FUNCTIONAL KNEE BRACING is often prescribed for

Transcription

1 1680 Effects of Functional Knee Bracing on Muscle-Firing Patterns About the Chronic Anterior Cruciate Ligament Deficient Knee Jay Smith, MD, Gerard A. Malanga, MD, Bing Yu, PhD, Kai-Nan An, PhD ABSTRACT. Smith J, Malanga GA, Yu B, An K-N. Effects of functional knee bracing on muscle-firing patterns about the chronic anterior cruciate ligament deficient knee. Arch Phys Med Rehabil 2003;84: Objective: To examine the effects of functional knee bracing on the muscle-firing patterns about the chronic anterior cruciate ligament (ACL) deficient knee in successful brace users. Design: Cross-sectional comparative clinical trial. Setting: Motion analysis laboratory. Participants: Ten active individuals with unilateral, isolated, chronic ( 18mo postinjury), ACL-deficient knees who subjectively reported improved function with a functional knee bracing. Intervention: Each subject completed 3 single-leg hop maneuvers on their ACL-deficient knee with and without their knee brace while surface electromyographic activity was recorded from the quadriceps, hamstring, and gastrocnemius muscles. Main Outcome Measure: Muscle onset latency. Results: Brace use significantly delayed the average onset of vastus lateralis activation before landing (123 47ms vs ms, P.001), though significant interindividual variations existed. Bracing significantly altered the onset latency in 1 or more muscles in 9 of 10 subjects. In 4 subjects, a favorable change in the firing pattern was seen, whereas only 1 subject exhibited an unfavorable change. Without bracing, 5 of the 10 subjects fired the hamstrings or gastrocnemius muscles first; with bracing, 7 of 10 fired these muscles first. Conclusions: Brace use in this population did not consistently result in more favorable muscle firing patterns during the single-leg hop maneuver. Interindividual responses to brace use indicate the need for further research to investigate the multiple strategies that may exist to stabilize the ACL-deficient knee. In the meantime, functional knee brace use among ACLdeficient patients remains empirical. Key Words: Anterior cruciate ligament; Braces; Proprioception; Rehabilitation by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation From the Department of Physical Medicine and Rehabilitation, Mayo Clinic Sports Medicine Center (Smith); and Motion Analysis Laboratory (An), Mayo Clinic, Rochester, MN; Department of Physical Medicine and Rehabilitation, Kessler Institute for Rehabilitation, West Orange, NJ (Malanga); and Center for Human Movement Sciences, Division of Physical Therapy, University of North Carolina, Chapel Hill, NC (Yu). Poster presented at the 62nd Annual Assembly of the American Academy of Physical Medicine and Rehabilitation, November 4, 2000, San Francisco, CA. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. Reprint requests to Jay Smith, MD, Mayo Clinic, 200 First St SW, Rochester, MN 55905, smith.jay@mayo.edu /03/ $30.00/0 doi: /s (03) FUNCTIONAL KNEE BRACING is often prescribed for anterior cruciate ligament (ACL) deficient individuals who desire to continue sports participation at an equal or modified level. 1-3 The rationale for prescribing a functional knee brace (FKB) is based primarily on research that suggest that the majority of ACL-deficient individuals report improved knee stability, reduced frequency of giving way, or improved performance while wearing a functional knee brace. 1,4-7 Despite the supportive subjective results among brace users, the exact mechanisms through which a functional knee brace may improve knee stability remain unknown. The ability of knee braces to mechanically control abnormal knee motions in ACLdeficient knees has been questioned Consequently, some authors 2,6,13-15 have proposed that functional knee bracing may improve neuromuscular control about the knee, possibly through proprioceptive mechanisms. Relatively few studies have attempted to elucidate the effects of functional knee brace on muscle control about the ACLdeficient leg during functional activities. Branch et al 16 demonstrated proportionately reduced quadriceps and hamstrings muscle activity among ACL-deficient patients during the landing phase of a cutting maneuver performed with a functional knee brace when compared with the unbraced condition. Branch suggested that the support offered by the functional knee brace resulted in the reduced need for dynamic muscle stabilization. 16 More recently, Nemeth et al 6 reported that ACL-deficient skiers wearing a functional knee brace demonstrated increased lateral hamstring electromyographic activity during periods of increased knee flexion. In contrast to Branch, 16 Nemeth 6 suggested that functional knee bracing may dynamically stabilize the ACI-deficient patient by stimulating increased hamstring activity Temporal muscle-firing patterns about the knee offer an additional method to study the neuromuscular effects of bracing. Normal individuals generally show a more favorable pattern of earlier reflex firing of the hamstrings (ACL agonists) relative to the quadriceps (ACL antagonists) when challenged by an external load. 17,19,20 This pattern is reversed to an unfavorable pattern of earlier quadriceps firing in many ACLdeficient individuals. 20 Among ACL-deficient patients, the extent of this pattern reversal correlates better with knee function than static laxity measures such as the Lachman test and KT-1000 arthrometer Delays in hamstring reflex muscle contraction appear to attenuate with time, 20 and some research suggests that this process may be facilitated via specific rehabilitative exercise. 22,24,25 Only 2 previously published studies have examined FKBinduced changes in muscle-firing patterns as determined by electromyography. Wojtys and Huston 20 reported that functional knee bracing actually delayed certain measures of reflex hamstring activation in response to an anterior tibial translatory load under laboratory conditions. Branch 16 found that functional knee bracing had no effect on the time to peak electromyographic activity in the preimpact phase of a cutting maneuver. Neither study evaluated the onset latencies of knee

2 KNEE BRACING, Smith 1681 muscle activation during functional maneuvers, during which knee stabilizing, feed-forward neuromuscular activity may be most evident. In addition, neither study 16,20 controlled for concomitant knee injuries, brace use experience and perceived efficacy by the user, or prior surgery. Population heterogeneity may therefore have confounded the results. The purpose of our study was to examine the effects of functional knee bracing on the muscle firing patterns about the chronic ( 18mo postinjury), unilateral, isolated ACL-deficient knee in individuals who were preselected as copers. 26,27 That is, all individuals subjectively reported improved function with FKB use. We hypothesized that functional knee bracing in this preselected population would favorably alter muscle onset latencies about the ACL-deficient knee during the preimpact phase of single-leg hop task, as manifested by earlier hamstring and/or gastrocnemius muscle activation relative to the quadriceps. 9,17,19,20,24,28,29 We hoped to increase the sensitivity to detect brace-induced muscle firing alterations by selecting a more homogenous group of subjects who subjectively felt that their braces (all used the same brace design) improved their activity tolerance. METHODS Participants A total of 10 individuals (7 men, 3 women) with chronic, isolated, unilateral ACL-deficient were recruited to participate by advertisement at the primary author s institution. Inclusion criteria included (1) age 18 to 40 years at the time of the study; (2) isolated ACL injury, defined as magnetic resonance imaging (MRI) or arthroscopically documented complete ACL disruption without clinical, surgical, or radiographic evidence of concurrent or subsequent articular or ligamentous injury (MRI or arthroscopy performed within 3mo of injury in all cases); (3) chronic injury ( 18mo) 9 ; and (4) regular successful brace use with a CTi functional knee brace. a Regular successful brace use was defined as the presence of all 3 of the following: (1) participation in sporting activities at least twice per week, (2) use of the functional knee brace during sporting activities, and (3) subject perception that functional knee bracing was necessary to allow him/her to participate at a desired activity level. Only subjects using the CTi functional knee brace were included because this brace was the most popular brace used in our area at the time of the study, and previous research 8,9 had indicated that potentially significant differences exist between brace designs. Therefore, use of a single brace type by all subjects was deemed desirable for the purposes of this study. Exclusion criteria included (1) concomitant knee injury or surgery (other than arthroscopic examination); (2) contralateral lower-limb injury; (3) documented neurologic, metabolic bone, or connective tissue disease; (4) regular ( 1d/wk) use of analgesic medications for any reason; (5) knee pain, effusion, range of motion limitation, or crepitus at the time of the study; (6) asymmetrical thigh atrophy greater than 1cm measured 10cm proximal to the medial knee joint line; (7) posterolateral corner laxity more than 5 asymmetry in tibial external rotation; (8) greater than grade 1 laxity recorded during clinical examination of the knee when tested in the medial, lateral, or posterior directions; (9) clinical examination suggestive of meniscal pathology; (10) use of a functional knee brace other than a CTi; or (11) any condition preventing the subject from completing the single-leg hop task 30 (eg, ankle sprains or instability, hip disease) or the inability to achieve 80% of the noninjured leg distance in a single leg hop. Informed consent was obtained for all subjects participating in the study. The study was approved by the institutional review board at Mayo Clinic. Clinical Data Acquisition All subjects completed a standard knee examination for atrophy, effusion, crepitus, motion loss, and ligamentous laxity. Ligamentous laxity was graded on a scale from 0 to 3 during the Lachman test, posterior drawer, and varus-valgus stress tests at 30 of knee flexion. 31 The pivot shift was graded a 0 (normal), 1 (pivot glide), 2 (clunk), or 3 (locking). 32,33 Static anterior laxity was measured with the KT-1000 arthrometer b by using a maximal manual difference technique between the affected and unaffected legs. 33 The KT-1000 is a portable measuring device that allows the examiner to quantify anterior tibial translation, similar to an anterior drawer test. 31,33 A maximal side-to-side difference of more than 3mm has been shown to indicate the presence of ACL deficiency. 33 The posterolateral corner was tested in the standard manner with the knee flexed at 30, 31 and meniscal testing was performed to document lack of knee pain or crepitus. 31 Each subject completed the Lysholm functional knee scale and an activity profile with and without a functional knee brace. 27,34 The single-leg hop task was adopted from Gauffin and Tropp 30 as a functional test for ACL-deficient individuals. To perform a single-leg hop, the subject stood on the affected leg with the arms behind the back. The subject then jumped forward as far as possible to land on the affected leg and hold that position until stability was gained. Each subject was allowed several trials of the single-leg hop task to determine his/her comfort level after a demonstration of the task by the primary author (JS). To participate in the study, each subject needed to achieve at least 80% of the distance achieved with the unaffected knee with his/her affected, unbraced knee. 30 Biomechanical Data Acquisition The MA-100B electromyography system c was used to collect surface electromyographic signals of the vastus medialis and lateralis, medial and lateral hamstrings, and medial and lateral gastrocnemius muscles during the single-leg hop test. Each subject s skin was prepared by lightly abrading with fine sandpaper and wiping with alcohol to reduce skin resistance. Recording electrodes were then placed onto the midportions of each muscle. Electrode placement followed standard techniques, 35 although slight modifications of up to 1cm were necessary to accommodate the surface electrodes when the functional knee brace was worm. For this reason, the functional knee brace was always placed on the subject, and the electrodes adjusted accordingly, regardless of the testing order. Electrode positions were then secured with tape. In this manner, for each subject the electrode position did not change between the braced and unbraced trials. Raw electromyographic signals were collected at a sampling rate of 2500Hz via bipolar Ag-AgCl surface electrodes with an interelectrode distance of 3cm. Surface electrodes were connected to a backpack device attached to the subject with an unobtrusive belt, and wires transferred the signals to a computer for recording (fig 1). A Bertec forceplate d was used to collect vertical ground reaction force signals during the landing of a single-leg hop at a sampling rate of 2500Hz through the same computer. After all electrodes were placed, electromyographic signals were visually confirmed for each muscle using manual muscle testing. Resting electromyographic signals were recorded from all muscle groups for 5 seconds while subjects rested supine on an examining table.

3 1682 KNEE BRACING, Smith Table 1: Subject Characteristics (N 10) Gender 7 men, 3 women Age (y) 34 (27 49) Age at injury (y) 27 (16 46) Side of injury 6 right, 4 left Injury documentation 6 MRI, 4 arthroscopy Time since injury (mo) 86 (30 228) Duration of brace use (mo) 57 (22 96) Lysholm knee score 93 (75 100) NOTE. Values are n or means with ranges. the muscle reached its peak 1-second activity for the postimpact phase. Statistical Analysis Descriptive analysis of muscle firing patterns was used to analyze common muscle-firing sequences with and without the functional knee brace. A 2-way analysis of variance (ANOVA) without repeated measures was used to detect significant differences in muscle onset latencies or time to peak activities by subject or condition (braced vs unbraced). Tukey post hoc tests were used when the ANOVA revealed significant difference. Statistical significance was set at P less than.05. RESULTS Fig 1. Subject landing on forceplate during single-leg hop without brace. Each subject then completed 3 successful trials of single-leg hop with the ACL-deficient knee under the braced and unbraced conditions while surface electromyographic activity was collected from the 6 lower-limb muscles (fig 1). A successful trial was defined as a trial in which the subject landed with his/her takeoff foot on the forceplate, held an upright position after landing, and achieved a distance of at least 80% of that achieved with the contralateral lower limb. By using a balanced randomization scheme, the order of braced versus unbraced was randomized but provided for 5 subjects completing braced-unbraced sequence, and 5 subjects completing the opposite sequence. No subject reported pain, instability, or injury during or after testing. Data Reduction Electromyographic signals were filtered though a band-pass filter with a lower-pass cutoff frequency of 350Hz and a high-pass cutoff frequency of 20Hz. Each single-leg hop trial was divided into a preimpact phase and a postimpact phase based on the vertical ground reaction force signal recorded from the forceplate at the time of landing. For reach muscle, onset latency was defined as the point at which the muscle s electromyographic activity achieved a value twice the resting signal. Onset latencies before impact (preimpact) were given negative values, and those subsequent to impact (postimpact) were given positive values. The time to peak muscle electromyographic activity was determined for each muscle as the time between impact with the forceplate and the point at which Participants Subject characteristics are shown in table 1. All subjects were at least 30 months post-acl injury and had been using their braces for a minimum of 22 months. Lysholm scores (table 1), clinical testing (table 2), and physical activity profiles (table 3) support the relative homogeneity of the subject population and profile consistent with copers. 26,27 Table 3 supports the preselection for individuals who were successful brace users as defined in this study. Only 1 subject performed level 2 activities (lateral motions) without his functional knee brace. In comparison, all 10 subjects performed level 1 or 2 activities while wearing their braces. All test subjects completed the 3 single-leg hops with and without a knee brace, and without complication. The number of repetitions needed to complete 3 successful single-leg hops ranged from 3 to 6 without the brace and 3 to 5 with the brace. Muscle Onset Latencies All 6 muscles achieved initial onset during the preimpact phase of the single-leg hop in both the braced and unbraced conditions. ANOVA and post hoc testing indicated that, on average, bracing significantly delayed the onset latency of the Table 2: Instrumented and Clinical Laxity Testing for Subjects (N 10) KT-1000 (max side-side) 3.3 (3 5)* Normal Grade 1 Grade 2 Grade 3 Anterior drawer Lachman test Pivot shift Posterior drawer test Quadriceps active test External rotation at * Value is mean with range.

4 KNEE BRACING, Smith 1683 Table 3: Subject Physical Activity Profiles (N 10) Sports Level Preinjury* Postinjury Without brace With brace * Each subject included only once, based on most intense activity completed on a regular basis (at least twice per week). Activity classes are as follows: level 1, jumping, pivoting, hard cutting (basketball, football, soccer); level 2, lateral motion, less intense jumping or hard cutting vs level 1 (baseball/softball, all racquet sports [except doubles tennis], downhill skiing, dance); level 3, other sports (jogging/running, swimming, cross-country skiing, doubles tennis, bicycling, golf); and level 4, walking, daily activities. vastus lateralis relative to the unbraced condition (F 43.2, P.001) (table 4). No other statistically significant changes were seen in onset latency. As shown in table 4, the preimpact muscle-firing order on average was hamstrings (either medial or lateral), followed by quadriceps (either vastus medialis or lateralis), and finally gastrocnemius (either medial or lateral) in both the braced and unbraced conditions. The ANOVA also revealed a statistically significant interaction between the subject and condition (braced vs unbraced) on muscle onset latencies (P.05 for all muscles), suggesting that subjects muscle onset latencies responded differently to bracing. Figure 2 shows this interaction effect with respect to the changes observed in the vastus lateralis. Although bracing significantly delayed the average vastus lateralis onset in the study population, the vastus lateralis onset latency was actually delayed in 3 subjects (P.05), earlier in 3 (P.05), and unchanged in 4 (P.05). Without bracing, half of the subjects activated the gastrocnemius or hamstring muscles first, whereas 7 of 10 subjects activated 1 of these muscles first while wearing their brace (table 5). Bracing significantly altered the onset latency of 1 or more muscles in 9 of the 10 subjects (table 5). In 4 of these 9, the muscle firing order did not qualitatively change (neutral). In 4 subjects, the firing order was favorably altered as determined by an earlier hamstring or gastrocnemius activation relative to the quadriceps (significantly earlier activation of the hamstring or quadriceps muscles and/or a significant delay in the quadriceps muscles). Only in subject 9 did brace use result in a statistically significant onset latency change that produced an unfavorable alteration in muscle firing order. Table 4: Mean Onset Latencies for Subjects (N 10) During Single-Leg Hop Lower-Limb Muscle MH LH VM VL MG LG Unbraced Onset* Firing order Braced Onset* Firing order NOTE. Larger values represent earlier onset. Abbreviations: LG, lateral gastrocnemius; LH, lateral hamstrings; MG, medial gastrocnemius; MH, medial hamstrings; VL, vastus lateralis; VM, vastus medialis. *Values are milliseconds prior to impact standard deviation (SD). P.001. Fig 2. The time (ms) before impact when the vastus lateralis muscle initially fired without (control) and with (brace) the functional knee brace. Bracing delayed the vastus lateralis onset on average, although significant interindividual differences are shown. Time to Peak Electromyographic Activity The time to reach peak 1-second electromyographic activity in the postimpact phase was also calculated for each of the 6 muscles. All 6 muscles achieved their peak 1-second values during the postimpact phase in both the braced and unbraced conditions. ANOVA did not reveal a statistically significant effect of bracing on time to peak electromyographic activity for any of the 6 muscles tested (table 6), nor did it reveal any significant interaction between subjects and brace effect. DISCUSSION The potential stabilizing role of the hamstrings and gastrocnemius muscles in the ACL-deficient knee has been well documented. 9,17,19,20,24,28,29 In the presence of ACL disruption, Table 5: Muscle Firing Sequence by Subject With and Without Bracing Subject Without Brace With Brace Qualitative Change 1 Q-H-G H-Q-G* Favorable 2 Q-H-G G-H-Q Favorable 3 H-G-Q H-G-Q NS 4 Q-H-G Q-H-G Neutral 5 Q-H-G G-H-Q Favorable 6 Q-H-G Q-H-G Neutral 7 G-Q-H G-H-Q Favorable 8 H-Q-G H-Q-G Neutral 9 H-G-Q Q-H-G Unfavorable 10 H-G-Q H-G-Q Neutral H first (%) G first (%) Q first (%) NOTE. Values based on onset latencies. All muscles fired prior to impact. Abbreviations: G, gastrocnemius (medial or lateral gastrocnemius); H, hamstrings (medial or lateral hamstrings); Q, quadriceps (vastus medialis or lateralis); NS, not significant. * Boldfaced letters indicate statistically significant changes in onset latency between the braced and unbraced conditions (P.05). Qualitative muscle firing order changes were considered favorable if the hamstrings or gastrocnemius fired earlier than the quadriceps with the brace versus without the brace.

5 1684 KNEE BRACING, Smith Table 6: Time to Peak Electromyographic Activity Postimpact for Subjects (N 10) During Single-Leg Hop Lower-Limb Muscle MH LH VM VL MG LG Unbraced Time to peak* Firing order Braced Time to peak* Firing order NOTE. Smaller values represent faster times to reach peak electromyographic activity. * Values are milliseconds after impact SD. these muscles may antagonize the anterior tibiofemoral shear forces imparted by an isolated quadriceps contraction and provide dynamic knee stability through a coordinated cocontraction with the quadriceps muscle group during functional activities. 9,17,19,27,29,36 However, the optimal firing pattern or balance of these muscles about the ACL-deficient knee is currently unknown. What is known is that some ACL-deficient individuals can accommodate the loss of the ACL, 5,26 and it appears that those with more normal neuromuscular control about the knee have better functional outcomes On an electrophysiologic level, this adaptation has been recorded primarily as earlier and/or greater activity in the hamstrings or gastrocnemius muscle groups. 16,20-22 Therefore, for our purposes, potentially favorable responses to functional knee bracing were considered an earlier activation (shorter onset latency) of either the hamstring or gastrocnemius muscles groups relative to the quadriceps muscle groups. The primary variable of interest was the initial onset latency during the preimpact phase of the single-leg hop maneuver. It is during this time that the effects of the functional knee brace on anticipatory muscle activation in preparation for landing would be manifest. Group data did show a statistically significant delay in the vastus lateralis onset latency with functional knee bracing, a potentially favorable effect (table 4). However, when considering the grouped data, this change did not alter the qualitative firing pattern of the lower-limb muscles; the medial and lateral hamstrings fired first on average, regardless of brace use (table 4). This observation would support the role of the hamstrings to decelerate knee extension in preparation for landing while also potentially preparing the knee to counteract anterior tibiofemoral shear forces to be generated by the quadriceps during the impact phase of landing. 9,17,19,20,30 The mean quantitative delay in vastus lateralis activation may, on average, provide the hamstrings more time to carry out these functions. There are no prior studies that directly compare the findings of our study. It is noteworthy that the general pattern of hamstring-quadriceps-gastrocnemius activation in the preimpact phase of the single-leg hop (table 4) is the same as the pattern of the intermediate and voluntary reflex responses recorded by Wojtys and Huston 20 in patients with chronic ACLdeficient knees ( 18mo postinjury), despite the fact that substantial methodologic differences exist between the 2 studies. Branch et al 16 used a methodology more similar to our study and reported that functional knee bracing failed to produce significant changes in muscle firing patterns in the preimpact phase of a cutting maneuver in 10 ACL-deficient patients. However, Branch 16 studied the time to peak electromyographic activity in the preimpact phase and not the initial onset of each muscle. There were significant interindividual differences in the effect of bracing on muscle onset latencies between the 10 study subjects (table 5). Bracing significantly altered at least 1 muscle onset latency in 9 of the 10 subjects, but no dominant pattern of change emerged. Four subjects exhibited statistically significant changes that resulted in a favorable qualitative change in muscle firing sequence (ie, relatively earlier hamstrings or gastrocnemius firing). Only 1 subject (subject 9) exhibited a statistically significant change that resulted in a qualitatively unfavorable response. Overall, 5 of 10 subjects exhibited a favorable muscle firing sequence without bracing, whereas 7 of 10 exhibited a favorable muscle firing sequence with bracing. It is also interesting to note that the results shown in figure 2 seem to suggest a convergence of vastus lateralis onset latencies between the 10 subjects when wearing the brace as compared with the more divergent values in the unbraced condition. Direct comparison between our study and previous studies is problematic. Wojtys and Huston 20 have published the only data specifically examining muscle-firing sequence in ACL-deficient individuals. However, they examined reflex activity in response to anterior tibial translatory loading and not during the preimpact phase of a functional activity as used in our study. Nonetheless, they 20 did note that chronic ACL-deficient individuals initially activated the gastrocnemius and/or hamstring muscles 68% to 85% of the time (depending on the reflex recorded) when the knee was challenged by an anterior tibiofemoral shearing force. In comparison, our subjects initially activated 1 of these muscles 50% of the time without the brace, and 70% of the time with the brace (table 5). By using a similar study design, Wojtys et al 9 later reported that functional knee bracing improves many spinal reflexes, while slowing some intermediate and voluntary reflex times. No consistent changes in muscle firing sequence were identified with brace use, a CTi brace was not 1 of the 6 FKB designs studied, and individual data on the 5 chronic ACL-deficient study subjects was not provided. Wojtys 9 expressed concern regarding the observation that braces may slow some reflexes times. We did not observe any statistically significant slowing of onset latency or time to peak 1-second activity in any muscles using our model. The variability in individual responses to bracing for onset latencies and time to peak electromyographic values was surprising given our specific attempt to select a homogenous study population. Differential individual responses to bracing as recorded by electromyographic parameters have been documented in previous studies using different methodologies. Osternig and Robertson 15 showed that a prophylactic knee brace significantly reduced quantitative electromyographic activity in 42% of all braced versus nonbraced comparisons of the knee flexors and extensors but significantly increased activity in 17% of the same comparisons. Data from Wojtys et al 20,24 indicated interindividual differences in muscle reflex patterns and a lack of a universal recruitment pattern in response to anterior tibiofemoral shear forces.

6 KNEE BRACING, Smith 1685 In our study, we had hoped to reduce a significant portion of this variability by preselecting a more homogenous study population than those used in these previous studies. All 10 of our subjects had chronic, isolated ACL-deficient knees and used their own CTi braces during the study. Because the average time from injury was 86 months at the time of study participation, we cannot be sure some subjects had not incurred additional structural knee changes. However, because none of our 10 subjects reported subsequent knee injuries, additional structural changes would have to be considered subclinical and therefore of questionable significance for our purposes. The similarities in symptoms, function, and laxity of our subject population ensured an optimal and reasonable degree of homogeneity. This population was difficult to recruit because of the high incidence of subsequent clinical injuries and pain among chronic ACL-deficient patients. 5,37 Therefore, we were limited to 10 subjects after a year of recruiting. In addition, we preselected for successful brace users based on the hypothesis that we would be more likely to detect significant muscle onset latency changes in individuals with high subjective brace use satisfaction. Several potential methodologic points may explain the interindividual responses to bracing. First, electromyographic recordings in biomechanical studies commonly exhibit relatively large physiologic variability. 16,38,39 Second, there may be additional confounding factors not strictly controlled in our study, such as brace fit, thigh muscle strength, gender, or age. Any of these factors may affect an individual s response to functional knee bracing. Third, functional knee bracing effects may attenuate with time. Our subjects were a minimum of 30 months postinjury. Bracing may have more dramatic and consistent effects earlier post-acl injury. Time-dependent kinetic, 13,40 kinematic, 13,40,41 and electrophysiologic 20 changes in ACL-deficient knees have been well documented. Fourth, more consistent responses may be evident in individuals with less static stability and poorer knee function. The KT1000 laxity measurements and activity profiles of our subjects resembled ACL deficient copers as defined by Daniel et al. 26 Many copers appear to have the ability to accommodate for ACL deficiency, and this ability is not necessarily dependent on wearing a brace. It is possible that bracing would produce a more dramatic and consistent change in onset latency in a more unstable knee. 42 Aside from the previously mentioned methodologic issues, it is possible that functional knee bracing does not produce consistent changes in muscle onset latency. If functional knee bracing promotes dynamic knee stability, the mechanism may involve factors other than onset latency. Recently, increased attention has been focused on the role of the entire lower limb as it relates to ACL injury and rehabilitation. 27 Perhaps the effects of a functional knee brace are more global and may only be detected by integrating electromyographic and kinetic-kinematic observations of the entire lower limb during functional activities. Such methods may be better positioned to explain the movement pattern changes reported in ACL-deficient or ACL-reconstructed patients wearing braces. 4,6,13,14,16 A second possibility is that there are multiple potential stabilizing strategies in the ACL-deficient knee. Our data do suggest an overall shift to a more favorable qualitative muscle-firing pattern (table 5). The lack of consistency in this shift may be an indicator that several different adaptive strategies exist to stabilize the knee. 27,30 The coping strategy of the individual would in part dictate that individual s response to bracing. This hypothesis would explain the variable changes in electromyographic parameters recorded in our study as well as others 6,9,15,20 and is supported by recent research suggesting that multiple strategies may be used to stabilize the ACL-deficient knee. 27,43 If this hypothesis is true, elucidation of the mechanism through which braces might improve knee stability may be delayed until the various coping patterns can be identified. It would also suggest that future studies should also analyze individual responses to bracing. As we showed (tables 4, 5; fig 2), analyzing a study population statistically in such cases may obscure potentially significant interindividual responses to bracing. Several study limitations are noteworthy. First, as discussed, there may be additional confounding factors not controlled for in this study. Second, we chose a value of twice the resting electromyographic signal amplitude to determine each muscle s onset latency. Although previous researchers have noted that the threshold for onset can significantly affect the results, 16 our results did not significantly change when we reanalyzed the data by using a value of 3 times the resting signal. Third, we used surface electrodes because of their greater reliability in gait and motion studies. 44 It is possible that fine-wire electrodes would have provided more precise electromyographic data to determine onset latency. However, we hoped to avoid the sampling error that can occur with indwelling electrodes. 39 Fourth, the single-leg hop may not have provided enough stress to manifest the beneficial effects of bracing. Although we cannot be certain, previous research 30 suggests that the singleleg hop provides a significant sagittal plane challenge to the ACL-deficient knee. Finally, our study may have lacked the statistical power to detect a true bracing effect. 45 The primary variable of interest was muscle onset latency during the preimpact phase of the single-leg hop. Because there are no prior studies with which to compare our methodology directly, power calculations are problematic and understandably imprecise. Pilot work in our laboratory, as well as research by Wojtys et al, 8,9 suggested that onset latencies would be approximately 120 to 140ms for the quadriceps and hamstrings. No prior data exist to estimate either random variation or effect size for the purposes of our study, which is essentially a pilot study. Assuming approximately 30% variability (table 4) and an estimated effect size of a 10% change in onset latency with bracing, post hoc power calculations (type I error, 2-tailed,.05; type II error,.20) suggest that up to 50 subjects may be required in a larger scale study to investigate the effect of functional knee bracing during the single-leg hop. 45 A 20% effect size would reduce this number to approximately 25 subjects. 45 These calculations suggest that our study may have contained insufficient power to detect a true functional knee brace effect. Unfortunately, because it took over 1 year to recruit our 10 subjects at a major medical center, completion of a larger scale study would be challenging. Future research may need to focus on a different model, as previously discussed. CONCLUSIONS In the 10 subjects with chronic, isolated, unilateral ACLdeficient knees preselected as successful brace users, functional knee brace use did not consistently result in more favorable muscle-firing patterns during the single-leg hop maneuver. Although no consistent pattern of change occurred, 5 of 10 subjects exhibited a favorable muscle-firing sequence (hamstrings or gastrocnemius first) without an FKB, whereas 7 or 10 exhibited a favorable sequence while wearing their braces. Significant interindividual responses to functional knee bracing with respect to muscle onset latencies exist. This suggests that individuals should be considered separately in future studies examining the effects of functional knee brace on electrophysiologic variables. In addition, this variability may be a manifestation of the multiple neuromuscular coping strategies that exist for ACL-deficient patients to stabilize their knees.

7 1686 KNEE BRACING, Smith Although significant changes in muscle onset latency were observed during functional knee bracing in our study, insufficient evidence exists to interpret these changes to recommend or not recommend these braces for ACL-deficient patients. Future study is required. References 1. Cawley P, France E, Paulos L. The current state of functional knee bracing. Am J Sports Med 1991;19: Styf J. Effects of functional knee bracing on muscle function and performance. Sports Med 1999;28: DeCoster L, Vailas J, Swart W. Functional ACL bracing: a survey of current opinion and practice. Am J Orthop 1995;24: Cook F, Tibone J, Redfern F. Dynamic analysis of a functional bracing for anterior cruciate ligament insufficiency. Am J Sports Med 1989;17: Noyes F, Matthews D, Moarr P, Grood E. The symptomatic anterior cruciate ligament deficient knee: Part II: The results of rehabilitation, activity modification, and counseling on functional disability. J Bone Joint Surg Am 1983;65: Nemeth G, Lamontagne M, Tho K, Ericksson E. Electromyographic activity in expert downhill skiers using functional braces after anterior cruciate ligament injuries. Am J Sports Med 1997;25: Mishra D, Daniel D, Stone M. The use of functional knee braces in the control of pathological anterior knee laxity. Clin Orthop 1989;Apr;(241): Wojtys E, Loubert P, Samson S, Viviano DM. Use of a knee brace for control of tibial translation and rotation: a comparison, in cadavera, of available models. J Bone Joint Surg Am 1990;72: Comment in: J Bone Joint Surg Am 1992;74: Wojtys E, Kothari S, Huston L. Anterior cruciate ligament functional brace use in sports. Am J Sports Med 1996;24: Jonsson H, Karrholm J. Brace effect on the unstable knee in 21 cases: a roentgen stereophotogrammetric comparison of 3 designs. Acta Orthop Scand 1990;61: Beynnon B, Howe J, Pope M, Johnson R, Fleming B. The measurement of anterior cruciate ligament strain in vivo. Int Orthop 1992;16: Beynnon B, Johnson R, Braden C, et al. The effect of functional knee bracing on the anterior cruciate ligament in the weightbearing and nonweightbearing knee. Am J Sports Med 1997;25: DeVita P, Torry M, Glover K, Speroni D. A functional knee brace alters joint torque and power patterns during walking and running. J Biomech 1996;29: Kuster M, Grob K, Kuster M, Wood G, Gachter A. The benefits of wearing a compression sleeve after ACL reconstruction. Med Sci Sports Exerc 1999;31: Osternig L, Robertson R. Effect of prophylactic knee bracing on lower extremity joint position and muscle activation during running. Am J Sports Med 1993;21: Branch T, Hunter R, Donath M. Dynamic EMG analysis of anterior cruciate deficient legs with and without bracing during cutting. Am J Sports Med 1989;17: Solomonow M, Baratta R, Zghou B. The synergistic action of the anterior cruciate ligament and the thigh muscles in maintaining joint stability. Am J Sports Med 1987;15: Renstrom P, Arms S, Stanwyck T, Johnson R, Pope M. Strain within the anterior cruciate ligament during hamstrings and quadriceps activity. Am J Sports Med 1986;14: O Conner J. Can muscle co-contraction protect knee ligaments after injury or repair? J Bone Joint Surg Br 1993;75: Wojtys E, Huston L. Neuromuscular performance in normal and anterior cruciate ligament deficient lower extremities. Am J Sports Med 1994;22: Beard D, Kyberd R, Fergusson C, Dodd C. Proprioception after rupture of the anterior cruciate ligament. J Bone Joint Surg Br 1993;75: Beard D, Kyberd R, O Conner J, Fergusson C, Dodd C. Reflex hamstring contraction latency in anterior cruciate ligament deficiency. J Orthop Res 1994;2: Walla D, Albright J, McAuley E, Martin RK, Eldridge V, El- Khoury G. Hamstring control and the unstable anterior cruciate ligament-deficient knee. Am J Sports Med 1985;13: Wojtys E, Huston L, Taylor P, Bastian S. Neuromuscular adaptations in isokinetic, isotonic, and agility training programs. Am J Sports Med 1996;24: Ihara H, Nakayama A. Dynamic joint control training for knee ligament injuries. Am J Sports Med 1986;14: Daniel D, Stone M, Dobson B, Fithian D, Rossman D, Kaufmann K. Fate of the ACL-injured patient. Am J Sports Med 1994;22: Rudolph KS, Fitzgerald GK, Snyder-Mackler L. Restoration of dynamic stability in the ACL deficient knee. In: Lephart S, Fu F, editors. The role of proprioception and neuromuscular control in management in articular pathology. Champaign: Human Kinetics; p Wojtys E, Wylie B, Huston L. The effects of muscle fatigue on neuromuscular function and anterior translation in healthy knees. Am J Sports Med 1996;24: Lass P, Kaalund S, le Fevre S, Arendt-Nielsen L, Sinkjaer T, Simonsen O. Muscle coordination following rupture of the anterior cruciate ligament: electromyographic studies of 14 patients. Acta Orthop Scand 1991;62: Gauffin H, Tropp H. Altered movement and muscular-activation patterns during one-legged jump in patients with old anterior cruciate ligament rupture. Am J Sports Med 1992;20: Magee D. Orthopedic physical assessment. 3rd ed. Philadelphia: WB Saunders; Slocum D, James S, Larson R, Singer K. A clinical test for anterolateral rotary instability of the knee. Clin Orthop 1976;Jul-Aug(118): Beynnon B, Ryder S, Konradsen L, Johnson R, Johnson K, Renstrom P. The effect of anterior cruciate ligament trauma and bracing on knee proprioception. Am J Sports Med 1999;27: Tegner Y, Lysholm J, Odensten M, Gillquist J. Evaluation of cruciate ligament injuries: a review. Acta Orthop Scand 1988;59: Delagi E, Iazzeti J, Perotto A, Morrison D. In: Perotto AO, editor. Anatomical guide for the electromyographer: the limbs and trunk. 3rd ed. Springfield: CC Thomas; p Goldfuss A, Morehouse C, LeVeau B. Effect of muscular tension on knee stability. Med Sci Sports Exerc 1973;5: Noyes F, Mooar P, Matthews D, Butler D. The symptomatic anterior cruciate ligament deficient knee: Part I: The long-term functional disability in athletically active individuals. J Bone Joint Surg Am 1983;65: Ciccotti M, Kerlan R, Perry J, Pink M. An electromyographic analysis of the knee during functional activities. Am J Sports Med 1994;22: Basmajian J, DeLuca C. Muscles alive: their functions revealed by electromyography. 5th ed. Baltimore: Williams & Wilkins; p DeVita P, Blankenship P, Skelly W. Effects of a functional knee brace on the biomechanics of running. Med Sci Sports Exerc 1992;24: Berchuck M, Andriacchi T, Bach B, Reider B. Gait adaptations by patients who have a deficient anterior cruciate ligament. J Bone Joint Surg Am 1990;72: Beynnon BD, Good L, Risberg MA. The effect of bracing on proprioception of knees with anterior cruciate ligament injury. J Orthop Sports Phys Ther 2002;32: Shiavi R, Limbird T, Borra H, Edmonstone M. Electromyographic profile of the knee joint musculature during pivoting: changes induced by anterior cruciate ligament deficiency. J Electromyogr Kinesiol 1991;1: Kadaba M, Wootten M, Gainey J, Gainey J, Gorton G, Cochran GV. Repeatability of phasic muscle activity: performance of surface and intramuscular wire electrodes in gait analysis. J Orthop Res 1985;3: Browner WS, Black D, Newman TB, Hulley SB. Estimating sample size and power. In: Hulley SB, Cummings SR, Browner WS, Grady D, Hearst N, Newman TB, editors. Designing clinical research. Philadelphia: Williams & Wilkins; p Suppliers a. Innovation Sports, Pauling, Foothill Ranch, CA b. MEDmetric Corp, 7542 Trade St, San Diego, CA c. Motion Lab System Inc, Old Hammond Hwy, Baton Rouge, LA d. Bertec Corp, 6185 Huntley Rd, Ste B, Columbus, OH