during gait)

(No) → copers and non-copers

Bulgheroni et al. (1997) [58]

At least 5 trials of walking at natural cadence (112 ± 5.1 steps/min), 20-m distance used to reach steady state of walking

VL, RF, BF, ST

Amplitude of EMG activity, EMG normalized to the maximum recorded signal amplitude during a single walking cycle

No

Gokeler et al. (2010) [59]

Single leg hop test for distance (arms behind back, maintained balance for at least 1 s after landing, 3 maximal trials for each limb; (IKDC, Rolimeter device for laxity testing)

Gmax, BF, ST, SM, VM, VL, RF, MG, LG, SO

Mean onset times (= preparatory activity before landing) of the EMG signals of each muscle

No

Hansen et al. (2017) [60]

Running on weight-supporting treadmill ("anti-gravity", Alter G, respectively) at 16 km/h with 6 different body weight conditions from 50% (half weight) to 100% (full weight-bearing) in random order

SM, SL, MG, LG, MH, LH

Soleus, gastrocnemius and hamstring cluster formed, SPM used to analyze entire time-dependent EMG signal, comparison of injured vs. non-injured leg and left vs. right leg; EMG signal normalized to its MVC value during 100% body weight running trials for each participant

No

Klyne et al. (2012) [61]

Controlled single leg hop on each limb (arms behind back, landing position hold for at least 1-2 s), length of the horizontal distance hopped was equal to the measured length of the lower leg; 3 successful trials

MG

Onset and offset of MG activation relative to take-off, during flight and landing, muscle activity (RMS), 7 temporal variables (ms, %activity)

No [59]

Knoll et al. (2004) [62]

Walking on treadmill at least 10 min at a constant speed of 2 km/h

VL, VM, BF, AL

Linear envelope EMG curve determined by root mean square method and normalized to average of peak EMG signal values of six gait cycles → EMG patterns during % of gait cycle

(No) → pre-operatively and follow-up (6 weeks, 4, 8, 12 months post-surgery)

Kuster et al. (1995) [63]

At least 5 trials of each task to obtain at least 10 cycles of EMG data for ensemble average processing; level walking and downhill walking on dismountable slope (6 m length, -19° gradient)

RF, BF, GC

Peak muscular activity at heel strike, just before heel strike; values normalized to subject's individual peak levels

No

Madhavan and Shields (2011) [64]

Single leg squat maneuver with random/unexpected perturbations at the start of the flexion phase (triggered compensatory reflex activity)

VM obliquus, RF, VL, LH, MH of exercised limb (reconstruced leg of ACLR subjects, pseudorandomly selected limb of healthy controls to counterbalance ACLR limbs)

Normalized long latency responses (= difference between the mean EMG of perturbation trials and the mean EMG of unperturbed trials, divided by the mean EMG of the unperturbed trials) between 50 and 200 ms after the onset of perturbation of quadriceps and hamstrings; peak velocity (cm/s); latency of peak LLR (= time to peak EMG activity between 50–200 ms following the perturbation); mean muscle EMG activity (%MVIC) in the 200 ms prior to perturbation, 50–200 ms after the perturbation, and 200–400 ms post perturbation

No

Ortiz et al. (2008) [65]

5 trials of a single legged 40-cm drop jump: standing initially on both feet on the 40-cm platform and then standing on the jumping leg, and then to drop when ready to do so, maximal-effort vertical jump on landing single legged on the center of the force plate, use of arms allowed for balance; 2 trials of a 20-cm up-down hop task, participant stood facing a 20-cm step and performed 10 consecutive jumps up to and down when ready. The 10 consecutives up and down hops composed 1 trial

GM, GMax, RF, LH, MH; dominant leg in noninjured women and reconstructed leg in ACLR women

Quadriceps/hamstring cocontraction ratios (values between 0 and 1; closer to 1 = excellent co-contraction, closer to 0 = poor co-contraction) and normalized EMG activity of lower extremity muscles (values between 0 and 1; effect sizes respectively)

No

Ortiz et al. (2011) [66]

Side-to-side hopping task that consisted of hopping single legged 10 times consecutively from side to side across 2 lines marked 30 cm apart on 2 individual force plates. The task was designated as a side hopping when the hop was to the opposite side of the stance leg and as crossover hopping when the hop was toward the side of the stance leg

GM, GMax, RF, LH, MH; dominant leg in noninjured women and reconstructed leg in ACLR women

Quadriceps/hamstring cocontraction ratios (values between 0 and 1; closer to 1 = excellent co-contraction, closer to 0 = poor co-contraction) and normalized EMG activity of lower extremity (values between 0 and 1; effect sizes respectively)

No

Patras et al. (2012) [67]

2 10-min treadmill runs on 2 occasions in the lab, 1 at a moderate (80%VO2max) and 1 at a high intensity (85–88% VO2max), EMG recordings at the 3rd, 5th, 7th, and 10th minute of the runs

VL, BF bilaterally: left leg of controls selected for analysis

Peak EMG amplitude during the stance phase

No

Swanik et al. (1999) [68]

4 functional activities: downhill walking (15°, 0.92 m/s), level running (2.08 m/s), and hopping (self-paced) and landing from a jump (20.3 cm)

VL, VM, MH, LH

Integrated EMG (microvolts x ms) normalized to mean amplitude of 3–6 consecutive test repetitions → mean area and peak integrated EMG of a 250 ms-period after ground contact = reactive muscle activity; testing order and leg assessed by random

No

Zebis et al. (2017) [69]

Standardized side cutting maneuver, CJM with the hands placed at the hip (akimbo), and maximal jump height was calculated

VL, BF, ST

EMG preactivity

(No) → single case, risk profile retrospective, pre-/post-surgery and post-intervention

  1. AL = adductor longus muscle; BF = biceps femoris muscle; CMJ = countermovement jump(ing); EMG = electromyography; GC = gastrocnemius muscles; GM = gluteus medius muscle; GMax = gluteus maximus muscle; GRF = ground reaction force; Hz = Hertz; LG = gastrocnemius lateral head; LH = lateral hamstring muscle; MG = gastrocnemius medial head; MH = medial hamstring muscle; ms = milliseconds; PKEM = peak knee extension moment; RF = rectus femoris muscle; SL = soleus lateralis muscle; SM = soleus medialis muscle; SO = soleus muscle; SPM = Statistical Parametric Mapping; ST = semitendinosus muscle; VL = vastus lateralis muscle; VM = vastus medialis muscle; vs. = versus; WA = weight acceptance