🧬 biology

Within-patient vastus lateralis median frequency asymmetry during isokinetic exercise after ACL reconstruction: a cross-sectional secondary analysis

This cross-sectional secondary analysis reveals that while sample entropy did not differentiate groups, median frequency asymmetry in the vastus lateralis muscle persists in the involved limb of ACL-reconstructed patients compared to healthy controls during isokinetic exercise, suggesting it could complement traditional torque-based assessments.

Original authors: Alex Kubiak, Farzad Haji Boloori, Ali Rezazadeh Shirazi, Zhuoyan Wu

Published 2026-08-19

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Alex Kubiak, Farzad Haji Boloori, Ali Rezazadeh Shirazi, Zhuoyan Wu

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

When a person tears the anterior cruciate ligament, the knee joint is damaged, but the recovery often involves a deeper, invisible struggle. Even after the ligament is surgically repaired and the tissue has healed, the muscles that straighten the leg often remain weak. This is not simply because the muscle has shrunk or lost strength; it is a problem of communication. The brain and spinal cord seem to place a reflexive brake on the motor neurons that control the quadriceps, a phenomenon known as arthrogenic muscle inhibition. This neural shutdown prevents the muscle from firing at full capacity, particularly the fast-twitch fibers needed for powerful movements. While doctors have long relied on measuring the force a patient can push against a machine to judge recovery, these force measurements only capture the final output, not the complex electrical signals the muscles are sending to produce that force. To truly understand if a patient is ready to return to sport, researchers are looking for ways to listen to the electrical chatter of the muscle itself, hoping to find a signature that reveals whether the nervous system is still holding back.

A recent study set out to listen to this electrical chatter in the vastus lateralis, a large muscle on the side of the thigh, using a technique called surface electromyography. The researchers analyzed data from twenty-two individuals, including twelve people who had undergone anterior cruciate ligament reconstruction and ten healthy adults. Each participant performed knee extension exercises on a specialized machine that moves the leg at a constant speed, testing them at two different rates: a slower pace of 90 degrees per second and a faster pace of 180 degrees per second. The goal was to see if the electrical signals from the injured leg looked different from the healthy leg, and if those differences changed depending on how fast the leg was moving. The team focused on two specific ways of analyzing the signal. The first was a measure of complexity, which looks at how predictable or chaotic the electrical pattern is over time. The second was a measure of the signal's pitch, known as median frequency, which indicates how fast the muscle fibers are firing.

The researchers found that the electrical signals did not show a clear difference in complexity between the injured and uninjured legs. While the data hinted that the signals from the injured leg might be slightly less complex, the pattern was not strong enough to be considered a definitive finding with this group of participants. This suggests that the way the muscle fibers fire in terms of their timing and order might not be the most sensitive indicator of the lingering neural inhibition after surgery. However, the study uncovered a very different story when looking at the pitch of the signal. The median frequency of the electrical activity in the injured leg was significantly lower than in the healthy leg. Specifically, the frequency in the injured leg was 11.7 Hertz lower than in the uninjured leg of the same person. This difference was consistent and clear, whereas the healthy control group showed no such gap between their left and right legs.

Importantly, this drop in frequency did not depend on how fast the leg was moving. Whether the machine was testing the leg at the slower speed or the faster speed, the injured leg consistently produced a lower-pitched signal. This finding challenges the idea that a faster test speed might be necessary to reveal the problem; the deficit was present and detectable at both speeds. The lower frequency suggests that the injured leg is failing to recruit the high-speed, powerful muscle fibers that are needed for explosive movement, leaving the muscle to rely more on slower, less efficient fibers. While the study was too small to use these electrical patterns to reliably distinguish between patients and healthy people in a broad sense, the clear difference within each patient's own body offers a promising new tool. It suggests that listening to the pitch of the muscle's electrical signal could provide a valuable addition to the standard force tests used by doctors, helping to identify when a patient's nervous system is still suppressing their muscle strength long after the surgery has healed.

Technical Summary: Within-patient Vastus Lateralis Median Frequency Asymmetry During Isokinetic Exercise After ACL Reconstruction

Problem Statement
Persistent quadriceps weakness following anterior cruciate ligament reconstruction (ACLR) is increasingly recognized as a neurophysiological deficit driven by arthrogenic muscle inhibition (AMI), rather than a purely mechanical issue. AMI reflexively suppresses the recruitment of high-threshold, fast-twitch motor units, a phenomenon that can persist long after structural healing. Current clinical assessment relies heavily on isokinetic torque and the Limb Symmetry Index (LSI). However, torque measures net mechanical output without revealing the underlying recruitment strategy of the motor unit pool. Furthermore, LSI thresholds (e.g., 90%) often underestimate functional asymmetry and poorly predict re-injury. While surface electromyography (sEMG) offers a window into motor unit firing patterns, the application of complexity and spectral analysis to the vastus lateralis (VL) during the specific context of isokinetic dynamometry—the standard modality for return-to-sport decisions—remains limited. Specifically, it is unclear whether signal complexity (Sample Entropy) or spectral content (Median Frequency) reveals deficits in the involved limb compared to the uninvolved limb and healthy controls, and whether testing speed (90°/s vs. 180°/s) moderates these deficits.

Methodology
This study is a cross-sectional, repeated-measures secondary analysis of a publicly available, de-identified sEMG dataset (Dryad repository).

Participants: 22 individuals (12 post-ACLR patients, 10 healthy controls).
Protocol: Participants performed isokinetic knee extensions at two speeds (90°/s and 180°/s).
Data Acquisition: Bipolar surface EMG signals were recorded from the vastus lateralis (VL) of both legs using a Delsys sensor. The dataset provided 88 trial-level observations.
Signal Processing:
- Signals were band-pass filtered (20–450 Hz), trimmed to the middle 60% of the contraction to ensure stationarity, resampled to 1000 Hz, and z-scored.
- Sample Entropy (SampEn): Computed using the Richman–Moorman algorithm ( $m=2, r=0.20 \times SD$ ) to quantify time-domain signal complexity. Due to non-normality, values were log(1+x) transformed for parametric modeling.
- Median Frequency (MDF): Calculated via Welch's method (512-sample window, 50% overlap) as the frequency below which 50% of spectral power lies.
Statistical Analysis:
- Primary Analysis: A linear mixed-effects model tested the interaction between cohort (patient vs. healthy) and limb (involved vs. uninvolved) on log-transformed SampEn, with speed as a covariate and participant as a random intercept.
- Secondary Analysis: Wilcoxon signed-rank tests assessed within-patient limb asymmetry for MDF and SampEn.
- Exploratory Analysis: Leave-one-out cross-validated logistic regression attempted to classify cohorts based on mean features, though this was noted as underpowered.
- Sensitivity: Multiple pre-specified and post-hoc sensitivity analyses were conducted to test robustness against parameter variations and outliers.

Key Results

Sample Entropy (Complexity): The primary hypothesis—that VL SampEn would be lower in the involved limb of ACLR patients compared to controls and the uninvolved limb—was not supported. The cohort $\times$ limb interaction was non-significant ( $\beta = 0.092$ , 95% CI -0.009 to 0.193; $p = 0.074$ ). Within-patient asymmetry in SampEn was also non-significant ( $p = 0.169$ ).
Median Frequency (Spectral Content): In contrast, a significant within-patient asymmetry was observed in MDF. The involved limb of ACLR patients exhibited a significantly lower MDF than their uninvolved limb (median difference $\Delta = -11.7$ Hz; $p < 0.001$ ). No such asymmetry existed in healthy controls ( $p = 0.599$ ).
Velocity Dependence: The hypothesis that MDF deficits would be speed-dependent (unmasked at specific velocities) was not supported. The MDF asymmetry was velocity-invariant, showing similar deficits at both 90°/s and 180°/s.
Classification: Leave-one-out classification of cohort based on these features performed at or below chance levels, attributed to the small sample size ( $n=22$ ) and the nature of the signal being within-subject rather than between-group.

Significance and Claims
The paper modestly claims that VL median frequency asymmetry serves as a detectable, velocity-invariant marker of neuromuscular deficits in post-ACLR patients during isokinetic testing.

Clinical Utility: The authors suggest that MDF asymmetry could complement traditional torque-based LSI as a "low-overhead" sEMG add-on for return-to-sport assessments. It highlights that the involved limb suffers from a reduction in high-frequency content (indicative of suppressed fast-twitch motor units) without a parallel loss of temporal signal complexity.
Mechanistic Insight: The dissociation between significant MDF reduction and non-significant SampEn reduction suggests that while the motor unit pool shifts toward slower conduction velocities, the temporal complexity of the composite firing pattern remains preserved at the timescales assessed by SampEn.
Limitations and Future Needs: The authors explicitly state that these findings are exploratory. The single-muscle (VL only) recording limits generalizability to other quadriceps compartments. The small sample size and lack of demographic/clinical covariates (e.g., time since surgery, graft type) prevent definitive clinical recommendations. The paper concludes that replication in a larger, demographically characterized cohort with multi-muscle instrumentation is required before SampEn or MDF can be validated as biomarkers for AMI.

Technical Summary: Within-patient Vastus Lateralis Median Frequency Asymmetry During Isokinetic Exercise After ACL Reconstruction

More like this