Proprioceptive deficit in patients with complete tearing of the anterior cruciate ligament.

OBJECTIVE: To investigate the existence of proprioceptive deficits between the injured limb and the uninjured (i.e. contralateral normal) limb, in individuals who suffered complete tearing of the anterior cruciate ligament (ACL), using a strength reproduction test.
METHODS: Sixteen patients with complete tearing of the ACL participated in the study. A voluntary maximum isometric strength test was performed, with reproduction of the muscle strength in the limb with complete tearing of the ACL and the healthy contralateral limb, with the knee flexed at 60°. The meta-intensity was used for the procedure of 20% of the voluntary maximum isometric strength. The proprioceptive performance was determined by means of absolute error, variable error and constant error values.
RESULTS: Significant differences were found between the control group and ACL group for the variables of absolute error (p = 0.05) and constant error (p = 0.01). No difference was found in relation to variable error (p = 0.83).
CONCLUSION: Our data corroborate the hypothesis that there is a proprioceptive deficit in subjects with complete tearing of the ACL in an injured limb, in comparison with the uninjured limb, during evaluation of the sense of strength. This deficit can be explained in terms of partial or total loss of the mechanoreceptors of the ACL.

Introduction

Proprioception is defined as the conscious capacity to perceive position, movement and the forces imposed on and produced by body segments. It has a crucial role in joint stability and postural and motor control.2, 3 Therefore, it is essential for adequate functioning of the joint structures during day-to-day activities and sports practice. The main ways of evaluating proprioception are through testing joint position sense (JPS), the perception threshold for passive movement5, 6, 7 and the sense of strength.8, 9

With regard to the knee joint, tearing of the anterior cruciate ligament (ACL) is the commonest injury, and its incidence has been increasing over the years. It has been estimated that in the United States, 95,000 people suffer injuries to this ligament every year.

The ACL functions together with other anatomical structures surrounding the knee so as to maintain static and dynamic balance. It has an important role in proprioceptive monitoring of mechanical receptors such as Pacini corpuscles and Ruffini endings.12, 13, 14, 15

Many studies have indicated that subjects with partial tearing of the ACL present proprioceptive deficits.16, 17, 18 These deficits can be considered to be factors predisposing toward knee instability: they adversely affect the activity, balance and strength of the quadriceps and increase the risk of new injuries to the knee.

The functional and proprioceptive levels of the knee in subjects with partial tearing of the ACL have been measured previously and most studies have used tests on JPS or on the threshold of detection of passive movement.13, 20 All of these studies have found deficits in the injured limb in comparison with the uninjured limb.14, 21, 22 The sense of strength has received more attention in the literature recently, but few data are available in relation to this paradigm for evaluating proprioception in the knee joint, given that there are no studies that have evaluated patients with total tearing of the ACL.

Thus, the objective of the present study was to use a strength reproduction test to investigate the existence of proprioceptive deficits between the injured limb and the uninjured limb (i.e. the normal contralateral limb), in individuals who have suffered total tearing of the ACL. In this regard, our study hypothesis was that individuals with total tearing of the ACL would present proprioceptive deficits in the injured limb, in comparison with the uninjured limb.

Materials and methods

Subjects

Sixteen volunteers from both sexes aged between 18 and 40 years (mean age, 27.6 ± 2.9; mean height, 172.2 ± 6.7; and mean weight, 74.4 ± 12.9) participated in this study. All of them presented total tearing of the ACL in one of their legs. Volunteers presenting any of the following were excluded from the sample: previous surgery in the leg with ACL tearing; any other type of injury to that limb; joint degeneration (characterized by joint crepitation in any of the knee compartments); chondral lesions diagnosed through magnetic resonance imaging; and/or signs of osteoarthrosis seen on radiographs of the knee. All the subjects were evaluated clinically by the same orthopedist and they signed a consent statement in which the objectives and conditions of the experiment were described in detail. This statement had been approved by our institution's ethics committee in accordance with Resolution 196/96 of the National Health Council.

Experimental procedure and task

An isokinetic dynamometer (CSMI, Humac Norm) was used, and before each test, the equipment was properly calibrated. The subjects were positioned to sit comfortably, with the lateral condyle of the femur aligned with the rotation axis of the apparatus and the ankle fixed to the rod of the knee assessment accessory by means of a Velcro strip (Fig. 1).

Fig. 1

Photo of an individual positioned in the dynamometer.

The following tests were performed:

Muscle strength and force reproduction tests were performed with the knee extended at 60°. In order to perform a maximum voluntary isometric strength (MVIS) test, the subjects warmed up and became familiarized with the equipment by means of five repetitions, without resistance imposed by the apparatus, performed over their entire range of joint motion. After this familiarization, the subjects performed specific warm-ups with three submaximal isometric contractions (with subjective effort of 20%, 40% and 60% of the maximum force), with a one-minute interval between them. The MVIS was conducted after a three-minute interval, and three attempts were made with three-minute intervals between them. The greatest instantaneous torque found was taken to be 100% of the MVIS. Each isometric contraction lasted for six seconds. The uninjured limb was evaluated first.

A reproduction test on the ipsilateral strength of the knee extensors was performed after leaving a 10-min interval following the MVIS test. The target intensity for the procedure was 20% of the MVIS. The procedure consisted of performing a reference contraction, in which visual feedback of the torque level was used. The subjects were instructed to maintain the desired force level. Immediately after the reference contraction, the subjects attempted to reproduce the force produced previously, as precisely as possible, without visual feedback. Three attempts were made, with three-minute intervals between them. Each isometric contraction lasted for six seconds.

Dependent variables

The individual error value for each attempt was determined as the difference between the reproduced force and the force experienced. The proprioceptive performance was determined by means of the values for the absolute error (AE), variable error (VE) and constant error (CE). Schmidt and Lee described the calculations for each variable, in detail. Briefly, the AE is obtained from the arithmetic mean of the individual errors in the modulus and determines the individual's accuracy in reproducing force; the VE is the standard deviation of the individual errors and determines the consistency of the reproductions performed; and the CE is the arithmetic mean of the individual errors with the signs and determines the tendency to reproduce the force above or below the target (bias). Only the period of the torque curves from two to six seconds was used for determining the AE, CE and VE. Pilot tests demonstrated that this would be the period needed for the subjects to stabilize the intensity of contraction, and within which there would be the least effect from fatigue while sustaining the force.

Statistical analysis

Descriptive statistics (mean ± standard deviation (SD)) were used to describe the data. The dependent variables were the AE, VE and CE. The data were subjected to the Shapiro–Wilk test of normality. Comparisons were made between the injured and uninjured limbs. The values determined for 20% of MVIS were compared using the t test for paired measurements. The calculations were made using the Statistical Package for the Social Sciences (SPSS Inc. Chicago, IL, USA). The significance level established was p ≤ 0.05.

Results

Time and cause of the injury and associated lesions

The mean length of time from the injury until the data gathering was 3.2 ± 1.6 years. In the majority of the cases, the injuries occurred during recreational soccer practice, all without contact (68.75%). Other cases occurred in relation to surfing (6.25%), falling (6.25%), playing handball (6.25%), playing basketball (6.25%) and suffering motorcycle accidents (6.25%).

Clinical examinations

These are presented in Table 1.

Table 1

Clinical examinations.

Lachman and anterior drawer
	−	+/+++	++/+++	+++/+++
Lachman	3	13	0	0
Anterior drawer	0	16	0	0
McMurray and Pivot
	−	+
Bocejo	8	8
Pivot	3	13

Values for absolute error (AE), variable error (VE) and constant error (CE)

After the data-gathering, the strength values were standardized in relation to body weight. This was done to make it possible to make comparisons between our subjects. The MVIS calculated for the uninjured limb was 3.2 ± 1.0 N/kg and for the injured limb, 3.0 ± 1.1 N/kg. No statistically significant difference was found for the MVIS (p = 0.059). The t test demonstrated a significant difference between the limbs regarding the variables AE (p = 0.05) and CE (p = 0.01) (Fig. 2). There was no difference regarding VE (p = 0.83). The means and standard deviations for the individual errors relating to strength are given in Table 2.

Fig. 2

Graphical representation of the absolute error (AE) results (A) and the constant error (CE) results (B).

Table 2

AE, VE and CE determined for 20% MVIS (mean ± SD).

	AE	VE	CE
ACL	4.3 ± 2.2%a	1.6 ± 1.2%	4.1 ± 2.3%a
Control	3.0 ± 1.3%	1.7 ± 1.1%	1.9 ± 2.1%

Significantly different from the control limb.

Discussion

The present study had the objective of determining whether patients with total tearing of the ACL presented proprioceptive deficits in the injured limb during assessment of their sense of strength. For this, muscle strength and force reproduction tests were performed on the injured and uninjured limbs. In this regard, we worked with the hypothesis that individuals with total tearing of the ACL would present proprioceptive deficits in the injured limb, in comparison with the uninjured limb. Our hypothesis was in line with the study by Héroux and Tremblay, who identified proprioception deficits in this same population using a weight discrimination test.

In particular, the sense of strength was evaluated in 16 patients with unilateral total tearing of the ACL, using 20% of their MVIS. Significant differences were found in the force reproduction test between the injured and uninjured limbs. The AE results demonstrated that the injured limb was less capable of accurately reproducing the force, given that the CE results demonstrated that although both limbs tended to overestimate the target, the injured limb overestimated this much more. With regard to the VE, we did not find any statistical difference, which shows that the individuals were consistent regarding the errors.

Using the JPS test, Lee et al. and Carter et al. found significantly different AE values between the injured and uninjured limbs. Héroux and Tremblay conducted a study on the sense of strength using the weight discrimination test and also obtained results that indicated lower acuity in the injured side. The results from our investigation corroborate the results from these three experiments. Few studies on the sense of strength have used the force reproduction test, which makes it difficult to compare the results.

Thus, one likely explanation for the low accuracy of force reproduction on the injured side may be partial failure of the process of calibrating the descending motor commands due to impaired assessment of the force signals resulting from muscle contraction. This possibility raises the question of which afferent information sources are susceptible to being affected by ACL injuries.

Among the receptors, the Golgi tendon organ (GTO) is considered to be the main source of afferent inputs coming from peripheral regions relating to muscle strength and tension.24, 25, 26 However, because the GTO is located in the muscle-tendon area, it should not be affected by ACL injuries. Thus, involvement of the GTO in the low acuity of our subjects seems very unlikely.

There is evidence that afferent stimulation of the ACL may influence knee flexor and extensor activity during voluntary contractions and that the sensory innervation of the joints is rarely recovered after injury. Our subjects’ difficulty in force reproduction can be attributed to loss of innervation of the mechanical receptors of the ACL, which thus reduces the quantity of sensory information relating to tension and force during the test.

It is also possible that some mechanical receptors located in the joint capsule that were spared from injury may nonetheless have been altered. Khalsa and Grigg investigated this possibility in an animal model and concluded that the afferent response capacity of the joint capsule was not significantly affected after complete transection of the ACL. Thus, we believe that there was some residual innervation in the injured knee, together with the inputs from the GTO, which remained intact in the quadriceps. This would explain the variability between individuals that was observed. It might also explain why the capacity to reproduce force, albeit reduced in comparison with the uninjured leg, was still relatively well conserved in our subjects.

In relation to the MVIS, our results did not find significantly different values: 3.2 ± 1.0 N/kg for the uninjured leg and 3.0 ± 1.1 N/kg for the injured leg. This differed from what was found by Héroux and Tremblay. We attribute this difference in results to the fact that our subjects were mostly practitioners of physical activity. However, our MVIS values were much higher than those found in the study by Héroux and Tremblay. This was probably because although they also used an isokinetic dynamometer, it was done at with the knee extended at 45°. We used a knee extension of 60°, which produced greater mechanical efficiency of the quadriceps, with the capacity to reach higher peak torque, according to the database of our laboratory (unpublished data). This makes it impossible to directly compare the results, because we used force values that were standardized according to weight.

Thus, our data corroborate the hypothesis that there is a proprioceptive deficit among subjects with total tearing of the ACL in one injured limb, in comparison with the uninjured limb during assessment of the sense of strength. This proprioceptive deficit seems to be better explained in the study by Hogesvorst and Brand, along with that of Khalsa and Grigg, who attributed the capacity for force reproduction to losses, or to the continuing existence of some residual innervation of the mechanical receptors of the ACL, along with inputs from the GTO, which remained intact in the quadriceps and reduced the quantity of sensory information.

In this regard, because of the absence of studies relating to this problem, we suggest that new studies should be conducted with the aims of expanding the knowledge on this subject and enabling comparison of results.

Conflicts of interest

The authors declare no conflicts of interest.