Eledromyographic Analysis of Knee Rehabilitation Exercises

Transcription

1 Eledromyographic Analysis of Knee Rehabilitation Exercises Stephen M. Cryzlo, MD' Robert M. Patek, MD' Marilyn Pink, MS, PTZ lacquelin Perry, MD3 T he knee is a frequently injured area that often requires assessment by a physician and a physical therapist. Some injuries need only rest or reassurance, while others require some form of therapy to return the patient to his or her preinjury state. Others may need surgical intervention and postoperative rehabilitation (21). Knee rehabilitative exercises continue to be defined and redefined by clinical experience, observations, and basic science research. The general goals of knee rehabilitation are to increase strength, endurance, and range of motion. To reach these goals, the exercises must be performed through a safe range of motion and be efficient and effective (24). Electromyography (EMG) is an appropriate tool to measure the relative intensity of muscle activity occurring during exercises or functional activities (16). Soderberg and Cook and Soderberg et al, in two separate studies during the 1980s. analyzed the surface EMG activity of four lower extremity muscles during quadricep femoris muscle setting (quad sets) and straight leg raises (SLR) (23,24). In both reports, more muscle activity was found in the vastus medialis, biceps femoris, and gluteus medius during quad sets than during SLR. The rectus femoris was Many exercises are used to strengthen the knee muscles, yet limited studies that evaluate the exercises exist. The purpose of this study was to describe and compare the muscle firing patterns in five knee muscles during five rehabilitative exercises, which were presumed either to strengthen a specific muscle group or to elicit a cocontraction. During short-arc knee extension, the medial and lateral vasti were significantly more active during the last 15" of extension than during other arcs (p <.05). The short-arc knee extension with hamstring cocontraction demonstrated significantly more activity in the rectus fernoris, vastus rnedialis oblique, and vastus lateralis than the biceps femoris and semimembranosus in the final 15" of extension (p <.05). With isometric knee cocontractions, no significant difference in muscle activity occurred in any of the arcs tested. During squatting, the rectus femoris, vastus medialis oblique, and vastus lateralis were significantly more active than the biceps femoris and semimembranosus during the descending, holding, and arising phases (p <.05). This information offers suggestions in selecting optimal knee rehabilitation exercises. Key Words: electromyograph y, rehabilitative exercises, cocon tractions ' Clinical Instructor, Orthopaedic Surgery, Northwestern University Medical School, Chicago, I1 Director and Assistant Administrator, Biomechanics Laboratory, Centinela Hospital Medical Center, 555 E. Hardy St., Inglewood, CA 'Consultant, Biomechanics laboratory, Centinela Hospital Medical Center, Inglewood, CA more active during SLR than during quad sets. This corroborated a finding by Skurja et al, who reported more EMG activity from the vasti muscles (vastus medialis obliques and vastus lateralis) during isolated isometric knee extension than during SLR and, conversely, more increased EMG values for the rectus femoris during SLR than during isometric knee extension (22). It is obvious from these studies that different lower extremity muscles are active at varying degrees of intensity with different exercises. A current area of controversy in knee rehabilitation is the program for the patient with an anterior cruciate ligament (ACL) reconstruction (10). All authorities agree that rehabilitation is important, but the specific approach remains an enigma (2). In the last I0 years, many changes in the postoperative ACL reconstruction therapy program have occurred. Emphasis has shifted from rigid immobilization and nonweight bearing to immediate continuous passive motion and weight bearing as tolerated to even full knee extension on the first postoperative day and participation in sports by 4-6 months (1 5,19,20). Sachs et al found the most common complica- Volume 20 Number I *July 1994 *JOSPT

2 tions following ACL surgery to be quadriceps weakening, flexion contracture, and patellofemoral pain (18). They stated that all three were interrelated. Rehabilitation aimed at quadriceps strengthening and improved extension placed the reconstructed ACL at risk by increasing strain on the ligament and increasing anterior tibial translation (2,5,7-10,17,27). A potential answer to this problem was simultaneous hamstring and quadriceps contraction-so-called cocontraction (26). Draganich et al reported hamstrings/quadriceps cocontraction in healthy subjects during knee extension (6). In their study, the greatest hamstring activity occurred at less than 10" of flexion. The authors concluded that the hamstrings act in concert with the ACL to prevent anterior tibial translation. Additionally, in a cadaveric study, Renstrom et al reported a potential strain reduction role of the hamstrings when coactivated with the quadriceps in regards to the ACL (16). Finally, an in vivo study by Kain et al determined that if hamstring muscles fired before quadriceps muscles and if simultaneous contractions of both were sustained, then a decreased strain on the ACL occurred (10). These studies suggest that cocontractions of the hamstring/quadriceps group can reduce stress on the ACL. Clinically, this concept of cocontraction has been incorporated into the rehabilitation program for ACL reconstructions (l,l9). To date, EMG studies (I 2,14, 23-25) on lower extremity muscle activity have concentrated on only a few rehabilitation exercises, namely the SLR and quad sets exercises. Many exercises remain untested. At this time, objective information on coactivation of hamstring and quadricep muscles is available during simple knee motions (3.6) but not during rehabilitative exercises. Additionally, previous research (I 2,14, 23-25) has used surface electrodes JOSPT Volume 20 Number I July 1994 to record lower extremity EMG activity during the different knee motions. A problem with surface EMG is potential crosstalk or contamination of the recording signals by neighboring muscle activity (3). Indwelling wire electrodes can avoid this concern. The purpose of this paper is twofold. The first purpose is to describe and compare the EMG activity of the rectus femoris, vastus lateralis, vastus medialis oblique, biceps femoris, and semimembranosus with the use of fine-wire electrodes during five rehabilitative exercises. The second purpose is to compare the EMG activity within a muscle throughout the arc of motion for each exercise. Significant acfivify, however, did occur in the knee extensors during all phases of the squat. MATERIALS AND METHODS This study was conducted in the Biomechanics Laboratory at Centinela Hospital Medical Center in Inglewood, CA. Twelve healthy subjects volunteered for participation in the study (nine males and three females), with ages ranging from 25 to 32 years (mean age = 29). None of the participants had a history of knee injury. They all had a full range of motion, and no atrophy of the lower extremity muscles was present. Muscle activity of the right vastus medialis oblique, vastus lateralis, rectus femoris, semimembranosus, and the long head of the biceps femoris was recorded using indwelling wire electrodes. Recording of the signal utilized the Basmajian (4) single needle technique. Following proper skin preparation and isolation of the specific muscle, dual 50-p insulated wires with 2-3-mm bared tips were inserted into the muscle using a 25-gauge hypodermic needle as a cannula. The wires from each muscle were attached to ground plates and taped to the patient's body. The signals from the leads were transmitted using an FM-FM telemetry system (Model 4200-A, Bio-Sentry Telemetry, Torrance, CA), which was capable of transmitting up to four muscles simultaneously. Correct wire electrode placement was confirmed by a manual test specific to the inserted muscle with the telemetry signal monitored on an oscilloscope (Model A, Tektronix, Beaverton, OR). Each subject wore a battery-operated FM transmitter belt pack oriented to prevent any restrictions in bodily movements. The EMG information was bandpass filtered at 100-1,000 Hz and recorded on a multichannel instrumentation recorder (Model 3968A. Hewlett-Packard, Palo Alto. CA) for later retrieval and review. One 16-mm high-speed motion picture camera (Model DBM-55, Teledyne Camera System, Arcadia, CA) operating at 50 frames per second was positioned at a right angle to the subject and recorded his or her performance. Marks were electronically placed on the film and EMG data to allow for synchroni7ation. Later, the films were reviewed and the exercises were divided into arcs of motion. To begin the testing, resting EMGs were first recorded. The EMGs were then taken during a maximal manual muscle test (MMT) for each muscle. The muscle test positions were in accordance with standard physical therapy guidelines (1 1). The subjects were instructed through the series of exercises being evaluated to ensure proper performance. Two trials of each exercise were taken, and a rest averaging 3 minutes was given between each trial, thus eliminating the potential for fatigue.

3 The exercises were performed in the following sequence: Straight Leg Raise With the subject supine, a straight leg raise (SLR) was performed until a hip flexion angle of 75" was reached. No weight was attached to the leg in order to mimic an early stage of a rehabilitation program. Short-Arc Knee Extension The subjects were seated and a 12-inch roll was placed behind their right thigh. Their right knee was flexed to 45" off the end of the table. An isotonic short-arc quadricep exercise (SAEX) was performed from 45" of knee flexion to full extension at a constant rate with a 12.5-lb weight attached to the ankle. A 12.5-Ib weight was utilized as it mimicked an early stage of a rehabilitation program. Short-Arc Knee Extension with Hamstring Cocontraction The subjects were asked to perform the same exercise as above and this time add hamstring cocontractions (SAEXHS). This was accomplished by isometrically contracting their hip into the thigh roll as they extended their knees from 45" of flexion to full extension. Again, a 12.5-lb weight was attached to the ankle for the exercise in order to mimic an early stage of a rehabilitation program. Squat The subjects stood and were asked to descend to 90" of knee flexion. They held this position for 3 seconds and then ascended. Isometric Knee Cocontraction The subjects returned to the supine position. A 12-inch roll was placed behind their right thigh, and their right knee was placed off the end of the table. An isometric quadricep with hamstring cocontraction exercise (ICO) was performed at 15, 30, and 45" of knee flexion and held for a count of five. After the films were processed, they were reviewed and the isotonic exercises were divided into arcs of motion which were synchronized with the EMG. The SLR was divided into 15" arcs of hip flexion as measured by a goniometer. The SAEX and SAEXHS were divided into 15" arcs of knee extension as measured by a goniometer. The squat was divided into the descend phase, the hold phase, and the ascend phase. Throughout all of these exercises, the EMG activity of the five muscles was stored for later retrieval and review. The EMG data were converted from analog to digital signals by computer (PDP 1 1/23. Digital Equipment Corporation, Bedford, MA) at a sampling rate of 2,500 Hz, rectified, and quantitated by computer integration. After excluding the noise identified by the resting recording, the peak 1 -second EMG signal during a maximal isometric MMT was selected as a normalizing value (1 00% MMT). Activity patterns during the 15" arcs were assessed every 20 msec and expressed as a percentage of the normalization base. Data Analysis The percent MMT for each muscle during each arc was averaged for the two trials in each subject. A mean and standard deviation were calculated. For each exercise, the arc with highest EMG activity for each muscle was identified and called the peak arc. One-way repeated measures analyses of variance (ANOVAs) with repeated measures were executed ( p <.05). Twenty-one separate ANOVAs were done to compare the EMG activity throughout the arc of motion within a muscle for each exercise. For example, the SLR has five arcs of motion (0-15 ", 15-30, 30-45" ". and 60-75"). and the vastus medialis oblique EMG activity was compared throughout these arcs (ANOVA 1). The EMG activity throughout these arcs was compared separately for the vastus lateralis, as it was for the rectus femoris (ANOVAs 2 and 3; note the rows in Table 1). Analyses of variance were done for the other exercises as well (note the rows in Tables 2-5). Seventeen ANOVAs were executed to compare the EMG activity between the muscles during each arc of motion for each exercise. For example, during the SLR, three muscles were tested. Thus, the vastus medialis oblique, vastus lateralis, and rectus femoris were compared at the 0-15" arc, and then at the 15-30" arc, the 30-45" arc, the 45-60' arc, and the 60-75" arc (note the columns in Table 1). Thus, for this exercise, five separate ANOVAs were done to compare the EMG activity between the muscles during each arc of motion. Likewise, ANOVAs were done for the other exercises (note AK of Motion (Degrees of hip flexion) 0-15' 15-30' 30-45' 45-60' 60-75' - - Comparison of Arcs (pc.05) 1 SO X SD X SO 1 SO SO VMO 26f 23 26f18 25f f 21 NSD VL 29f34 28f26 29f NSD RF 42f18 38f15 36f15 33f13 34f20 NSD Comparison NSD NSD NSD NSD NSD of muscles NSD = No signiiicant diiierence. VMO = Vastus medialis oblique. RF = Rectus femoris. TABLE 1. Means and standard deviations for the straight leg raise exercise (SLR) (muscle activity as a percent of manual muscle test). Volume 20 Number 1 July 1994 JOSPT

4 VMO VL RF Arc of Motion (Degrees of knee extension) Comparison of Arcs 45-30' 30-15' 15-0' (P <.W - Ti SD Ti SD X SD ' > 30-IS', 45-30" 14f f ' > 30-IS", 45-30' 24 f f f 29 NSD Comparison of muscles NSD NSD NSD (p <.05) NSD = No significant difference. VMO = Vastus medialis oblique. RF = Rectus femoris. TABLE 2. Means and standard deviations for the short-arc knee extension exercise (SAEX) (muscle activity as a percent of manual muscle test). VMO VL RF BF SM An: of Motion (Degrees of knee extension) Comparison of Arcs 45-30' 30-15' 15-0' ( p <.05) - Ti SD X SD Ti SD f f ' > 30-IS', 45-30" 19f 8 39f f ' > 30-IS', 45-30' 15 f f f ' > 30-1 So, 45-30' 25f f 16 12f ' > 15-0' f f 1 1 NSD Comparison of muscles NSD NSD VM0,VL > RF > BF,SM (p <.05) NSD = No significant difierence. RF = Rectus femoris. VMO = Vastus medialis oblique. BF = Biceps femoris. SM = Semimembranosus. TABLE 3. Means and standard deviations for the short-arc knee extension with hamstring cocontraction exercise (SAEXHS) (muscle activity as a percent of manual muscle test). VMO VL RF BF SM Comparison of muscles (P <.05) Arc of Motion Descend Hold A d Comparison of A m 0-90' 90' 90-0' (P <.05) f 16 hold, ascend > descend 16f f f 18 hold, axend > descend hold > descend 1+2 2f2 4f5 NSD f 2 3f4 NSD VMO,VL,RF VMO,VL,RF VMO,VL,RF > BF,SM > BF,SM > BFISM VM0,VL NSD = No significant difference. RF = Rectus femoris. VMO = Vastus medialis oblique. BF = Biceps femoris. SM = Semimembranosus. TABLE 4. Means and standard deviations for the squat exercise (squat) fmuscle activity as a percent of manual muscle test). the columns in Tables 2-5). Five ANOVAs were done to compare EMG activity of the peak arc of the five exercises for each muscle (Table 6). When a significant difference was seen, a post hoc Tukey multiple comparison was done. The level of significance was accepted at p <.05. Straight Leg Raise During the SLR, the rectus femoris consistently had more muscle activity than the vastus medialis oblique and vastus lateralis in each 15 ' arc tested (Table I, Figure 1). This was best seen during the first 15" of the SLR when the rectus femoris measured 42% MMT, while the vastus medialis oblique and vastus lateralis measured 26 and 29% MMT, respectively. However, this difference and the others seen during the remaining arcs of motion were not statistically significant ( p >.05). When further analyzing the EMG activity of each individual muscle from arc to arc, again no significant difference was seen, although all three muscles demonstrated a trend of less EMG activity as the leg was progressively raised (ie., rectus femoris 42% MMT at 0-1 5' and 34% MMT at 60-75'). Short-Arc Knee Extension During this exercise, the muscle activity of the rectus femoris, vastus medialis oblique, and vastus lateralis increased as the knee approached full extension (Table 2. Figure 2). However, when comparing the activity of the three muscles during each arc, no one muscle was significantly more active than the next. When further analyzing the individual muscle activity from arc to arc, significance was found. During the 15-0" arc, the vastus medialis oblique had significantly more activity (56% MMT) than during the previous two arcs (28 and 16% MMT). The same JOSPT Volume 20 Number l July 1994

5 Arc of Motion 45' 30' 15' Ti so K so Ti so Comparison of Arts ( p <.05) VMO NSD VL f NSD RF 24f f 18 NSD BF 21 f 21 24f f 16 NSD SM f f 22 NSD Comparison NSD NSD NSD of muscles NSD = No significant difference. VMO = Vastus medialis oblique. RF = Rectus lemoris. BF = Biceps fernon's. SM = Semimembranosus. TABLE 5. Means and standard deviations for the isometric knee cocontraction exercise (KO) (muscle activity as a percent of manual muscle test). Peak Arc % MMT Comparison of Exercises SLR SAEX SAEXHS Squat ICO VMO SAEX,SAEXHS > SLR,Squat,lCO SAEX,SAEXHS > SLR,Squat,lCO No significant differences between exercises IC0,SAEXHS > 3 34 SAEX,SLR,Squat ICO > SAEX,SAEXHS,SLR,Sauat VMO = Vastus medialis oblique. SLR = Straight leg raise exercise. SAEX = Short-arc knee extension exercise. RF = Rectus femoris. SAEXHS = Short-arc knee extension with hamstring cocontraction exercise. BF = Biceps femoris. 1CO = Isometric knee cocontraction exercise. SM = Semimembranosus. TABLE 6. Percent manual muscle test and significant differences for the peak arc of each muscle during each exercise. was true for the vastus lateralis, demonstrating significantly more activity during the 15-0" arc (58% MMT) than earlier in the exercise (27 and 14% MMT). Although the activity of the rectus femoris increased as the knee extended (4 1 % MMT at 15-0" compared with 28 and 24% MMT at 30-15" and 45-30" arcs, respectively), this trend was not significant. Short-Arc Knee Extension With Hamstring Cocontraction No one muscle had significantly more EMG activity than the others during the 45-30" or 30-15" arc (Table 3, Figure 3). However, during the 15-0" arc, the activity of both the vastus rnedialis oblique and vastus lateralis (83 and 8 1 % MMT, respectively) was greater than the rectus femoris (50% MMT), biceps femoris (1 2% MMT), and semimembranosus (1 3% MMT, p <.05). Additionally, the EMG activity of the rectus femoris (50% MMT) during this arc when compared with the biceps femoris (1 2% MMT) and semimembranosus ( 1 3% MMT) was significantly more. When comparing individual muscle activity from arc to arc, the vastus rnedialis oblique had significantly more activity (83% MMT) in the 15-0" arc than during the 45-30" (29% MMT) or " arcs (45% MMT). This was also true of the vastus lateralis. Its EMG activity was greater during the 15-0" arc (8 1 % MMT) than during the 45-30" (19% MMT) or 30-15" arcs (39% MMT). The rectus femoris also demonstrated more activity during the 15-0" arc (50% MMT) than during the 45-30" (15% MMT) or " arcs (25% MMT). The biceps femoris had greater activity (p <.05) during the 45-30" arc (25% MMT) than during the 15-0" arc (12% MMT). Squat In all phases of this exercise (descend, hold, and ascend), the vastus medialis oblique, vastus lateralis, and rectus femoris had more activity (p <.05) than the biceps femoris and semimembranosus (Table 4, Figure 4). The vastus medialis oblique and vastus lateralis (38 and 31 % MMT. respectively) demonstrated more activity than the rectus femoris (1 6% MMT) during the ascend phase (90-0") of the squat (p <.05). During the descending and holding phase of the squat, there was no statistical difference between the vastus medialis oblique, vastus lateralis, and rectus femoris. When comparing individual muscle activity from phase to phase, the vastus medialis oblique and vastus lateralis had significantly more activity during the hold phase (3 1 and 28% MMT, respectively) and ascend phase (38 and 3 1 % MMT, respectively) than during the descend phase (20 and 16% M MT, respectively). For the rectus femoris, the most activity occurred during the hold phase (38% MMT) when compared with the descend (22% MMT) and ascend (16% MMT, p <.05) phases. There was no significant difference in the activity of the biceps femoris or semimembranosus from phase to phase. The most activity for Volume 20 Number 1 July 1994 JOSPT

6 STRAIGHT LEG RAISE EXERCISE (SLR) the rectus femoris, no exercise revealed significantly more activity than the next. The biceps femoris demonstrated the most activity during the ICO exercise at 15" (28% MMT) and also during the 45-30" arc (25% MMT) in the SAEXHS exercise ( p <.05). The semimembranosus revealed the greatest activity during the ICO at 15" (34% MMT, p <.05) (Table 6). 0-15' 15-30' 30-45' 45-60' 60-75' ARCS OF MOTION OVMO ~ V L ~ R F FIGURE 1. Muscle activity during the straight leg raise (means f standard deviations). VMO = Vastus medialis oblique, VL = Vastus lateralis, RF = Rectus lemoris. SHORT ARC KNEE EXTENSION EXERCISE (SAEX) ' 30-15' 15-0' ARCS OF MOTION FIGURE 2. Muscle activity during the short-arc knee extension (means + standard deviations). VMO = Vastus medialis oblique, VL = Vastus lateralis, RF = Rectus lemoris. these two muscles, the biceps femoris and semimembranosus, was during the ascend phase of the squat (4 and 3% MMT, respectively). Isometric Knee Cocontradion At each flexion angle tested (1 5, 30, and 45"). there was no significant difference in the EMG activity of any of the five muscles (Table 5, Figure 5). When comparing individual muscle activity from angle to angle, once again, none of the five muscles showed significant differences (p >.05). Comparison of Exercises Upon comparing all five of the exercises, the vastus medialis oblique and vastus lateralis demonstrated significantly more activity during the 15-0" arc of the SAEX (56 and 58% MMT) and SAEXHS (83 and 8 1 % MMT) exercises than during any portion of the SLR, squat, or ICO exercises (p <.05). With regard to DISCUSSION Unlike work by Soderberg and Cook (23) and Soderberg et al (24), this study did not find the SLR exercise to provide significantly more EMG activity for the rectus femoris than the vastus medialis oblique and vastus lateralis. In fact, none of the five exercises examined during this study provided significantly more activity for the rectus femoris than the rest. Rather, the rectus femoris was equally active when compared with the other muscles throughout all five exercises. This difference may represent the use of fine wire electrodes used in this study compared with the use of the surface electrodes in other studies, or it may be due to low statistical power with a sample size of only 12 subjects. Our results during the SAEX exercise reflect the findings of Lieb and Perry (I 3). Their study concluded that a 60% increase in force was needed to complete the final 15" of extension. It was during those same final 15" in our study that we recorded a doubling of EMG activity for the vastus medialis oblique (28 to 56% MMT) and VL (27 to 58% MMT) when compared with the motion arc of 30-15". Additionally, this exercise portrayed the vastus medialis oblique and vastus lateralis with identical muscle action throughout the motion arcs. Our study suggests that the SAEX and SAEXHS exercises are the most effective exercises in generating vastus medialis oblique and vastus lateralis EMG activity. Clinically, JOSPT Volume 20 Number I July 1994

7 SHORT ARC KNEE EXTENSION WITH HAMSTRING CO-CONTRACTION (SAEXHS) ' 15-0" ARCS OF MOTION OW0 HVL QRF NBF HSM FIGURE 3. Muscle activity during the short-arc knee extension with hamstring cocontraction (means + standard deviations). VMO = Vastus medialis oblique, VL = Vastus lateralis, RF = Rectus femoris, BF = Biceps femon's, SM = Semimembranosus. SQUAT OW0 HVL URF HBF HSM FIGURE 4. Muscle activity during the squat (means + standard deviations). VMO = Vastus medialis oblique, VL = Vastus lateralis, RF = Rectus femoris, BF = Biceps femoris, SM = Semimembranosus. ISOMETRIC KNEE CO-CONTRATION (ICO) KNEE FLEXION ANGLE FIGURE 5. Muscle activity during the isometric knee cocontraction (means + standard deviations). VMO = Vastus medialis oblique, VL = Vastus lateralis, RF = Rectus femoris, BF = Biceps femoris, SM = Semimembranosus. if knee motion is not contraindicated, patients would benefit most with these two exercises to restore vastus medialis oblique and vastus lateralis strength as well as rectus femoris strength. Of major interest when this study began were the cocontraction exercises: SAEXHS and ICO. With regard to the ICO exercise, our study demonstrated balanced or matched EMG activity for both the knee flexors and extensors. No single muscle or muscle group studied revealed significantly more activity than the other at each angle tested, including 30 and 15" of flexion. During the SAEXHS exercise, balanced EMG activity occurred between the hamstrings and quadriceps groups during the 45-30" and 30-15" motion arcs. Balance, however, did not carry over into the final 15-0" motion arc. Rather, significantly more extensor activity was demonstrated than hamstring activity. Clinically, knee cocontraction exercises have been advocated in the rehabilitation program for patients who have undergone ACL reconstructive surgery (1 9). The theory is that hamstring contraction during knee extension will prevent anterior tibia1 translation (27) as well as prevent increased strain on the ACL graft (2,8,10,17) and possibly avoid knee flexion contractions. We agree with this assumption and further suggest that effective hamstring activity must occur for the exercise to be considered a cocontraction. Other studies have revealed EMG activity in the hamstring group (less than 10% MMT) with knee extension (3.6). However, we believe this level of EMG activity is too small to be considered a balanced cocontraction. Our study revealed no significant differences between quads and hamstring activity at every arc except during the 15-0" arc of SAEXHS. This would suggest that the ICO exercise and SAEXHS exercise minus the final 15" of extension may be safe to perform in the ACLreconstructed patient; also, as afore- Volume 20 * Number I *July 1994 *JOSPT

8 mentioned, there was a low statistical power. The squat, or 90" knee bend, is another exercise used for the patient with an ACL reconstruction (20). This study clearly indicates the virtually nonexistent EMG activity of the biceps femoris and semimembranosus muscles during all phases of this exercise. Significant activity, however, did occur in the knee extensors (rectus femoris, vastus medialis oblique, and vastus lateralis) during all phases of the squat. This evidence of a lack of cocontraction suggests that the therapist should utilize the squat with caution in the patient with a reconstructed ACL. This study provides descriptive evidence for muscle activity in healthy subjects during early-stage rehabilitation programs. These reference data offer a baseline upon which to compare future studies done on patients with specific knee disorders. CONCLUSIONS Three key points can be reported from this study: 1 ) The final 15" of extension during the short-arc knee extension exercise or short-arc knee extension with hamstring cocontraction exercise demonstrated the greatest EMG activity for the vastus medialis oblique and vastus lateralis. 2) Balanced cocontractions occurred with short-arc knee extension with hamstring cocontraction and isometric knee cocontraction exercises at all angles and arcs except during the final 15" of extension in the short-arc knee extension with hamstring cocontraction exercise. 3) The squat exercise provided all quadriceps activity but no hamstring coactivity. JOSPT REFERENCES 1. Antich TI, Brewster CE: Modification of quadriceps femoris muscle exercises during knee rehabilitation. Phys Ther 66: , Arms SW, Pope MH, lohnson RI, Fischer RA, Arvidsson I, Erickson E: The biomechanics of anterior cruciate ligament rehabilitation and reconstruction. Am I Sports Med , Barratta R, Solomonow M, Zhov BH, Letson D, Chuinard R, D'Ambrosia R: Muscular coactivation: The role of the antagonist musculature in maintaining knee stability. Am I Sports Med , Basmajian /V: Muscles Alive (4th Ed), Baltimore: Williams & Wilkins, Daniel DM, Stone ML, Barnett P, Sachs R: Use of the quadriceps active test to diagnose posterior cruciate ligament disruption and measure posterior laxity of the knee. I Bone loint Surg 7OA: , Draganich LF, laeger RI, Kralj AR: Coactivation of the hamstring and quadriceps during extension of the knee. I Bone loint Surg 71A: , Crood ES, Suntay WI, Noyes FR, Butler DL: Biomechanics of the knee-extensor exercise. I Bone loint Surg 66A: , Henning CE, Lynch MA, Click KR: An in vivo strain gauge study of elongation of the anterior cruciate ligament. Am I Sports Med 13:22-26, lurist KA, Otis /C: Anteroposterior tibiofemoral displacements during isometric extensor effects. Am / Sports Med 13: , Kain CC, McCarthy /A, Arms SW, Pope MH, Steadmore lr, Monshe PR, Shinely RA: An in vivo analysis of the effect of transcutaneous electrical stimulation of the quadriceps and hamstring on anterior cruciate ligament deformation. Am I Sports Med 16: , I. Kendall HO, Kendall FP, Wadsworth CE: Muscles Testing and Function (Vol 2), Baltimore: Williams & Wilkins, Knight KL, Martin /A, Londeree BR: EMC comparison of quadriceps femoris activity during knee extension and straight leg raises. Am I Phys Med Rehabil58:57-69, Lieb F/, Perry I: Quadriceps function. / Bone loint Surg 5OA: , Moller BN, Krebs B, Tidemand C, Aaris K: Isometric contractions in the patellofemoral pain syndrome. Arch Orthop Trauma Surg 105:24-27, Paulos L, Noyes FR, Crood E, Butler DL: Knee rehabilitation after anterior cruciate ligament reconstruction and repair. Am I Sports Med 9: , Perry I, Bekey CA: EMC-force relationship in skeletal muscle. Crit Rev Biomed Eng 7: 1-22, Renstrom P, Arms SW, Stanwyck TS, lohnson RI, Pope MH: Strain within the anterior cruciate ligament during hamstring and quadriceps activity. Am I Sports Med 14:83-87, Sachs RA, Daniel DM, Stone ML, Carfein RF: Patellofemoral problems after anterior cruciate ligament reconstruction. Am I Sports Med 17: , Seto /L, Brewster CE, Lombardo S/, Tibone / : Rehabilitation of the knee after anterior cruciate ligament reconstruction. I Orthop Sports Phys Ther 1 1 (l):8-18, Shelbourne KD, Nitz P: Accelerated rehabilitation after anterior cruciate ligament reconstruction. Am I Sports Med 18: , Shields CO Ir.: Manual of Sports Surgery, New York: Springer-Verlag, Skurja M, Perry I, Cronley :, Hislop H: Quadriceps action in straight leg raise versus isolated knee extension. Phys Ther 60:582, Soderberg CL, Cook TM: An electromyographic analysis of quadriceps femoris muscle setting and straight leg raising. Phys Ther 63: , Soderberg CL, Miner SD, Arnold K, Henry 7, Chatterson lk, Poppe DR, Wall C: Electromyographic analysis of knee exercises in healthy subjects and in patients with knee pathologies. Phys Ther 76: , Wild /I, Franklin TD, Woods CW: Patellar pain quadriceps rehabilitation: An EMC study. Am I Sports Med 10: 12-27, Yasuda K, Sasaki 1: Exercise after anterior cruciate ligament reconstruction. Clin Orthop 220: , Yasuda K, Sasaki 7: Muscle exercise after anterior cruciate ligament reconstruction. Clin Orthop 220: , 1987 JOSPT Volume 20 Number I July 1994