DEBRA A. VAN HORN, MS, PT, ATC,t JOYCE L. MACKINNON, EdD, PT,S PHILIP L. WITT, MS, PT, DANIEL N. HOOKER, PhD, PT, ATCll

0196-6011/88/0907-0254$02.00/0 THE JOURNAL OF ORTHOPAEOIC AN0 SPORTS PHYSICAL THERAPY Copyright 0 1988 by The Orthopaedic and Sports Physical Therapy Sections of the American Physical Therapy Association Comparison of the Effects of the Anderson Knee Stabler and McDavid Knee Guard on the Kinematics of the Lower Extremity during Gait* DEBRA A. VAN HORN, MS, PT, ATC,t JOYCE L. MACKINNON, EdD, PT,S PHILIP L. WITT, MS, PT, DANIEL N. HOOKER, PhD, PT, ATCll The purpose of this study was to compare gait patterns among subjects wearing Anderson Knee Stabler braces, McDavid Knee Guards, and no braces. Fifteen male subjects were filmed while running on a treadmill at 4 mph and 8 mph without a brace, with an Anderson Knee Stabler, and with a McDavid Knee Guard. Fourteen gait variables were measured for each brace and speed condition. Analysis of the variables with multivariate ANOVA indicated that there was an increase in hip and knee flexion and knee angular velocity with and without braces at 8 mph as compared to 4 mph, a decrease in knee extension when either brace was worn, and minimal gait pattern differences with the Anderson Knee Stabler as compared with the McDavid Knee Guard (all results p < 0.05). The results of this study demonstrate that no clear superiority exists between the braces' effect on the gait characteristics measured. Therefore, other parameters should be considered when making brace selections. The knee is one of the most traumatized joints of people participating in contact sports22 and injuries to this joint tend to have grave consequences. Often injuries to the knee complex require a longer loss of time from sports participation than most other injuries.24 Fifty percent of those athletes sustaining knee injuries during the 1975 to 1982 college football seasons missed more than 7 days of participation because of their injuries,23 and athletes that sustain knee injuries tend to have residual difficulties long after the initial injury.15 Sports medicine professionals and researchers are aware of the vulnerability of the knee joint and have been searching for methods to protect it. 'This study was completed in partial fulfillment of Ms. Van Horn's master's degree, University of North Carolina at Chapel Hill. t Physical Therapy Department, Aspen Valley Hospital. 200 Castle Creek Road. Aspen. CO 81 61 1. *Division of Physical Therapy, University of New England, 11 Hills Bend Road, Biddeford, ME 04005. 5 Division of Physical Therapy, Department of Medical Allied Health Professions, Medical School Wing E 222 H, University of North Carolina at Chapel Hill. Chapel Hill. NC 27514. 11 Athletic Trainer, Physical Therapist, Sports Medicine Program, Student Health, University of North Carolina at Chapel Hill. Chapel Hill, NC 27514. One method now commonly in use is requiring athletes to wear preventative knee braces. Although these braces are thought to be capable of limiting the number of injuries occurring in a contact sport such as football, in which 75% of all sports related knee injuries occur,16 this contention has neither been supported nor refuted by research. Two preventative knee braces frequently used by college and professional football teams are the Anderson Knee Stabler and the McDavid Knee Many athletes complain that their speed and cutting agility are reduced when wearing these braces, and that running in the braces is awkward. While there have been several studies done to determine if athletes are able to perform agility skills while wearing these braces,'o, 14.18.19.28 whether or not these braces interfere with an athlete's normal running gait has not been documented. This is an important concern, since any alteration in normal running gait could lead to a biomechanical imbalance" and perhaps result in a variety of chronic overuse injuries.16 Therefore, the purpose of this study

JOSPT January 7988 ANDERSON KNEE STABLER/MCDAVID KNEE GUARD 255 was to document and compare gait patterns in individuals as they ran on a treadmill at 4 and 8 mph wearing an Anderson Knee Stabler (Omni Scientific, Inc., Martinez, CA), a McDavid Knee Guard (McDavid, Clarendon Hills, IL), and no brace. REVIEW OF LITERATURE When studying gait patterns of individuals whose running styles may be altered by wearing a preventative knee brace, it is important to briefly review what is known about normal running biomechanics, the difference between overground and treadmill running, and preventative knee braces. Several researchers have documented the biomechanical components of running.2.6, 13.17.26 Mann and Hagy17 studied the run- ning gait of 13 competitive runners and found that step length, cadence, and velocity increased as the speed of gait increased. They also observed that the range of motion at the hip, knee, and ankle joints increased as the gait speed increased. Sinning and ~orsyth,~' testing seven competitive college runners, reported findings similar to Mann and Hagy when they measured hip, knee, and ankle flexion and extension. Using electrogoniometers and measuring hip, knee, and ankle range of motion as the subjects ran on a treadmill at speeds ranging from 6.8-14.8 mph, these researchers found that maximum hip flexion increased as running speeds increased, and hip extension did not vary significantly. Maximum knee flexion also increased as gait speed increased, while maximum knee extension decreased slightly. Ankle joint range of motion remained constant at the various running speeds. In a similar study, Gollnick and Karpovich7 reported slightly different measurements; Sinning and ~orsyth~~ stated that these slight differences might be caused by the different running speeds used in the two studies. Sinning and Forsyth also measured changes in angular velocity for the hip, knee, and ankle joints. The average angular velocities of hip extension and flexion increased linearly as the running speeds increased. The angular velocity of knee extension increased more than knee flexion velocity did during the swing phase of the subjects' running gaits, while the changes in angular velocity for extension and flexion of the ankle were similar. Some researchers have questioned whether treadmill gait is equivalent to gait on a level surface.27 Research comparing walking and running on a level walkway versus on a treadmill has yielded conflicting concl~sions.~"~~'~~~~~~ Nelson et a1.,21 studying the biomechanics of overground and treadmill running in male athletes, documented that during treadmill running the support phase tended to be longer and the flight phase shorter in comparison to overground running. Elliott and Blanksby5 noted that for both the male and female subjects they tested while the subjects jogged at an average velocity of 5.4 m/sec, stride length decreased, stride rate increased, and the flight phase period decreased when running on a treadmill as compared to overground. Dal Monte et reported a similarity between track and treadmill running in trained runners, although they did document a reduction in the vertical displacement of the center of gravity during treadmill running. In light of this controversy, and until more conclusive research is presented, treadmill studies such as this should be used with caution to draw conclusions about overground running. There seems to be little scientific research published that thoroughly investigates preventative knee braces. The majority of brace studies are unpublished statistical comparisons of individual college football team injuries before and after these teams began using preventative knee Most of these studies were conducted over a short period of time and are not statistically significant.12 Anderson et al.' identified nine professional football players who had histories of medial knee joint injuries, who subsequently used the Anderson Knee Stabler for 2 years and who had no recurrence of knee injury. However, five of these athletes played in three or fewer games, thereby markedly reducing their chances of injury. During the 1980-1 981 football season at Oregon State University, 28 knee injuries were reported. The following season the team's linemen began wearing the Anderson Knee Stabler and only four injuries o~curred.~' At the University of North Carolina at Chapel Hill, there was no significant difference in the number of medial collateral ligament injuries sustained by football players during the years the players wore braces compared to the years they did not wear braces; however, there was a decrease in the number of MCL injuries requiring surgery during the years the athletes were braced.12 Some research has been done to study the effects of knee braces on athletic performance. Houston and Gormann," studying the effects of three derotational braces on isokinetic perform-

256 VAN HORN ET AL JOSPT Vol. 9, No. 7 ance measures, found that peak torque values during extension were higher for subjects not wearing a brace. In another aspect of this study, the authors found higher blood lactate levels in subjects wearing braces as they rode a bicycle ergometer; the authors speculated that these higher levels might be present because of the braces interfering with circulation and oxygen delivery. In contrast to results reported by Houston and Gormann," Hansen8 found that measurements of strength, power and endurance in quadriceps and hamstring muscles were not significantly different when subjects wore an Anderson Knee Stabler as compared to measurements taken when the subjects were not braced. Studies performed by MayIg and Martindale1* reported no significant difference in subjects' performance times during a run through a 40-yard maze with and without McDavid Knee Guards. Hawkins'' found no significant difference in a subject's running speed in a 30-yard dash with and without braces; Tore~celli~~ did find a significant decrease in a subject's forward running speed while wearing a knee brace. Finally, the relative strength of preventative knee braces also warrants study. Results in a study by PaulosZ3 suggest that intact ligaments in cadaver knees are stronger than a brace when an external static force is applied. It is clear that preventative knee braces need to be studied further. METHOD Subjects Fifteen healthy male students at the University of North Carolina at Chapel Hill volunteered to participate in this study. Age range was 18 to 35 years old with a mean age of 25. No subject had a history of ligamentous knee surgery. Materials A 16-mm Bolex (Bolex International, Yverdon, Switzerland) motor-drive camera was used to film the subjects. The camera was supported by a tripod and was perpendicular to the plane of motion. It was set at a speed of 64 frames/sec with an f-stop of 4.0 and an exposure time of 0.0025 seconds. Data gathered were measured from the developed film with a Vanguard Motion Analyzer (Vanguard Instrument Corp., Melville, NY). Data measurements were taken twice to ensure accu- racy. Angular measurements varied one or two degrees between the two measurements, and this measurement error should be considered when examining the results of this study. The Anderson Knee Stabler, a double hinged single sided metal brace, and the McDavid Knee Guard, a single hinged brace of an impact resistant plastic, were used in this study. The braces, designed to fit people of various heights, were new and on loan from their manufacturer. Procedure Subject Preparation Each subject signed a consent form and answered a background questionnaire prior to participating in this study. Each subject was allowed 5 minutes to stretch and warm up his muscles prior to running. Adhesive tape markers were placed over the lateral hip, knee, and ankle joints of each subject's left leg. A 1-inch diameter blackened dot was placed on the adhesive tape to mark the center of these joints. The subjects wore the braces on both knees. Adhesive tape markers were placed on the center point of the braces that corresponded with the center of the knee joint. During a pilot study it was determined that the braces did not slip and that they remained in place during all of the trials. The braces did not slip during the actual study, either. Therefore, the center marker on the braces was used to represent the center of the knee joint for angular measurements when the knee joint was unable to be visualized. Testing Procedures The 15 male volunteers were filmed while they ran on a treadmill at 4 and 8 mph. These speeds are comparable to a fast walk and a fast run, respectively. Subjects ran a total of six trials, three at 4 mph and three at 8 mph with an Anderson Knee Stabler, a McDavid Knee Guard, and without a brace. The order in which the subject ran the trials was balanced so as to limit the influence of fatigue or a training effect on any of the conditions. The subjects were filmed for approximately 10 sec at each speed for each of the three conditions. Analysis Specific components of each subject's running pattern were measured twice from one gait cycle during each trial condition. These measurements included: 1) joint angle measurements of the left hip, knee, and ankle during heel strike; 2) joint

JOSPT January 1988 ANDERSON KNEE STABLER/MCDAVID KNEE GUARD 257 angle measurements of the left hip, knee, and ankle during push off; 3) maximal velocities of knee extension and flexion; 4) maximal extension and maximal flexion of the left hip, knee, and ankle. Statistical Analysis A multivariate-anova was used to compare the statistical differences between running without a brace, running with an Anderson Knee Stabler, and running with a McDavid Knee Guard at 4 and 8 mph. The trial condition of running without a brace served as the control data to compare to the two experimental conditions of running with the braces. RESULTS Speed and Brace lnteraction A statistical significance was shown for the speed and brace interaction of three measured gait variables (Table 1). These gait components included the maximal ankle plantarflexion angle, hip flexion angle at foot strike, and knee extension angle at foot strike. The maximal ankle plantarflexion angle was less when subjects wore the Anderson Knee Stabler and jogged on the treadmill at 4 rnph compared to when subjects jogged without a knee brace or with the McDavid Knee Guard. The mean value for plantarflexion at 4 mph for the Anderson brace condition was 12.3', 11.2' for the McDavid brace condition, and 11.go for no brace condition. There was a statistically significant increase in hip flexion angle at foot strike when subjects wore the McDavid Knee Guard and ran on the treadmill at 8 mph compared to when subjects ran without a brace or with the Anderson Knee Stabler. The mean value for hip flexion angle at foot strike at 8 rnph was 28.2O for the McDavid Knee Guard trial, 26.7' for the Anderson Knee Stabler trial, and 26.2' for the control trial. There was also a statistically significant decrease in knee extension angle at foot strike when subjects wore the McDavid Knee Guard and ran on the treadmill at 8 mph compared to when subjects ran without a brace or with the Anderson Knee Stabler. The mean for knee extension at foot strike at 8 mph for the McDavid brace condition was 20.3' flexion, for the Anderson brace condition was 18.4O flexion, and 17.8' flexion for the control condition. TABLE 1 Significant mean values Variable Speeds Means (O) (4 mph) (8 mph) Maximal hip extension angle -3.9' -6.1' Maximal hip flexion angle 26.8 35.7 Maximal knee flexion angle 80.6 101.7 Hip flexion angle at foot strike 17.4 18.8 Knee extension angle at push-off -23.2' -1 7.6' Knee extension velocity (radlsec) 8.1 10.9 Knee flexion velocity (radlsec) 6.7 9.5 Variable Braces Means Control Anderson McDavid Maximal knee extension angle -1 3.4* -1 4.9' -1 5.3' Maximal knee flexion angle 90.9 90.9 92.7 Hip flexion angle at foot strike 23.2t 23.7 24.4t Knee extension angle at foot -1 7.3'7-1 8.3* -1 8.7*t strike Knee extension angle at push-off -1 9.2't -20.7' -21.3't Brace x Speed lnteraction Means Variable Control Anderson McDavid Maximal ankle dorsiflexion 4 mph: 11.9 12.3t 11.2 Maximal ankle plantarflexion 8 mph: 8.9 9.8 7.7 Hip flexion angle at foot 4 mph: 20.2 20.7 20.5 strike 8 mph: 26.2 26.7 28.2t Knee extension angle 4 mph: 16.7 18.3 17.2 (-) Indicates that the joint did not reach complete extension. t Statistically significant means. Speed The results of the multivariate analysis of variance for speed show a statistical significance for 7 of the 14 gait components (Table 1). These variables are maximal hip extension angle, maximal hip flexion angle, maximal knee flexion angle, hip flexion angle at foot strike, knee extension angle at push-off, maximal angular velocity of knee extension, and maximal angular velocity of knee flexion. Six of these gait variables had greater mean values at 8 rnph as compared to 4 mph. The mean value for the maximal hip extension angle was 3.85' flexion at 4 mph and 6.1 ' flexion at 8 mph. The mean measurement for knee flexion angle at 4 rnph was 80.6' and at 8 mph was 101.7'. The mean for the maximal hip flexion angle measured 26.8' at 4 mph and 35.7' at 8 mph. The mean value for the hip flexion angle at foot strike was 17.4' at 4 rnph and 18.8' at 8 mph. The mean for maximal angular velocity of knee extension was 8.05 radianslsec during 4 rnph and 10.9

258 VAN HORN ET AL JOSPT Vol. 9, No. 7 radianslsec at 8 mph. The mean value for maximal angular velocity of knee flexion was 6.7 radians/ sec at 4 rnph and 9.5 radianslsec at 8 mph. The knee extension angle at push off at 4 rnph was less than the same angle measured at 8 mph. The mean measurement for this gait component was 23.17' flexion at 4 rnph and 17.6' flexion at 8 mph. Brace There was a statistically significant difference shown for the brace variable in five of the gait components measured (Table 1).These gait components are maximal knee extension angle, maximal knee flexion angle, hip flexion and knee extension angles at foot strike, and knee extension angles at push-off. There was less maximal knee extension when subjects wore the Anderson Knee Stabler or the McDavid Knee Guard as compared to when the braces were not worn. The mean value for the maximal knee extension angle when subjects did not wear a brace was 13.4' flexion. The mean measurements for the Anderson Knee Stabler condition was 14.86' flexion and for the McDavid Knee Guard condition was 15.3' flexion. There was an increased in hip flexion at foot strike and a decrease in knee extension at foot strike and at push-off when the McDavid Knee Guard was worn compared to when a brace was not worn. There was not a statistically significant difference between the mean values of these gait components for the Anderson Knee Stabler condition and the other two conditions. The mean value for hip flexion at foot strike for the McDavid brace condition was 24.4', for the no brace condition was 23.3', and for the Anderson Knee Stabler condition was 23.7'. The mean of the knee extension measurements at foot strike with the McDavid Knee Guard was 18.7' flexion, 18.3' flexion with the Anderson Knee Stabler, and 17.3' flexion without a brace. The mean value of the knee extension angle at push-off for the McDavid knee brace condition measured 21.3' flexion, for the Anderson Knee-Stabler condition was 20.7' flexion and without a brace condition was 19.2' flexion. The measurements for maximal knee flexion were statistically greater when the McDavid Knee Guard was worn compared to when the Anderson Knee Stabler was worn. There was also a statistically significant difference noted when four of the gait components were measured at varying speeds and under braced and nonbraced conditions, and when taking into account the speedbrace interaction. These variables include the maximal ankle dorsiflexion angle (Table I), ankle dorsiflexion angle at a foot strike, and ankle flexion angle at push-off (Table 1) and hip extension angle at push-off. DISCUSSION Speed This present study found a statistically significant difference in subjects' maximal hip extension angle and maximal hip flexion angle, hip flexion angle at foot strike and hip extension angle at push-off, and maximal angular velocity of knee extension and flexion measurements at 4 and 8 mph. These results are in agreement with other studies that compared gait pattern changes at different velocities. Both Mann and Hagy's17 and Sinning and F~rsyth's*~ research studies reported a slight decrease in knee extension at the faster speeds. The present study showed a statistically significant decrease in knee extension at push-off during 4 rnph compared to 8 mph. Eleven of the 15 subjects had more knee flexion at push-off at 4 rnph compared to this measurement at 8 mph. This reported difference between Sinning and Forsyth's and Mann and Hagy's studies and this present research study may be caused by the differences in the running speeds observed. Sinning and Forsyth's treadmill study did not report subjects' gait patterns at a speed comparable to 4 mph. Therefore, the present study's reported decrease in knee extension may be accurately characteristic of subjects' gait pattern of 4 rnph on a treadmill. Sinning and Forsyth reported that the range of motion of the ankle did not significantly change at the different velocities. Ankle range of motion during the present study did not statistically differ when subjects jogged at 4 rnph or ran at 8 mph. The present study observed a statistically significant increase in the maximal velocities of knee extension and knee flexion at 8 rnph when compared to these same measurements at 4 mph. Sinning and Forsyth also reported a significant increase in the velocities of knee extension and flexion at the faster speeds.

JOSPT January 1988 ANDERSON KNEE STABLER/MCDAVID KNEE GUARD 259 Braces This study showed a decrease in maximal knee extension when the subjects wore the Anderson Knee Stabler and the McDavid Knee Guard in comparison to when a brace was not worn. A decrease in knee extension was also seen at foot strike and push off when the McDavid Knee Guard was worn compared to when a brace was not worn. This decrease in complete knee extension during a subject's gait pattern may have been directly influenced by the wearing of the braces. Also noted was an increase in hip and knee flexion angles at foot strike when the McDavid Knee Guard was worn compared to when no brace was worn. This increase may be due to a subject compensating for feeling uncomfortable while wearing the braces by increasing his hip flexion to lower his center of gravity, thus increasing his stability. Speed and Brace Interaction Three gait components had a statistically significant difference for speed and brace interaction. These components were the maximal ankle plantarflexion angle, and hip flexion and knee extension angles at foot strike. The maximal plantarflexion angle was less when subjects wore the Anderson Knee Stabler and jogged on the treadmill at 4 rnph compared to when subjects jogged without a brace or with the McDavid Knee Guard. There was also a statistically significant increase in hip flexion and a decrease in knee extension angles at foot strike when subjects ran at 8 rnph wearing the McDavid Knee Guard when compared to running with the Anderson Knee Stabler or without a brace. The mean value for hip flexion at foot strike at 8 rnph for the McDavid knee brace condition was 2.0' greater than the mean value for the Anderson knee brace condition, and 1.5' greater than the control condition. There was a 2.5' decrease in knee extension at foot strike at 8 rnph when subjects ran with the McDavid Knee Guard compared to when a brace was not worn. Knee extension at foot strike decreased 1.go when the McDavid Knee Guard was worn compared to when the Anderson Knee Stabler was worn. These measurement differences are minimal and may not alter an athlete's gait enough to affect performance. However, when looking at high level performance individuals even small alterations in gait patterns may have implications for athletic performance. Further research may be able to provide an explanation for these results. CONCLUSIONS AND RECOMMENDATIONS The following conclusions are based on the results of this study: 1) There was an increase in hip and knee flexion when the running velocity increased from 4 to 8 rnph (p < 0.05). 2) There was an increase in knee extension and flexion velocities when the treadmill speed increased from 4 to 8 mph. 3) There was a statistically significant decrease in the maximal knee extension when subjects ran with the Anderson Knee Stabler and the McDavid Knee Guard compared to when subjects ran without a preventative knee brace (p < 0.05). 4) There was a statistically significant decrease in the hip flexion angle at foot strike when the McDavid Knee Guard was worn compared to when a brace was not worn (p < 0.05). 5) There was a statistically significant decrease in the knee extension angles at foot strike and push-off when subjects wore the h4cdavid Knee Guard compared to when a brace was not worn (p < 0.05). 6) There was a statistically significant increase in knee flexion when the subjects ran wearing the McDavid Knee Guard compared to when subjects ran wearing an Anderson Knee Stabler (p < 0.05). 7) There was statistically significant decrease in ankle plantarflexion during the Anderson Knee Stabler trial at 4 rnph compared to the McDavid Knee Guard and the control trials (p < 0.05). 8) There was a statistically significant increase in hip and knee flexion angles at foot strike when subjects ran at 8 rnph and wore the McDavid Knee Guard compared to when subjects ran at 8 rnph and did not wear a brace or wore the Anderson Knee Stabler (p < 0.05). The results of this research study indicate that there are only minimal differences in how the Anderson Knee Stabler and the McDavid Knee Guard affect subjects' gait patterns compared to when a brace is not worn. It was also observed that the braces did not differ significantly when compared to each other in respect to their individual effects on subjects' gait patterns. Further investigations need to be made on these preventative knee braces to include: 1 ) Cinematographic analysis of the biomechanics of running overground while subjects wear the preventative knee braces.

260 VAN HORN ET AL JOSPT Vol. 9, No. 7 2) Running subjects that are familiar with and have previously used the preventative knee braces. These subjects could include college and professional football players. 3) Measurements of subject's stride length while wearing braces. A shortened stride length would increase an athlete's running speed, by requiring him to take more steps. This finding would have an impact on an athlete's performance. Based on the results of this study, no clear superiority exists between the two braces' effects on the gait characteristics measured. Therefore, other parameters should be used to make equipment purchase decisions. REFERENCES 1. Anderson G, Seman SC, Rosenfeld RT: The Anderson Knee Stabler. Phys Sports Med 7:125-127,1979 2. Atwater AE: Cinematographic analysis of human movement. Exerc Sports Sci Rev 1 :217-258,1973 3. Cavanagh PR, Gregor JR: Knee joint torque during the swing phase of normal treadmill walking. J Biomech 8:337-344. 1975 4. Dal Monte A, Fucci S, Manoni A: The treadmill used as a trainer and simulator instrument in middle and long distance running. Med Sport 8: Biomechanics 111:359-363,1973 5. Elliot BC. Blanksby BA: A cinematographic analysis of overground and treadmill running by males and females. Med Sci Sports Exerc 12:257-261.1980 6. Gilbert JA, Maxwell GM, McElhaney JH, Clippenger FW: A system to measure the forces and moments at the knee and hip during level walking. J Orthop Res 2:281-288.1984 7. Gollinick PD, Karpovich PV: Electrogoniometric study of locomotion and of some athletic movements. Red Q 35:357-369.1964 8. Hansen BL: The effects of the Anderson Knee Stabler on strength, power and endurance of the quadriceps and hamstring muscle groups. Unpublished research paper, University of Colorado, Boulder. 1981 9. Hansen BL, Ward JC, Diehl RC: The preventative use of the Anderson Knee Stabler in football. Phys Sports Med 1375-81, 1985 10. Hawkins HA: Effect of the Arco Knee Guard on running and speed. Unpublished manuscript, lndiana State University, Terre Haute, 1977 11. Houston ME, Gormann PH: Leg muscle performance of athletes with and without knee support braces. Arch Phys Med Rehabil 6431-432,1982 12. Hunter SL: Effects of the Anderson Knee Stabler on injury prevention. Unpublished manuscript, University of North Carolina at Chapel Hill, 1983 13. James SL, Brubaker CE: Running mechanics. JAMA 221:1014-1016,1972 14. Johnson A: Effects of wearing a knee guard on agility. Unpublished manuscript, lndiana State University. Terre Haute. 1969 15. Kaplan EB: Some aspects of functional anatomy of the human knee joint. Clin Orthop 2318-29, 1962 16. Klafs DE, Arnheim DD: Modem Principles of Athletic Training. Ed 4. St. Louis: Mosby Co, 1977 17. Mann RA, Hagy J: Biomechanics of walking, running and sprinting. Am J Sports Med 8345-350.1980 18. Martindale A: Effects of Arm Knee Guard on agility. Unpublished manuscript, lndiana State University, Terre Haute, 1977 19. May TL: Effect of Arco Knee Guard on agility determined by the forty yard maze test. Unpublished manuscript, lndiarla State University, Terre Haute, 1971 20. Moore M: Preventative knee braces catching on in football. Phys Sports Med 11:21,1983 21. Nelson RC. Dillman CJ, Lagasse P, Bickett P: Biomechanics of overground versus treadmill running. Med Sci Sports Exerc 4223-240,1972 22. O'Donaghue D: Treatment of Injuries to Athletics, Ed 3, pp 29-31. Philadelphia: WB Saunders Co, 1976 23. Paulos L: Lateral knee braces in football: do they prevent injury? Phys Sports Med 14:119-126,1986 24. Powell JW: Pattern of knee injuries associated with college football 1975-1 982. Ath Train 3:104-109, 1985 25. Rall KL, McElroy GL, Keats TE: A study of long term effects of football injury to the knee. Missouri Med 61:435-438,1964 26. Ryan AJ: Knee braces to prevent injuries in football: a round table. Phys Sports Med 14:108-118,1986 27. Sinning WE, Forsyth HL: Lower limb actions while running at different velocities. Med Sci Sports Exerc 2:28-34, 1970 28. Smidt GL: Biomechanical analysis of knee flexion and extension. J Biomech 6:79-92,1973 29. Torescelli TA: A comparison of the effects of selected preventative knee braces on running speed and agility. Unpublished master's thesis, University of North Carolina at Chapel Hill. 1985 30. Von lngen Schenau GT: Some fundamental aspects of the b ii mechanics of the overground versus treadmill running. Med Sci Sports Exerc 4233-240,1972