Hide All

2000 Publications

ELECTROMYOGRAPHIC PREDICTORS OF RESIDUAL WEAKNESS AND IMPAIRED FUNCTION FOLLOWING ANTERIOR CRUCIATE LIGAMENT McHugh MP, Tyler TF, Nicholas SJ, Browne MG, Gleim GWNicholas Institute of Sports Medicine and Athletic Trauma, New York, NY - last modified 2013-02-10 00:00
Transactions of the 46th Annual Meeting of the Orthopaedic Research Society, Orlando Florida, March 12-15, 2000, Abstract #0475.


Introduction: Despite aggressive rehabilitation residual quadriceps weakness is a common complication following anterior cruciate ligament reconstruction. Clinically it is important to identify patients at risk of developing residual weakness.

Surprisingly, preoperative strength is not a good predictor of residual weakness following reconstruction (McHugh). It is possible that the early postoperative loss of quadriceps function is a better predictor of residual weakness than preoperative strength.

Alternatively, EMG measurements may provide a more sensitive measure of preoperative and early postoperative quadriceps function. Decreased amplitude (iEMG) of the surface EMG signal has been used to indicate quadriceps inhibition following ACL reconstruction (Tyler).

However, concommittant effects on the frequency content of the EMG signal were not reported. Decreased median frequency (MF) has been demonstrated in ACL deficient patients and was attributed to fast-twitch fiber atrophy (McNair).

Therefore, the purpose of this study was to examine the contributions of preoperative and early postoperative measures of quadriceps strength, iEMG and MF to predicting residual weakness and impaired function following ACL reconstruction.

Methods: Isometric quadriceps strength measurements were made on 40 patients (28 men, 12 women) prior to ACL reconstruction (pre-op), 5 weeks following surgery (early post-op) and 6 months following surgery. Five weeks post-op was chosen because aggressive quadriceps strengthening began at that time. Six months post-op was chosen because patients were permitted to return to full activities at that time.

Strength tests were performed at 30° knee extension to minimize patellofemoral compression and limit the potential for pain. Following warm-up threee maximum contractions were performed on both the involved and noninvolved sides. Quadriceps surface EMG signals were recorded during all of the strength tests. Pairs of electrodes were placed over the rectus femoris, vastus lateralis and vastus medialis muscles with a 3 cm interelectrode distance.

The iEMG was computed by integrating the rectified EMG signal for each 5 sec contraction. MF was computed from 4096 point Fast Fourier Transforms of of the raw signal for each contraction. Additionally, patients performed a single leg hop test 6 months postoperatively.

Following warm-up, the peak of three single leg hops was recorded for the involved and noninvolved sides. Deficits in strength, iEMG, MF and the hop test were computed as: [(noninvolved-involved)/noninvolved] x 100. Residual weakness was defined as a strength deficit >20% 6 months postoperatively and normal strength was defined as a strength deficit <10%.

Paired T-tests with Bonferroni corrections for multiple comparisons were used to test for differences between the involved and noninvolved sides for each of the four dependent variables. Multiple regression analyses were used to determine which pre-op and early post-op measures were predictive of strength and hop test deficits 6 months postoperatively.

Results: Patients had significant deficits in strength, iEMG and MF preoperatively and at both postoperative time intervals. Hop test deficit at 6 months was 23+/-3% (p<0.01). Surprisingly, iEMG deficit was unrelated to MF deficit at any of the three time intervals (r=-0.16 to -0.22; p=0.91 to 0.18), indicating that these two measures were independent. Thirteen patients (33%) had residual weakness 6 months postoperatively and 16 patients (40%) had normal strength. Pre-op MF deficit was the best indepent predictor of the strength deficit at 6 months (r=0.59, p<0.01). The addition of early post-op strength deficit improved the prediction (R=0.71).

Pre-op MF deficit was 13+/-3% in patients with residual weakness compared to -2+/-3% in patients with normal strength at 6 months (p<0.001). Pre-op iEMG deficit was the best independent predictor of hop test deficit (r=0.51, p<0.01). The addition of pre-op MF deficit improved the prediction (R=0.60). Preoperative strength deficit was not a significant predictor in either multiple regression equation. Deficits in hop test and strength at 6 months were correlated (r=0.62, p<0.01).

Conlusions: Fifty percent of the variability in quadriceps strength deficit 6 months following surgery was explained by pre-op MF deficit (35%) and early post-op strength deficit (15%). Thirty six percent of the variability in hop test deficit 6 months following surgery was expalined by pre-op iEMG deficit (26%) and pre-op MF deficit (10%).

These data indicate that preoperative EMG analyses of quadriceps function are predictive of residual quadriceps weakness and functional impairment following ACL reconstruction. It is important to note that iEMG and MF deficits appeared to be independent measures.

IEMG is sensitive to the extent of motor unit activation and therefore iEMG deficits can be attributed to quadriceps inhibition. MF is sensitive to muscle fiber conduction velocity which is higher in larger fiber diameters.

For example, lower MF is associated with a greater proportion of slow-twitch fibers (small diameter) (Gerdle). Additionally, at a given level of activation, MF is higher at short versus long muscle lengths because fiber diameter is increased when muscles are shortened (Potvin). Therefore lower MF in the present study can be attributed to a general decrease in muscle fiber diameter secondary to atrophy. Lower vastus lateralis MF in ACL deficient patients, demonstrated previously, was attributed to selective fast-twitch fiber atrophy (McNair).

However, it is not possible to determine whether lower MF reflects selective fast-twitch fiber atrophy orsimply non-selective atrophy of all fibers. In conclusion, EMG amplitude and frequency analyses can provide an indication of quadriceps inhibition and atrophy, respectively, following ACL injury and recostruction.

In conclusion, numerous factors, such as, chronicity of injury, knee joint effusion, decreased weightbearing, pain and graft harvesting can affect postoperative quadriceps function. Despite the myriad of potential confounding factors, preoperative EMG measures of quadriceps function were strongly predictive of residual weakness and impaired function following ACL reconstruction.


Gerdle et al. Acta Physiol Scand 134:341-346, 1988

McHugh et al. J Orthop Sports PhysTher 27:407-411, 1998

McNair and Wood. Electromyogr Clin Neurophysiol 33:43-48, 1993

Potvin, J.R. J. Appl. Physiol. 82:144-151, 1997.

Tyler et al. Clin Orthop 357:141-148, 1999.