Weakness and wasting of the quadriceps femoris muscle is a common finding in patients after knee trauma and knee immobilization. Particularly in ACL-deficient knees it can persist over years and may be a decisive factor on the clinical outcome of ACL ruptures treated conservatively (6,21) or operatively (4,12,34). Often the measurable muscle weakness is greater than can be expected from muscle atrophy by disuse alone; the inability to fully activate the muscle voluntarily is considered to be responsible for the lack of correlation between muscle weakness and muscle atrophy (11,21,31,32). It has been shown that the deficiency of voluntary muscle activation occurs after meniscectomy (27), experimental knee infusion (9,35), painless chronic effusion (18) and extensive knee injuries (15,24). Only a few studies investigated voluntary quadriceps muscle activation after isolated ACL rupture (16,29). These studies revealed a minor deficit of quadriceps muscle activation in patients with ACL deficiency.
However, recent studies using more sensitive twitch-interpolation techniques have shown that even normal subjects may have some deficit in muscle activation (10). Therefore, it remains questionable whether the small amount of activation deficit described previously is attributable to an isolated ACL-rupture at all. With respect to that, the present study was designed to confirm or reject the hypothesis that isolated ACL deficiency leads to a lack of maximal voluntary quadriceps muscle activation in comparison with an age-, gender- and activity-matched control group.
Twenty-two male patients suffering from ACL deficiency and 19 healthy, male volunteers were included into the study. The patient group comprised subjects ranging in age from 14.2 to 45.5 (median 26.3) yr who had had a unilateral, isolated ACL tear. All diagnoses were confirmed by arthroscopy following the experimental measurements. Patients exhibiting any type of knee-joint damage beside the ACL rupture or presenting a knee effusion were excluded from this study. All patients were dissatisfied with the condition of their injured knee, complaining of instability at the least under sports activity. The trauma-to-measurement interval was 119 d (median) with a range of 31 d to 22 months.
The control group consisted of healthy male volunteers and was matched to the patient group regarding age (range 15.3 to 48.2, median 28.2 yr) and activity level. Except for arthroscopy, the experimental setup and testing procedure was identical to the patient group. All experimental procedures were approved by the local ethics committee and all subjects gave consent after being informed of the testing procedure and the nature of the study.
The quantification of the maximal voluntary muscle activation is based on the principle that if all the muscle fibers are fully activated voluntarily, a superimposed external muscle stimulation (twitch) will not produce any additional muscle force. However, if the muscle cannot be fully contracted voluntarily, additional force is generated by superimposed muscle stimulation. Providing over 25% of the maximal voluntary contraction, the additional force generated by superimposing electrical stimulation has a linear relationship to the voluntarily elicited initial force (5,7,25). This linearity is used to extrapolate the true maximal contraction at imaginary 100% muscle activation to calculate the extent of voluntary activation as a percentage.
Electrical stimulation was performed using a constant current stimulator (Dantec Counterpoint MK II, Skovlunde, Denmark) applying single, square-wave stimuli with 100 mA amplitude and 500 μs duration. Self-written software was used for data acquisition and for highly sensitive, automated twitch detection as described by Hales and Gandevia (13). The sensitivity and reliability of this method has been investigated (1,13,25).
All patients were seated in an upright position on a purpose-built chair with their hips and knees flexed to 90°. The lower legs were fixed using a nonextendible leather strap to the lever arm of a force-measuring device based on an analog strain gauge. For electrical muscle stimulation, aluminum plate electrodes of 5 × 10 cm, covered with thin, saline-soaked sponges for better skin contact, were strapped to the middle of the quadriceps muscle (cathode) and to the distal quadriceps muscle 10 cm above the patella (anode).
First, subjects were instructed under intense encouragement to fully extend their knee (5 s) for determination of the maximum voluntary contraction and for maximum potentiation of the twitch response (33). Immediately after twitch potentiation, the subjects performed isometric contractions with 90, 75, 50, 25, and 100% of their maximal voluntary contraction by matching the visualized torque level to the line on the monitor of the desired torque level. When the torque was kept stable three single stimuli were applied to the muscle.
Nonparametric paired sample tests (Wilcoxon signed ranks) were performed to compare the injured with the uninjured side of patients and the left with the right leg of controls. For comparison between the control group and the patient group, independent two sample tests (Mann-Whitney-U) were used. In correlation analysis the Spearman rank coefficient of correlation was calculated. The level of statistical significance was set to an alpha-level of P < 0.05 referring to two-tailed tests. SPSS (Chicago, IL) statistical software was used for all calculations. Unless specified otherwise, the results are given as mean ± SEM.
Maximal voluntary contraction of the quadriceps muscle.
Figure 1A shows the cumulative distribution function of the maximal voluntary contraction of the quadriceps muscle in the patient group and in the control. Maximal voluntary contraction values of the injured legs in the patient group are clearly left shifted, demonstrating the considerably lower maximal voluntary contraction (153 ± 9.6, range 82–246 Nm) compared with the legs of the control group (216 ± 9.5, range 120–384 Nm) (P < 0.001). A slight left shift of the dashed line (patients, uninjured side) with respect to the dotted line (controls) in Figure 1A indicates that the quadriceps muscles of the uninjured side in the patient group was slightly weaker (189 ± 12.4, range 85–365 Nm) compared with the controls not attaining significance (P = 0.057). The differences of the maximal voluntary contraction between the injured and the uninjured side of the patients were significant (P < 0.001). In the control group, the maximal voluntary contraction of the left quadriceps (217 ± 13.7, range 134–384 Nm) was equal to one of the right quadriceps (215 ± 13.6, range 120–362 Nm) (P = 0.31). Within the evaluated time range, no correlation was found between voluntary maximal contraction and time after trauma on the injured (correlation coefficient, −0.23, P = 0.301) and uninjured side (correlation coefficient, −0.231, P = 0.301).
Voluntary activation of the quadriceps muscle.
Figure 2 shows a typical linear relation between initial knee extension torque and the additional twitch-torque induced by superimposing electrical stimuli in a normal subject (A) and in a patient with ACL deficiency (B). The healthy subject achieved a maximal voluntary contraction of 213 Nm with the left leg and only a small additional torque of 2.8 Nm could be elicited by electrical stimulation at maximal voluntary contraction. The superimposed twitch torque increased linearly with decreasing initial torque, yielding an extrapolated twitch torque of 30 Nm in the resting muscle and resulting in a calculated voluntary activation of 94%. The right quadriceps muscle was marginally weaker (203 Nm) while presenting the same amount of voluntary activation. In contrast, the patient achieved a lower maximal voluntary contraction of 105 Nm with a higher twitch-torque of 5 Nm at maximal voluntary contraction, resulting in a voluntary activation of 79% in the injured leg (Fig. 1B). The quadriceps muscle of the uninjured side was stronger (153 Nm), although showing almost the same activation deficit (voluntary activation 80%) compared with the injured side. Despite the similarity of the voluntary activation on both sides of the patient, superimposed twitch amplitudes on the injured side were lower for comparable initial torque.
An evaluation of all subjects showed that the ability to fully activate the quadriceps muscle was significantly decreased in the patients compared with the control group. The differences were significant for the injured side (P = 0.026) and the uninjured side (P = 0.031). The patients’ cumulative distribution functions of the voluntary activation (Fig. 1B) demonstrate a slight left shift compared with that in the controls (91.1 ± 0.8, range 79.4–99.2%; dotted line) and the shapes are altered exhibiting a long tail toward low voluntary-activation values. Furthermore, the cumulative distribution functions show that the voluntary activation of the quadriceps muscles of the injured side (83.9 ± 2.3, range 57–95.2%) and the uninjured side (84.7 ± 2.2, range 60.8–98.8%) was equal (P = 0.453). Sixty-four percent of the patients presented a voluntary activation under the lower bound of the 95% confidence interval of mean of the controls (CI 89.0–93.7%), and 23% of the patients presented a voluntary activation of less than 80% in the injured leg. There was no correlation between voluntary activation on the injured side and time after trauma (correlation coefficient −0.07, P = 0.76). The uninjured side presented a tendency of increased voluntary activation with prolonged history of injury without attaining statistical significance (correlation coefficient 0.330, P = 0.133).
True maximal contraction of the quadriceps muscle.
The true maximal contraction of the patients’ quadriceps muscles of the injured leg (181 ± 9.1 Nm) was higher than the measured maximal voluntary contraction (P < 0.001). However, this true maximal contraction was still lower compared with one of the control group (238 ± 10.0 Nm) (P = 0.003). Figure 1C shows additionally that the cumulative distribution function of the true maximal contraction on the uninjured side (222 ± 12.7 Nm) matches almost the cumulative distribution function of the controls; the minor differences do not attain statistical significance (P = 0.360). No correlation was found between true maximal contraction and time after trauma on the injured (correlation coefficient −0.247, P = 0.268) and uninjured side (correlation coefficient −0.176, P = 0.434).
The present study shows that patients with symptomatic, isolated, subacute and chronic ACL deficiency have only a moderate but statistically significant deficiency of voluntary quadriceps activation compared with an age-, gender-, and activity-matched healthy control group. Furthermore, the inability to fully activate the quadriceps muscle voluntarily affects the injured and the uninjured side to the same extent.
A voluntary-activation deficit in patients with subacute ACL rupture has also been described by Snyder-Mackler et al. (29). In this study a semiquantitative variant of the twitch interpolation technique was applied to 12 patients with subacute and eight patients with chronic ACL deficiency. Voluntary-activation deficits were found only in 9 of the 12 patients of the subacute group, implying that the voluntary-activation deficit may vanish with time after injury. In the present study, however, no correlation was found between voluntary-activation deficit, maximal voluntary contraction, true maximal contraction, and time after injury on the injured side. This might be explained in part by the relatively narrow time range of up to 22 months after trauma of our patients compared with that of the chronic group of Snyder-Mackler et al. with a time-to-injury interval of up to 5 yr. On the other hand, the voluntary activation of the uninjured side showed a tendency of improvement over the time. This indicates that a voluntary-activation deficit might be reversible in general, but ongoing irritations by knee-joint pathology prevent recovery in ACL-deficient knees.
A bilateral quantification of the voluntary-activation deficit in 10 patients with isolated ACL rupture was performed before and after physiotherapy by Hurley et al. (16). Assuming 100% voluntary muscle activation as “normal,” they found a voluntary activation of approximately 92% for the patients on both sides, indicating a bilateral deficit of about 8%. The presence of such a small bilateral deficit on average was confirmed in the present study although our voluntary-activation values were lower than those of Hurley et al. (16). This difference in the two studies might be explained by the different twitch interpolation methods. The technique for twitch detection without adaptive signal amplification by Hurley et al. is less sensitive than the technique by Hales and Gandevia (13) used in the present study. In accordance to our results, the presence of incomplete voluntary muscle activation in normal subjects has been described in several other studies (1,10) using the highly sensitive variant of twitch-interpolation technique similar to the one in the present study. Consequently, it is necessary to evaluate eventual voluntary-activation deficits only with respect to a proper control group. As shown in Figure 1B there are voluntary-activation deficits of ACL patients even when compared with those of controls. However, the vast majority of patients exhibited a voluntary-activation distribution almost identical to the control group, whereas about a quarter of the patients are responsible for a long distribution tail toward low voluntary-activation values, the tail being in turn the reason for the average decline of voluntary activation in the ACL-deficient group.
The neurophysiological mechanisms of voluntary muscle activation deficits are not yet fully understood. There is evidence that knee joint receptors contribute to the regulation of muscle tone in posture and movement via influence on the gamma muscle loop to regulate joint stiffness and joint stability (17). The existence of inhibition of spinal neurons receiving nociceptive afferent inflow by descending pathways is well established in cat studies (26). It has been shown that unilateral acute inflammation increases the effectiveness of tonic descending inhibition, resulting in less hyperexcitability for the afferent input from the inflamed knee as well as for the input from regions of the contralateral leg (8,23,26). The experiments do not explain entirely the neurophysiology of voluntary-activation deficits but point toward central mechanisms adjusting the bilateral fusimotor-muscle-spindle system in cases of joint pathology.
Furthermore, it is of clinical interest to know whether the unconscious down-regulation of the voluntary quadriceps femoris muscle activation is a specific reaction to the instability after ACL-rupture because it is necessary for motor balance of the knee joint or whether a voluntary-activation deficit is a more nonspecific and undesirable effect. In the first case, afferents specifically stimulated by pathological transversal movements in the knee joint might increase inhibitory descending pathways on the quadriceps muscle to avoid anterior shift comparable with the supposed “quadriceps avoidance gait” in ACL deficiency (2). This would call into question whether it is useful to attempt to overcome a voluntary-activation deficit of the quadriceps and therefore interfering with its regulatory effect. In the second case, inhibiting afferents might lead to the voluntary-activation deficit stimulated by nonspecific pathology like trauma, infection, or effusion. The results of previous studies (27) showing that joint pathology without instability causes a voluntary-activation deficit clearly support the latter hypothesis, i.e., a nonspecific reaction functioning as “protection reflex” to avoid further joint or soft-tissue damage. In this context the contralateral inhibition may be regarded as a tool to maintain a bilateral balance of motor output.
The evidence of bilateral voluntary-activation deficits after ACL-rupture has impact on the critical evaluation of functional muscle tests using the contralateral extremity as reference. Commonly, the functional deficit is defined as the ratio of the ipsilateral to the contralateral muscle function as a percentage (4,14,19,28,30). The present study shows that a voluntary-activation deficit in the uninjured side leads to reduction of the maximal voluntary contraction, resulting in an underestimation of the functional deficit. Especially in patients with trauma yielding high voluntary-activation deficits, the false assessment of quadriceps function reaches levels of relevance. This study was performed on isometric torque measurements. However, the contralateral voluntary-activation deficit might also affect the outcome of isokinetic tests (e.g., the Cybex-Score) and more complex functional tests (like the “one-leg-hop” in the IKDC-form) (14) in which the uninjured leg is used as reference.
Discussing the clinical relevance of a voluntary-activation deficit of the quadriceps femoris muscle in patients with ACL deficiency, the individual results of the patients have to be considered rather than the collective ones for the whole patient group. The study showed that the voluntary activation in the patient group exhibited a distribution with a long tail toward low voluntary-activation values quite different from that of a proper control group (Fig. 1B). There were patients without any voluntary-activation deficit, whereas others presented voluntary-activation values of less than 80%. For those patients suffering from high voluntary-activation deficits one can expect only little benefit from physiotherapy based on voluntary cooperation. On the other hand, electrical muscle stimulation of the quadriceps femoris muscle might be successful particularly in those patients. The variability of voluntary-activation deficits after ACL-rupture might explain the conflicting results of studies investigating the success rate of this nonvoluntary muscle training (3,20,22). With respect to the patients presenting a high voluntary-activation deficit, one should note, however, that the presence of any knee injury in addition to the ACL rupture was excluded, indicating that the low voluntary-activation values were not a result of more severe joint damage. Furthermore, the clinical examinations did not reveal any peculiarities of the patients with high voluntary-activation deficits in comparison with those without apparent deficits.
Despite the fact of a nonuniform voluntary-activation deficit distribution, the quadriceps muscle of the injured leg presented a considerable rather uniform true maximal contraction decline compared with that in the normal leg, indicating that disuse atrophy was present in addition to a voluntary-activation deficit. This is emphasized especially by the fact that the true maximal contraction distribution of the uninjured side was identical to that of controls. Therefore, the maximal voluntary contraction deficit of the uninjured side could be explained by the apparent voluntary-activation deficit alone. It was claimed that intense physiotherapy is able to improve the maximal voluntary contraction deficit but not voluntary-activation deficits (16). However, the purpose of this study was not to assess the effectiveness of a particular rehabilitation regime since each patient’s program was individually structured, leaving it unclear whether specific modalities like biofeedback, electrical muscle stimulation, or proprioceptive training might have an effect on voluntary-activation deficits. Furthermore, stabilization of the knee joint by ACL-reconstruction might eliminate voluntary-activation deficits. These suppositions remain to be confirmed or rejected by prospective studies since they are of interest in other areas as well as in sports medicine.
This research war supported by Deutsche Forschungsgemeinschaft grant AW5/2–1 and AW5/2–2.
1. Allen, G. M., S. C. Gandevia, and D. K. McKenzie. Reliability of measurements of muscle strength and voluntary activation using twitch interpolation
. Muscle Nerve 18:593–600, 1995.
2. Andriacchi, T. P. Functional analysis of pre- and post-knee surgery: total knee arthroplasty and ACL reconstruction. J. Biomech. Eng. 115:575–581, 1993.
3. Arvidsson, I., H. Arvidsson, E. Eriksson, and E. Jansson. Prevention of quadriceps wasting after immobilization: an evaluation of the effect of electrical stimulation. Orthop. 9:1519–1528, 1986.
4. Arvidsson, I., E. Eriksson, T. Haggmark, and R. J. Johnson. Isokinetic thigh muscle strength after ligament reconstruction in the knee joint: results from a 5–10 year follow-up after reconstructions of the anterior cruciate ligament in the knee joint. Int. J. Sports Med. 2:7–11, 1981.
5. Awiszus, F., B. Wahl, and I. Meinecke. Influence of stimulus cross talk on results of the twitch-interpolation technique at the biceps brachii muscle. Muscle Nerve 20:1187–1190, 1997.
6. Baugher, W. H., R. F. Warrem, J. L. Marshall, and A. Joseph. Quadriceps atrophy in the anterior cruciate insufficient knee. Am. J. Sports Med. 12:192–195, 1984.
7. Bulow, P. M., J. Norregaard, B. Danneskiold-Samsoe, and J. Mehlsen. Twitch interpolation
technique in testing of maximal muscle strength: influence of potentiation, force level, stimulus intensity and preload. Eur. J. Appl. Physiol. 67:462–466, 1993.
8. Cervero, F., H. G. Schaible, and R. F. Schmidt. Tonic descending inhibition of spinal cord neurones driven by joint afferents in normal cats and in cats with an inflamed knee joint. Exp. Brain Res. 83:675–678, 1991.
9. Deandrade, J. R., C. Grant, and A. S. Dixon. Joint distension and reflex muscle inhibition in the knee. J. Bone Joint Surg. 47:313–324. 1965.
10. Dowling, J. J., E. Konert, P. Ljucovic, and D. M. Andrews. Are humans able to voluntarily elicit maximum muscle force? Neurosci. Lett. 179:25–28, 1994.
11. Elmqvist, L. G., R. Lorentzon, C. Johannsson, and A.R. Fuglmeyer. Does a torn anterior cruciate ligament lead to change in the central nervous drive of the knee extensors? Eur. J. Appl. Physiol. 58:203–207, 1988.
12. Elmqvist, L. G., R. Lorentzon, C. Johansson, M. Langstrom, M. Fagerlund, and A. R. Fugl.-Meyer. Knee extensor muscle function before and after reconstruction of anterior cruciate ligament tear. Scand. J. Rehabil. Med. 21:131–139, 1989.
13. Hales, J. P. and S. C. Gandevia. Assessment of maximal voluntary contraction with twitch interpolation
: an instrument to measure twitch responses. J. Neurosci. Methods 25:97–102, 1988.
14. Hefti, F., W. Muller, R. P. Jakob, and H. U. Staubli. Evaluation of knee ligament injuries with the IKDC form. Knee. Surg. Sports Traumatol. Arthrosc. 1:226–234, 1993.
15. Hurley, M. V., D. W. Jones, and D. J. Newham. Arthrogenic quadriceps inhibition and rehabilitation of patients with extensive traumatic knee injuries. Clin. Sci. (Colch. ) 86:305–310, 1994.
16. Hurley, M. V., D. W. Jones, D. Wilson, and D. J. Newham. Rehabilitation of quadriceps inhibited due to isolated rupture of the anterior cruciate ligament. J. Orthop. Rheum. 5:145–154, 1992.
17. Johansson, H., P. Sjolander, and P. Sojka. Actions on gamma-motoneurones elicited by electrical stimulation of joint afferent fibres in the hind limb of the cat. J. Physiol. (Lond.)
375:137–52: 137–152, 1986.
18. Jones, D. W., D. A. Jones, and D. J. Newham. Chronic knee effusion and aspiration: the effect on quadriceps inhibition. Br. J. Rheumatol. 26:370–374, 1987.
19. Kannus, P., M. Jarvinen, R. Johnson, et al. Function of the quadriceps and hamstring muscles in knees with chronic partial deficiency of the anterior cruciate ligament: isometric and isokinetic evaluation. Am. J. Sports Med. 20:162–168, 1992.
20. Lieber, R. L., P. D. Silva, and D. M. Daniel. Equal effectiveness of electrical and volitional strength training for quadriceps femoris muscles after anterior cruciate ligament surgery. J. Orthop. Res. 14:131–138, 1996.
21. Lorentzon, R., L. G. Elmquist, M. Sjostrom, M. Fagerlund, and A. R. Fuglmeyer. Thigh musculature in relation to chronic anterior cruciate ligament tear: muscle size, morphology, and mechanical output before reconstruction. Am. J. Sports Med. 17:423–429, 1989.
22. Morrissey, M. C., C. E. Brewster, C. L. J. Shields, and M. Brown. The effects of electrical stimulation on the quadriceps during postoperative knee immobilization. Am. J. Sports Med. 13:40–45, 1985.
23. Neugebauer, V. and H. G. Schaible. Evidence for a central component in the sensitization of spinal neurons with joint input during development of acute arthritis in cat’s knee. J. Neurophysiol. 64:299–311, 1990.
24. Newham, D. J., M. V. Hurley, and D. W. Jones. Ligamentous knee injuries and muscle inhibition. J. Orthop. Rheum. 2:163–173, 1989.
25. Norregaard, J., J. J. Lykkegaard, P. M. Bulow, and B. Dannekiold-Samsoe. The twitch interpolation
technique for the estimation of true quadriceps muscle strength. Clin. Physiol. 17:523–532, 1997.
26. Schaible, H. G., V. Neugebauer, F. Cervero, and R. F. Schmidt. Changes in tonic descending inhibition of spinal neurons with articular input during the development of acute arthritis in the cat. J. Neurophysiol. 66:1021–1032, 1991.
27. Shakespeare, D. T., M. Stokes, K. P. Sherman, and A. Young. Reflex inhibition of the quadriceps after meniscectomy: lack of association with pain. Clin. Physiol. 5:137–144, 1985.
28. Shelbourne, K. D., H. J. Whitaker, J. R. McCarroll, A. C. Rettig, and L. D. Hirschman. Anterior cruciate ligament injury: evaluation of intraarticular reconstruction of acute tears without repair: two-to-seven year follow-up of 155 athletes. Am. J. Sports Med. 18:484–488, 1990.
29. Snyder-Mackler, L., P. F. Deluca, P. R. Williams, M. E. Eastlack, and A. R. Bartolozzi. Reflex inhibition of the quadriceps femoris muscle after injury or reconstruction of the anterior cruciate ligament. J. Bone Joint Surg. (Am.) 76:555–560, 1994.
30. Snyder-Mackler, L., A. Delitto, S. W. Stralka, and S. L. Bailey. Use of electrical stimulation to enhance recovery of quadriceps femoris muscle force production in patients following anterior cruciate ligament reconstruction. Phys. Ther. 74:901–907, 1994.
31. Spencer, J. D., K. C. Hayes, and I. J. Alexander. Knee joint effusion and quadriceps reflex inhibition in man. Arch. Phys. Med. Rehabil. 65:171–177, 1984.
32. Stokes, M. and A. Young. The contribution of reflex inhibition to arthrogenous muscle weakness. Clin. Sci. 67:7–14, 1984.
33. Thomas, C. K., R. S. Johansson, G. Westling, and B. Bigland-Ritchie. Twitch properties of human thenar motor units measured in response to intraneural motor-axon stimulation. J. Neurophysiol. 64:1339–1346, 1990.
34. Tibone, J. E. and T. J. Antich. A biomechanical analysis of anterior cruciate ligament reconstruction with the patellar tendon: a two-year follow-up. Am. J. Sports Med. 16:332–335, 1988.
35. Wood, L., W. R. Ferrell, and R. H. Baxendale. Pressures in normal and acutely distended human knee joints and effects on quadriceps maximal voluntary contractions. Q. J. Exp. Physiol. 73:305–314, 1988.