<%BANNER%>

Control Strategies during Gait Termination

Permanent Link: http://ufdc.ufl.edu/UFE0022703/00001

Material Information

Title: Control Strategies during Gait Termination Elucidating The Mechanisms of Ankle Instability
Physical Description: 1 online resource (88 p.)
Language: english
Creator: Inamdar, Amruta
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: ankle, gait, instability, muskuloskeletal, neuromuscular, orthopedic, termination
Applied Physiology and Kinesiology -- Dissertations, Academic -- UF
Genre: Applied Physiology and Kinesiology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Chronic ankle instability (AI) is a functionally debilitating condition that occurs after the first episode of injury to the ligaments of the ankle joint. Both feedback and feedforward mechanisms of neuromuscular control have been reported to be affected in people suffering from AI. However no previous researchers have examined these deficits in the same population. In this study we used the gait termination model to challenge both the feedback and feedforward mechanisms of neuromuscular control in the same population of subjects suffering from AI and compared the results to controls. Planned gait termination was used to reveal deficits in feedforward neuromuscular control whereas unplanned gait termination was used to reveal the deficits in the feedback neuromuscular control. Thus the purpose of this study was to reveal the type of alterations in neuromuscular control that exist in AI. Force, stability and EMG were used as outcome measured to reveal these deficits. The study was a single session, single subject mixed model design. Twenty participants were recruited in the AI group (age 20.2?1.2 years, height 169.8?9.7 cm and mass 74.2 ?20.2 kg) and 20 participants were recruited in the control group (age 20.4? 1.6 years, height 164.3 ?7.9 cm and mass 64.2 ?10.6 kg). The Ankle Injury Questionnaire was used to screen the participants. Each eligible participant performed both unplanned and planned gait termination trials. Normal walking trials were interspersed to avoid anticipation of stop. Ten trials of both planned and unplanned gait termination were captured with each limb serving as a lead limb in 5 trials. Ground reaction forces were collected using two force plates (0.4 m x 0.6 m) (Type 4060-10 Bertec Corporation, Columbus, Ohio). The GRF data along X, Y and Z axes were use to calculate dynamic postural stability index (DPSI) and its subcomponents (APSI, MLSI and VSI). A KONIGSBERG T-42AL-8T (Konigsberg Instruments Inc, California) telemetric electromyography unit was used to record muscle activity of Tibialis Anterior (TA), Soleus (Sol) and Gluteus Medius (GM) using bipolar 1-mm x 10-mm Ag/AgCl surface electrodes. A 3-way MANOVA (2:group x 2:limb x 3:condition) with repeated measures on the last factor revealed significance differences in propulsive force and braking force. Both propulsive force and braking force in the AI group were higher than the control group during unplanned gait termination and the AI group relied more on lead limb strategy during gait termination. A 4-way MANOVA (2:group x 2:limb x 6:phases x 2:condition) with repeated measures on the last factor revealed significance differences in average amplitude of TA, Sol and GM. The average amplitude of TA for the involved limb was less than the uninvolved limb during all four subphases of stance. The average amplitude of TA, Sol and GM was higher during unplanned gait termination than during planned gait termination. Average amplitude of Sol in the AI group was less than that in the control group. This finding failed to explain the result that the AI group generates a higher braking force and suggested that another muscle might be responsible for producing the braking force in the AI group during gait termination. A 3-way MANOVA (2:group x 2:limb x 2:condition) with repeated measures on the last factor revealed higher DPSI and APSI scores in the AI group and involved limb during unplanned gait termination than during planned gait termination. All these finding suggest that feedback and feedforward deficits of neuromuscular control coexist in AI.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Amruta Inamdar.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Tillman, Mark D.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022703:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022703/00001

Material Information

Title: Control Strategies during Gait Termination Elucidating The Mechanisms of Ankle Instability
Physical Description: 1 online resource (88 p.)
Language: english
Creator: Inamdar, Amruta
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: ankle, gait, instability, muskuloskeletal, neuromuscular, orthopedic, termination
Applied Physiology and Kinesiology -- Dissertations, Academic -- UF
Genre: Applied Physiology and Kinesiology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Chronic ankle instability (AI) is a functionally debilitating condition that occurs after the first episode of injury to the ligaments of the ankle joint. Both feedback and feedforward mechanisms of neuromuscular control have been reported to be affected in people suffering from AI. However no previous researchers have examined these deficits in the same population. In this study we used the gait termination model to challenge both the feedback and feedforward mechanisms of neuromuscular control in the same population of subjects suffering from AI and compared the results to controls. Planned gait termination was used to reveal deficits in feedforward neuromuscular control whereas unplanned gait termination was used to reveal the deficits in the feedback neuromuscular control. Thus the purpose of this study was to reveal the type of alterations in neuromuscular control that exist in AI. Force, stability and EMG were used as outcome measured to reveal these deficits. The study was a single session, single subject mixed model design. Twenty participants were recruited in the AI group (age 20.2?1.2 years, height 169.8?9.7 cm and mass 74.2 ?20.2 kg) and 20 participants were recruited in the control group (age 20.4? 1.6 years, height 164.3 ?7.9 cm and mass 64.2 ?10.6 kg). The Ankle Injury Questionnaire was used to screen the participants. Each eligible participant performed both unplanned and planned gait termination trials. Normal walking trials were interspersed to avoid anticipation of stop. Ten trials of both planned and unplanned gait termination were captured with each limb serving as a lead limb in 5 trials. Ground reaction forces were collected using two force plates (0.4 m x 0.6 m) (Type 4060-10 Bertec Corporation, Columbus, Ohio). The GRF data along X, Y and Z axes were use to calculate dynamic postural stability index (DPSI) and its subcomponents (APSI, MLSI and VSI). A KONIGSBERG T-42AL-8T (Konigsberg Instruments Inc, California) telemetric electromyography unit was used to record muscle activity of Tibialis Anterior (TA), Soleus (Sol) and Gluteus Medius (GM) using bipolar 1-mm x 10-mm Ag/AgCl surface electrodes. A 3-way MANOVA (2:group x 2:limb x 3:condition) with repeated measures on the last factor revealed significance differences in propulsive force and braking force. Both propulsive force and braking force in the AI group were higher than the control group during unplanned gait termination and the AI group relied more on lead limb strategy during gait termination. A 4-way MANOVA (2:group x 2:limb x 6:phases x 2:condition) with repeated measures on the last factor revealed significance differences in average amplitude of TA, Sol and GM. The average amplitude of TA for the involved limb was less than the uninvolved limb during all four subphases of stance. The average amplitude of TA, Sol and GM was higher during unplanned gait termination than during planned gait termination. Average amplitude of Sol in the AI group was less than that in the control group. This finding failed to explain the result that the AI group generates a higher braking force and suggested that another muscle might be responsible for producing the braking force in the AI group during gait termination. A 3-way MANOVA (2:group x 2:limb x 2:condition) with repeated measures on the last factor revealed higher DPSI and APSI scores in the AI group and involved limb during unplanned gait termination than during planned gait termination. All these finding suggest that feedback and feedforward deficits of neuromuscular control coexist in AI.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Amruta Inamdar.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Tillman, Mark D.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022703:00001


This item has the following downloads:


Full Text

PAGE 1

1 CONTROL STRATEGIES DURING GAI T TERMINATION: ELUCIDATING THE MECHANISMS OF ANKLE INSTABILITY By AMRUTA D. INAMDAR A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Amruta D. Inamdar

PAGE 3

3 To my parents who taught me to dream and my husband who gave me the strength and support to fulfill those dreams.

PAGE 4

4 ACKNOWLEDGMENTS As I look back at m y journey and accomplishmen ts during the last two years, I realize and acknowledge the efforts and suppor t of my professors, family and friends. Along the way they have all become an important part of my journey, and an important part of my life. Primarily I wish to thank Dr. Mark Tillma n, my advisor, for giving me a chance to complete my graduate education. I am truly grat eful to him for providing me the opportunity to fulfill my goals. I appreciate his continuous support and interest towards my thesis and coursework. Under the able guidance of Dr. Tillman, I have grown academically, professionally and more importantly as an individual. I am also grateful to Dr. Erik Wikstrom for his guidance during the course of my graduate study, for supporting and advising me through different stages of the research project. I wish to thank Dr. Hass for extending his expertise durin g the course of my thesis. I thank all my committee members towards the successful completion of my thesis. I express my humble gratitude to my parent s, whose vision and trust imbibed in me the confidence to achieve my goals. Most importantly I am thankful to my loving husband for making my dream and vision a rea lity. His love and belief in me inspired me to strive and achieve the best. I truly believe that he forms the essence and the origin of my happiness. Thank you, Devendra. I also want to thank my dear fr iends, Saurabh Wagh and Srikant Vallabhajosula, for their continuous support and help during these two years.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES.........................................................................................................................9 ABSTRACT...................................................................................................................................10 CHAP TER 1 INTRODUCTION..................................................................................................................12 Chronic Ankle Instability.......................................................................................................12 Hypotheses..............................................................................................................................15 Forces..............................................................................................................................15 Propulsive force........................................................................................................ 15 Propulsive time......................................................................................................... 15 Braking force............................................................................................................16 Braking time.............................................................................................................16 Postural stability....................................................................................................... 16 Muscle Activation...........................................................................................................17 Distal muscles (TA and Soleus)............................................................................... 17 Proximal muscle (Gluteus Medius).......................................................................... 17 Significance of the Study........................................................................................................17 2 LITERATURE REVIEW.......................................................................................................19 Introduction................................................................................................................... ..........19 The Talocrural Joint........................................................................................................... .....19 Tibiofibular Ligament..................................................................................................... 20 Medial and Lateral Ligam ent Complexes.......................................................................20 Muscular Reinforcement of the Ankle Joint................................................................... 22 Chronic Ankle Instability.......................................................................................................23 Mechanical Instability..................................................................................................... 23 Pathologic laxity....................................................................................................... 24 Athrokinematic insufficiencies................................................................................25 Degenerative changes in th e ankle joint and synovium ........................................... 25 Functional Instability.......................................................................................................26 Feedback and Feedforward Mechan ism s for Postural Control............................................... 27 Dynamic Postural Stability Index........................................................................................... 29 Human Gait.............................................................................................................................31 Introduction................................................................................................................... ..31 Gait Initiation...................................................................................................................31

PAGE 6

6 Rhythmic Walking........................................................................................................... 32 Gait Termination............................................................................................................. 33 Quantifying Gait Termination................................................................................................ 33 Factors Affecting Gait Termination........................................................................................ 34 Stimulus Delay................................................................................................................ 35 Stimulus Probability........................................................................................................ 35 Velocity of Gait............................................................................................................... 36 Gait Termination as a Model for Elucidation of the Feedforward and Feedback Mechanism s in Ankle Injury............................................................................................... 37 3 METHODS.............................................................................................................................39 Recruitment.................................................................................................................... .........39 Inclusion Criteria.............................................................................................................40 Exclusion Criteria............................................................................................................40 Instrumentation................................................................................................................ .......40 Force Platform.................................................................................................................40 The Motion Capture System............................................................................................ 41 Electromyography (EMG)............................................................................................... 42 Testing and Subject Preparation............................................................................................. 43 Static Trial.......................................................................................................................44 Procedure for Testing Unplanned Gait Termination....................................................... 44 Procedure for Testing Pla nned Gait Term ination............................................................ 45 Procedure for calculation of the DPSI............................................................................. 45 Statistics...........................................................................................................................46 Statistics for Kinetic and Kinematics.............................................................................. 46 Statistics for Stability......................................................................................................46 Statistics for Electromyography (EMG)..........................................................................47 4 RESULTS...............................................................................................................................48 Demographics.........................................................................................................................48 Velocity....................................................................................................................... ............48 Relative Time of Stimulus Delivery....................................................................................... 49 Force.......................................................................................................................................49 Propulsive Force............................................................................................................... ......50 Propulsive Time............................................................................................................... 51 Braking Force..................................................................................................................51 Braking Time...................................................................................................................52 Stability...................................................................................................................... .............52 Anteroposterior Stability Index (APSI)........................................................................... 52 Mediolateral Stability Index (MLSI)............................................................................... 53 Vertical Stability Index(VSI)..........................................................................................53 Dynamic Postural Stability Index (DPSI)....................................................................... 54

PAGE 7

7 Electromyography (EMG)......................................................................................................54 Tibialis Anterior..............................................................................................................54 Soleus..............................................................................................................................56 Gluteus Medius................................................................................................................57 5 DISCUSSION.........................................................................................................................60 Gait Velocity...........................................................................................................................60 Relative Timing of Stimulus Delivery .................................................................................... 61 Forces......................................................................................................................................62 Propulsive Force.............................................................................................................. 62 Braking Force..................................................................................................................65 Propulsive Time and Braking Time................................................................................ 67 Stability...................................................................................................................... .............67 Electromyography (EMG)......................................................................................................69 Distal Muscles (TA and Soleus)...................................................................................... 69 Proximal Muscle (Gluteus Medius)................................................................................. 72 Synopsis of the Results....................................................................................................73 Limitations.................................................................................................................... ..........76 Conclusion..............................................................................................................................77 APPENDIX A ANKLE INJURY QUESTIONNAIRE.................................................................................. 78 B STATISTICAL TABLE.........................................................................................................79 LIST OF REFERENCES...............................................................................................................82 BIOGRAPHICAL SKETCH.........................................................................................................88

PAGE 8

8 LIST OF TABLES Table page 3-1 Electrode placement:....................................................................................................... ...44 4-1 Velocity (m/s) (Mean SD).............................................................................................. 49 4-2 Relative timing of stimulus delivery(s) (Mean SD) ....................................................... 49 4-3 Propulsive Force (N) (Mean SD).................................................................................... 50 4-4 Propulsive Time (s) (Mean SD)...................................................................................... 51 4-5 Braking Force (N) (Mean SD)........................................................................................51 4-6 Braking Time (s) (Mean SD).......................................................................................... 52 4-7 Anteroposterior Stability Index(APSI) (Mean SD). ....................................................... 53 4-8 Mediolateral Stability Index(MLSI) (Mean SD) ............................................................ 53 4-9 Vertical Stability Index (VSI) (Mean SD)...................................................................... 53 4-10 Dynamic Postural Stability Index(DPSI)(Mean SD)...................................................... 54 4-11 Average Amplitude (TA) (Mean SD)............................................................................. 55 4-12 Average Amplitude (Soleus) (Mean SD)....................................................................... 57 4-13 Average Amplitude (GM) (Mean SD)............................................................................59 B-1 Statistics for force..............................................................................................................79 B-2 Statistics for stability..........................................................................................................80 B-3 Statistics for EMG (Soleus)...............................................................................................80 B-4 Statistics for EMG (TA).................................................................................................... .80 B-5 Statistics for EMG (GM)................................................................................................... 81

PAGE 9

9 LIST OF FIGURES Figure page 1-1 Nomenclature of the limbs during gait termination........................................................... 14 2-1 Lateral ligaments of the ankle joint................................................................................... 21 2-2 Medial ligaments of the ankle joint...................................................................................22 2-3 Left hip anterior velo city position plot for young and ol der adults during two-step stopping..............................................................................................................................36 3-1 Placement of the cameras and the force plates.................................................................. 41 3-3 Force curve with sub pha ses of the of lead lim b................................................................ 43 3-4 Electrode placem ent........................................................................................................... 44 3-5 Placement of the retro reflective markers and the wireless EMG transmitter................... 45 5-1 Comparison of propulsive force betw een limbs and across groups................................... 64 5-2 Comparison of braking force between limbs and across groups....................................... 66 5-3 Average amplitude of TA in limb x phase interaction....................................................... 70 5-4 Average amplitude of Soleus in condition x phase interaction.......................................... 72

PAGE 10

10 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science CONTROL STRATEGIES DURING GAI T TERMINTAION: ELUCIDATING THE MECHANISMS OF ANKLE INSTABILITY. By Amruta D Inamdar August 2008 Chair: Mark Tillman Major: Applied Physiology and Kinesiology Chronic ankle instability (AI) is a functionally debilitating condition th at occurs after the first episode of injury to the ligaments of the ankle joint. Both feedback and feedforward mechanisms of neuromuscular cont rol have been reported to be a ffected in people suffering from AI. However no previous researchers have examin ed these deficits in the same population. In this study we used the gait termination model to challenge both the f eedback and feedforward mechanisms of neuromuscular c ontrol in the same population of subjects suffering from AI and compared the results to controls. Planned gait termination was used to reveal deficits in feedforward neuromuscular contro l whereas unplanned gait termina tion was used to reveal the deficits in the feedback neuromuscular control. Th us the purpose of this study was to reveal the type of alterations in neuromuscu lar control that exist in AI. Force, stability and EMG were used as outcome measured to reveal these defic its. The study was a single session, single subject mixed model design. Twenty participants were recruited in the AI group (age 20.2.2years, height 169.8.7cm and mass 74.2 20.2kg) and 20 part icipants were recruited in the control group (age 20.4 1.6years, height164.3 .9cm and mass 64.2 .6kg). The Ankle Injury Questionnaire was used to screen the participan ts. Each eligible par ticipant performed both unplanned and planned gait termination trials. Normal walking trials were interspersed to avoid

PAGE 11

11 anticipation of stop. Ten trials of both planne d and unplanned gait termination were captured with each limb serving as a lead limb in 5 trials. Ground reaction fo rces were collected using two force plates (0.4m x 0.6m) (Type 4060-10 Bertec Corporation, Co lumbus, Ohio). The GRF data along X, Y and Z axes were use to calculate dynamic postural stability index (DPSI) and its subcomponents (APSI, MLSI and VSI). A KONIG SBERG T-42AL-8T (Konigsberg Instruments Inc, California) telemetric elec tromyography unit was used to record muscle activity of Tibialis Anterior (TA), Soleus (Sol) and Gluteus Medius (GM) usi ng bipolar 1-mm x 10-mm Ag/AgCl surface electrodes. A 3-way MANOVA (2:group x 2 :limb x 3:condition) with repeated measures on the last factor revealed signi ficance differences in propulsive force and braking force. Both propulsive force and braking force in the AI gr oup were higher than the control group during unplanned gait termination and the AI group reli ed more on lead limb strategy during gait termination. A 4-way MANOVA (2:group x 2:limb x 6:phases x 2:condition) with repeated measures on the last factor revealed significance differences in average amplitude of TA, Sol and GM. The average amplitude of TA for the involved limb was less than the uninvolved limb during all four subphases of st ance. The average amplitude of TA, Sol and GM was higher during unplanned gait termination than during pl anned gait termination. Average amplitude of Sol in the AI group was less than that in the control group. This finding failed to explain the result that the AI group generates a higher brak ing force and suggested that another muscle might be responsible for producing the braking fo rce in the AI group during gait termination. A 3-way MANOVA (2:group x 2:limb x 2:condition) with repeated measures on the last factor revealed higher DPSI and APSI scores in th e AI group and involved limb during unplanned gait termination than during planned gait termination. All these fi nding suggest that feedback and feedforward deficits of neuromuscular control coexist in AI.

PAGE 12

12 CHAPTER 1 INTRODUCTION Chronic Ankle Instability A lateral ankle sprain is one of the most comm only occurring musculoskeletal injuries which affects both athletes and non athletes (Osb orne & Rizzo, Jr., 2003). The incidence of ankle sprains is extremely high, with approximately 23,000 ankle sprains occurring each day in the United States (Hertel, 2002). In addition, this incidence could be an unde restimate as 56.8% of individuals do not seek conventional medical tr eatment (Gerber, Williams, Scoville, Arciero, & Taylor, 1998). The process of healing from this injury may be complete in three to six weeks, however, the functional recovery of the injured ankle may be difficult and sometimes impossible to achieve (McKay, Goldie, Payne, & Oakes, 2001). Injury of the ankle ligament commonly leads to a condition called Chronic A nkle Instability (CA I) in which the person complains of weakness and giving way or rolling over. Fift y to 75% of the people who sprain the lateral ligaments of their ankle suffer chronically from pain and instability (Yeung, Chan, So, & Yuan, 1994). While the exact cause of incomplete f unctional recovery of the ankle is unknown, preliminary evidence suggests that both feedback and feed forward neuromuscular control may play a role (Delahunt, Monaghan, & Caulfi eld, 2006; Freeman, Dean, & Hanham, 1965; Monaghan, Delahunt, & Caulfield, 2006). Stimulation of a corrective response within the corresponding system after sensory detection is often considered as a feedback control mechanism. In contrast, feedforward control mech anisms have been described as anticipatory actions occurring before the sensory detection of a homeostatic disruption (Johansson & Magnusson, 1991; Riemann & Lephart, 2002). The inju ry to the ligament complex that occurs

PAGE 13

13 during trauma to the ankle may affect the mecha noreceptors and the proprioceptors in the ankle joint (Hertel, 2002). This may result in decreas ed afferent input through these receptors. Thus, during gait the proprioceptive input is decreased causing an alteration in the feedback neuromuscular control of the ankle. In addition, researchers have examined different movement and muscle recruitment patterns in patients with CAI. Changes in the movement patterns and muscle recruitment after the first episode of ankle injury appear to exist (Delahunt et al., 2006). As the pre programmed movement patterns demons trate alterations, the feed forward mechanism of neuromuscular control may also be a contribu ting factor towards Ankle Instability (AI). Significance of gait terminatio n as a model for studying AI: To date, researchers have tried to uncover the mechanism of AI by usi ng a jump landing protocol (Ross & Guskiewicz, 2004b; Wikstrom, Tillman, Chmielewski, Cauraugh, & Borsa, 2007). However, the activity of jump landing has multiple degrees of freedom (due to multiple lower extremity joints) and a person may be successful in maintaining balan ce in spite of alterations in normal ankle strategies. Alterations in the ne uromuscular control of the ankle if any, could be revealed by an experimental model involving gait termination. Ga it termination involves a rapid deceleration of the forward momentum of the body during steady gait. A safe termination of gait requires a complex interaction of the neuromuscular system (Hase & Stein, 1998). Furthermore, the gait termination model replicates a functional activ ity and does not allow for multiple degrees of freedom. It also possesses a known and repeat able set of neuromuscular responses (Bishop, Brunt, & Marjama-Lyons, 2006; Bishop, Brunt, K ukulka, Tillman, & Pathare, 2003; O'Kane, McGibbon, & Krebs, 2003) and gait termination e xperiments can be constructed to challenge both feedforward and feedback neuromuscular control (Bishop et al ., 2006). Changes in neuromuscular controls in an unstable ankle may be studied by designing experiments to

PAGE 14

14 modulate the style of gait termin ation. Moreover a better underst anding of these changes could be possible by specifically studying the force and muscle activity of the lead limb (the first limb to land on the second force plate) and the trail limb (the first limb to land on the force plate) during gait termination. Figure 1-1 Nomenclature of th e limbs during gait termination, FP1 = Force Plate 1 and FP2 = Force Plate 2 Planned stopping or termination of gait challenges the feedforward mechanisms of neuromuscular control. Similarly, sudden or Unplanned gait termination will challenge the feedback mechanism of neuromuscular control to maintain stability and balance. Analyses of force, postural stability and musc le activation ankle joint during planned and unplanned gait termination may reveal the alterations in feedforward and feedback mechanisms of neuromuscular control respectively.

PAGE 15

15 Hypotheses The central hypothesis of this st udy is that differences betwee n feedback and feed forward m echanisms of neuromuscular cont rol related to ankle mechanics w ould be revealed by using the gait termination model. The following are the specific hypotheses for the proposed study. Forces Propulsive force The propulsive force during unplanned gait term in ation would be greater than the propulsive force during planned gait termination. Unplanned gait termination implies an unanticipated and sudden stop which does not provide time to reduce the propulsive force as during planned gait termination. The propulsive force for the AI group would be lower than that of the control group. Ligament injuries that accompany repeated an kle sprains in CAI contribute to feedback deficits in neuromuscular control (Freeman et al., 1965). Furthermore, the average amplitude of muscles responsible for generating the propu lsive force at the ankle (TA and Sol) has been documented to be significantly lower in the AI group as compared to the controls during normal gait (Delahunt et al., 2006) The propulsive force of the involved limb would be lower than the propulsive force of the uninvolved limb during gait termination. The evid ence supporting hypothesis b) explains this hypothesis as well. Propulsive time Propulsive tim e is defined as the time from the instant of heel strike to point of maximum propulsive force generation in stance limb. The propulsive time for unplanned gait terminat ion would be shorter than that observed during planned gait termination. During gait or ga it termination an impulse (I = F x t) must be created within each limb. Time and force ar e inversely related hence the propulsive time was shorter when the propul sive force is higher. The propulsive time for the AI group would be longer than that of the control group. In addition to the inverse relationship between fo rce and time, this hypothesis is based on the research findings of Nyska (2003) who dem onstrated slowing down of weight transfer from heel strike to toe o ff during normal gait in AI. The propulsive time of the involved limb would be longer than that of the uninvolved limb. Similar to the hypotheses above, slowing down of weight transfer during stance phase could result in a longer propulsive time in the injured limb.

PAGE 16

16 Braking force The braking force during unplanned gait term in ation would be greater than that seen during planned gait termination. Unplanned gait termination implies an abrupt termination of gait. Hence a greater braking force wa s generated to reduce the bodys forward momentum. The braking force for the AI group would be higher than that of the control group. Decreased conduction velocity in the tibial and peroneal ne rves has been reported in subjects having chronic ankle instability (J azayeri Shooshtari, Didehdar, & Moghtaderi Esfahani, 2007). This finding may indicate a decr ease in the braking force generated by Sol when the injured limb is the lead limb. Likewise, it may also indicate a decrease in the activity of the TA which is re sponsible for producing a braking force when the AI limb is a trail limb. Hence a much higher braking force of the uninjured limb will be necessary to arrest the forward momentum of the body during gait termination. The braking force of the involved limb would be lower than the uninvolved limb. As stated above, a decrease in conduction ve locities of tibial a nd peroneal nerves of the injured limb may restrict force generation. Braking time Braking time for the unplanned gait term inati on would be shorter th an that observed in planned gait termination. Assuming that brak ing force is increased as hypothesized above and understanding that a similar impulse must be generated to term inate gait during both planned and unplanned trials, unp lanned gait termination should require a shorter braking time. Braking time for AI group would be shorter than that seen in the control group. An Assumption based on the above hypothesis that the braking force in AI group would be higher during gait termination and understand ing the impulse relationship (I= F x t), braking time in AI group would be shorter. The braking time of the involved limb would be significantly longer than that seen for the uninvolved limb. The hypothesis th at braking force generated by the injured limb would be lesser than the uninjured limb along with th e understanding of the impulse relationship could mean a longer braking time in th e involved limb during gait termination. Postural stability The AI group would be less stable than the control group. W ikstrom and colleagues (2007) demonstrated deficits in the dynamic postural st ability in the subject with history of AI using a single leg hop-s tabilization model. Unplanned gait termination would be less stab le than planned gait termination. Unplanned gait termination implies an abrupt stop in ga it. An associated higher braking force as hypothesized above could cause unplanned ga it termination to be less stable.

PAGE 17

17 The involved limb would be less stable than the uninvolved limb. Jump landing models have demonstrated a decreased dynamic postu ral stability in the involved limb subjects with AI. (Ross & Guskiewicz, 2004a; Wikstrom et al., 2007; Docherty, Valovich McLeod, & Shultz, 2006; Ergen & Ulkar, 2008; Evans, Hertel, & Sebastianell i, 2004; Ross et al., 2004a; Wikstrom et al., 2007). However this is the first study using a gait termination model to investigate the deficits in dynami c postural stability in subjects with AI. Muscle Activation Distal muscles (TA and Soleus) The average am plitude of muscle recruitment fo r the distal muscles (TA, Sol) for the AI group would be decreased as compared to the controls. A decrease in average amplitude of TA and Sol is reported in subjects with AI during normal gait (Monaghan et al., 2006) Hence a decrease in average amplitude of dist al muscles is expect ed even during gait termination. The average amplitude of the distal muscle s (TA and Sol) for unplanned gait termination would be increased relative to planned ga it termination. Unplanned gait termination involves an unanticipated stop with presumab ly increased force generation. Hence it is likely that the distal muscles show higher av erage amplitude than that expected in a planned stop. The average amplitude of the distal muscle s (TA and Sol) for involved limb would be lesser than that seen in the uninvolved limb. The average amplitude of th e distal muscles of the injured limb has been reported to be lower than that of the uninjured limb (Delahunt et al., 2006) and conduction velocity of the tibial and peroneal nerves is reduced. Proximal muscle (Gluteus Medius) The average am plitude of muscle recruitment for the proximal muscle (GM ) for the AI group would be higher as compared to the c ontrols. A decreased reliance on the distal musculature for gait termination could im ply an increased reliance on the proximal muscles for arresting the forward mo mentum during gait termination. The average amplitude of GM during unplanne d gait termination would be greater than that seen in the planned gait termination. This hypothesis is based on a similar reasoning as that of the above hypothesis The average amplitude of GM in the involved lim b would be greater than that seen in the uninvolved limb. Significance of the Study Many authors have investigated the alterations in either the f eedback or the feed forward mechanisms in individuals with AI. However the potential difference in the feedback and feed

PAGE 18

18 forward mechanisms in the same set of subj ects with ankle instability remains unexamined. Thus, this study is designed to examine the und erlying alterations in the feed forward and feedback mechanisms in the same population of AI subjects. Specifically, I compared the ground reaction forces and dynamic postural stability dur ing gait termination in AI subjects and controls. Potential differences in muscle activa tion strategies will also be evaluated in both groups during planned and unplanned gait term inations. This was useful in understanding the alteration in the neurophysiology in AI. Until the mechanism of AI is elucidated, prophylactic and rehabilitative interventions will remain ineffective for this widely prevalent and functionally debilitating musculoskeletal injury.

PAGE 19

19 CHAPTER 2 LITERATURE REVIEW Introduction The ankle joint bears m ore weight per unit ar ea than any other join t in the body (Fallat, Grimm, & Saracco, 1998) and is one of the mo st commonly injured joints of the body. The extreme forces that are placed on the ankle join t make it more susceptible to injury (Fallat et al. 1998). Seventy-five percent of the injuries caused to the ankle are ligamen t injuries and 85 % of those are of the lateral ligament complex cau sed by inversion sprain s (Baumhauer, Alosa, Renstrom, Trevino, & Beynnon, 1995). The rate of r ecurrence of these sprains is very high and often leads to long term symptoms like pain a nd instability (Baumhauer et al., 1995; Verhagen, de Keizer, & van Dijk, 1995; Yeung et al., 1994). Chronic symptoms related to ankle instability cause individuals to be less activ e over their lifespan (Verhagen et al. 1995). Additionally the recurrent episodes of ankle injury may predispos e the joint to ankle osteoarthritis (Gross & Marti, 1999; Harrington, 1979; Hertel, 2002; Ve rhagen et al., 1995). The following review examines the various theories present in the literature regarding instability of the ankle and the underlying neurophysiological reasons behind it. Further it evalua tes gait termination as a model to study these changes in ankle instability. More specifically, in sight into the mechanisms of ankle injury may be gained by examining the anatom y of the talocrural joint, differing degrees of ankle instability, neuromechanical control mechanisms, and human gait. The Talocrural Joint. The ankle, o r talocrural joint, is a synovial hi nge joint that connects the distal ends of the tibia and fibula in the lower limb with the proximal end of the talus in the foot. The distal ends of the tibia and fibula (medial and lateral malleolus) articulate with each other to form a concave

PAGE 20

20 surface called the Ankle Mortise. The talus articulates with this ankle mortise to form the talocrural joint (Loudon, 1998; Lynch, 2002; Nyska, 2002; Loudon, 1998). The lateral malleolus of the fibula and the me dial malleolus of the tibia along with the inferior surface of the distal tib ia articulate with three facet s of the talus to form the bony articulation of the joint. The mediolateral stabilit y of the ankle joint is maintained by the medial and the lateral malleolus of the tibia and the fibul a. The medial malleolus extends to the body of the talus, however the lateral malleolus fails to extend as low as the medial malleolus Thus the anatomical structure of the ankle joint is eff ective in preventing the ev ersion or medial joint sprains making the body more susceptible to lateral ankle sprains or inversion sprains (Baumhauer et al., 1995; Hertel, 2002; Meir Nyska, 2002; Smith & Reischl, 1986). Due to this unique anatomical alignment the ankle joint is heavily dependent on its lateral ligament complex and musculature to stabilize the joint and to prevent lateral ankle sprains (Baumhauer et al., 1995; Hertel, 2002; Meir Ny ska, 2002). The three major structures that stabilize the ankle are the tibiofibular ligament, the medial ligament complex and lateral ligament complex. Tibiofibular Ligament The tibiof ibular ligament is formed by the distal and proximal extremes of the interosseous membrane that traverses the entire lower le g, connecting the tibia and fibula. The oblique arrangement of the tibiofibular ligaments aid in the distributi on of force placed upon the lower leg and stabilizes the ankle from rotationa l forces during activity (Meir Nyska, 2002). Medial and Lateral Ligament Complexes The ankle joint is bound m edially by the strong deltoid ligament and laterally by three ligaments, namely, the anterior talofibular ligamen t, the posterior talofibular ligament, and the calcaneofibular ligament (See Figure2-1,Figure2-2). The deltoid ligament supports the medial

PAGE 21

21 side of the joint, and is attached at the medial malleolus of the tib ia and connects in four places to the sustentaculum tali of the calcaneus, calcaneonavicular ligament, the navicular tuberosity, and to the medial surface of the talus. The Deltoid ligament is stronger than any of the ligaments considered individually (Hoppenfel d S., 1976) The lateral ligament complex is made up of the anterior and posterior talo fibular ligaments and the cal caneofibular ligaments. The anterior and posterior talofibular ligaments support the lateral side of the joint from the lateral malleolus of the fibula to the dorsal an d ventral ends of the talus. The calcaneofibular ligament is attached at th e lateral malleolus and to the lateral surface of the calcaneus (Delahunt et al., 2006; Kapandji, 1987; Nyska, 2002; Kapandji, 1987). During an inversion trauma the anterior talofibular ligament is the first to undergo stress and hence is most susceptible to injury. The orientation of the fibers of the calcanofibular ligament is more vertical and hence it is injured only in grade 2 sprains (partial rupture of the stabilizing ligamnets) of the ankle joint. The poste rior talofibular ligament is the strongest of the lateral ligaments and is least susceptible to injury (Hoppenf eld, 1976). Each of these ligamentous structures provides passive support for the ankl e joint although they do not act alone. Active restraints (muscles) also add stability. Figure 2-1 Lateral ligaments of the ankle joint

PAGE 22

22 Figure 2-2 Medial ligaments of the ankle joint Muscular Reinforcement of the Ankle Joint. The ligam entous framework of the ankle jo int is further reinforced by the surrounding muscles. These muscles are divided into four main compartments of the lower leg with each compartment serving a different function (H oppenfeld S., 1976). The anterior compartment consists of the TA and toe extensors. Thes e muscles serve as the primary and secondary dorsiflexors of the foot respect ively. The lateral compartment hol ds the peroneus longus and peroneus brevis muscles, which are the main ever tors of the foot and limit excessive inversion during activity. The tibialis posteri or and toe flexors make up the deep posterior compartment.

PAGE 23

23 These muscles are secondary invertors and pl antar flexors of the foot respectively. The superficial posterior compartment is made up of the gastrocnemius and Sol muscles, the main plantar flexors of the foot and main stabilizers of ankle motion (Kapandji.I.A, 1987; Meir Nyska, 2002). While all the muscles mentioned are important, weakness of the peroneals and gastrocnemius/Sol complex would more significantly put a person at risk to injury, specifically to inversion ankle sprains (Kannus & Rens trom, 1991; Monaghan et al., 2006). An initial inversion sprain can lead to long term disa bility known as chroni c ankle instability. Chronic Ankle Instability As indicated, the ankle joint m ay not fully r ecover after the first ep isode of ankle injury. The recurrence rate of ankle sprain s is as high as 70% after the in itial trauma to the ankle joint (McKay et al., 2001). The symptoms that are comm only seen in the recurr ent episodes are pain, swelling, tenderness and the feeling of giving way or rolling over of the ankle. The residual symptoms of the first episode of ankle sprain affect 55% to 72% of patients at 6 weeks to 18 months post injury (Hertel, 2002). The frequency of complications and br eadth of longstanding symptoms after ankle sprain has led to the s uggestion of a diagnosis of the sprained ankle syndrome(Fallat et al. 1998; Hertel, 2002) and to the conclusi on that there is no such thing as a simple ankle sprain.(Fallat et al. 1998; Verhagen et al. 1995). These recurrent episodes of ankle sprains after the initial trauma are term ed as Chronic Ankle Instability (CAI). The literature on ankle injury delineates two main theori es that may lead to chronic ankle instability namely mechanical instability and functional instability. Mechanical Instability Mechanical instability can be caused due to th e incongruence of anatom ic al structures at the ankle joint after the first ep isode of injury that may lead to abnormal joint mechanics. The

PAGE 24

24 contributing factors to mechanical instability are pathologic laxity, arthro kinematic restrictions and degenerative changes in the ankle joint and synovium (Hertel, 2002). Pathologic laxity Ras mussen and Tovborq-Jensen (1982) first de scribed the method for graphic recording of rotatory movements in osteoligamentous ankle preparations. In this study they recorded the mobility patterns of the ankle with intact ligam ents and after successively cutting the lateral collateral ligaments of the ankle in the anteroposterior direction. They concluded that talar tilt increases in the frontal plane and internal rotatio n of the talus increases in the horizontal plane beyond physiological limits as the de gree of injury is gradually in creased. Thus they suggested that there is an altered joint m echanics with injury to the anterior talofibular ligament (AFTL) and the calcanofibular ligamnet (CFL). Hollis and colleagues (1995) studied the effect of simulated ankle ligamentous injury on ankle-subt alar joint complex laxity by testing 36 intact ankles after gradually loading th em with stress in the anteroposte rior and mediolateral direction. The AFTL and the CFL were then gradually severe d and the joints were reassessed. They found that there was a dramatic increase in talar tilt and rotation along with greater movement in the subtalar joint. Other research ers have also shown that the CFL is the second most common ligament injured along with the ATFL (Hollis, Bl asier, & Flahiff, 1995). Recent studies have investigated the effect of serial sectioning of the ankl e joint ligaments in cadavers. The authors have reported that the sectioning of the ATF ligam ent resulted in external rotation of the fibula. Furthermore injury to ATF and Deltoid ligaments resu lted in greater laxity than the injury to the PTF(Beumer et al., 2006; Teramoto, Kura, Uchiyama, Suzuki, & Yamashita, 2008). Thus pathologic laxity in the talocrural and the distal tibiofibular joint is caused by the injury to the ligamentous complex. This can cause excessive acce ssory motions in the ankle joint leading to CAI.

PAGE 25

25 Athrokinematic insufficiencies Another factor contributing to the m echanical instability is deviation in the normal joint arthrokinematics at the talocrural and the subtal ar joints(Hertel, Denegar, Monroe, & Stokes, 1999). One of the reasons for recurrent ankle spra ins could be a deviance from the physiological position of the involved joint surface s at the talocrural joint. It is suggested that there is a decrease in the posterior fibular glide observed after the initial ep isode of ankle injury (Hertel, 2002; Mulligan BR., 1995). Also, the lateral malleol us of the fibula may have a more posterior dislocation at the ankle joint. As seen earlier, injury to the lateral ligaments may allow excessive internal rotation of the talus (Rasmussen & Tovborg-Jensen, 1982). It is postulated that the decrease in posterior fibular glide along with th e abnormal internal rotation of the talus causes the ATF to remain in a lax position predisposin g the ankle to inversi on sprains (Hertel, 2002). Thus changes in the normal joint arthrokinematics that result after the first episode of injury may cause mechanical instability and make the ankle joint more susceptible to injury Degenerative changes in the ankle joint and synovium Repetitive ankle sprains have been known to cause degenera tive changes and osteoporosis and osteophytes in the ankle jo int. G ross and Marti (1999) obse rved that volleyball players who had unstable ankles had more osteophytes and sy novial sclerosis as compared to the players without any history of ankle in jury. Sugimoto et al. (1997) co nducted a radiographic examination of 136 subjects with acute ankle sprains and 85 subjects with chronic an kle instability. They concluded that the varus tilt of the tibial plafond is more often seen in patients with chronic ligament instability of the ankle than in patients with acute ligament sprain (Sugimoto, Samoto, Takakura, & Tamai, 1997).

PAGE 26

26 Functional Instability Freem an et al. (1965) was the pioneer in prov iding the evidence that proprioceptive input is necessary for dynamic stability of the ankle. A ligament injury often results in injury to the joint mechanoreceptors. He proposed that the inst ability at the ankle is due to a lack of proprioceptive input from the joint mechanorecepto rs. However, subsequent research revealed that insufficient input from the mechanoreceptors could not explain the instability of the ankle joint completely (Lephart, Pincivero, & Rozzi, 19 98). The concept of insufficient neuromuscular control was proposed by Lephart et al. (1998) wh o proposed that neuromuscular control and joint stabilization are primarily medi ated by the central nervous syst em and the spinal cord. These researchers also suggested that proprioception and the accompanying neuromuscular feedback mechanisms provide an important component fo r the establishing and maintaining functional joint stability. Therefore, functi onal stability of the ankle join t is dependent on neuromuscular mechanisms as well. Further, recent research also suggests that mechanical and functional instabilities do not totally correla te and that functional instabi lity may exist with or without mechanical instability(Tropp, 2002). Functional ankl e instability is reported to impair the performance and quality of life in sporting and non sporti ng population (Fallat et al., 1998; Gerber et al., 1998; Gross et al., 1999; Lync h, 2002; Osborne et al., 2003; Tropp, 2002). Hence further research on functional ankle instabilit y is justified. The importance of sensory, proprioceptive and postural control systems as cont ributors towards the functional instability is described in the subsequent section. Impairment of sensation and prop rioception and postural control: Ankle injuries are known to affect afferent sensation in the lower limb (Lephart et al., 1998). Bullock and Saxton (1995) also found significant sensory deficits in the affected lower limbs of subjects who suffered ankle injuries and subsequent instability. These decreases in sensation are often

PAGE 27

27 associated with direct injury to the peron eal nerve which may occur during ankle sprains. Postural stability and proprioceptio n are also known to be affected by ankle instability (Docherty et al., 2006; Ergen et al., 2008; Ev ans et al., 2004; Ross et al., 2004a ; Wikstrom et al., 2007), but are normally evaluated in static situations. Wikstrom et al. (200 6) suggested that dynamic ankle instability can be a significant factor hindering an athletes pe rformance and that aggressive rehabilitation or surgery may not help the ankl e recover completely. These authors also suggest that the dynamic postural stability index (DPSI) can be used as an effective tool to detect the difference between the stable and functionally unstable ankle. The measurement of DPSI involves the use of a Jump landi ng model (the participant ju mps to a height 50% of his maximum vertical leap and then lands on one limb on the force pl ate) and Single leg stance model has been used until now as a primary model to detect deficits in dynamic postural stability (Ross et al., 2004a; Wikstrom et al., 2007; Docherty et al., 2006; Ergen et al., 2008; Evans et al., 2004; Ross et al., 2004a; Wikstrom et al., 2007). The dynamic postural control can be effectively maintained by the body with the help of different feedback and feedforward mechanisms inherent to it. Hence the study of possible alterations in these mechanisms could help us understand the instability of th e ankle with a better perspective. Feedback and Feedforward Mechanisms for Postural Control Johansson and Magnusson (1991) suggested that the stim ulation of a corrective response within the corresponding system af ter sensory detection is often considered as feedback control. In contrast, feedforward control mechanisms have been described as anticipatory actions occurring before the sensory detection of a homeostatic disr uption (Johansson et al., 1991; Riemann et al., 2002). The ankle joint is prone to sprains and injuries when there is a sudden shift or deceleration of the bodys COM (Johansson et al., 1991; Wikstrom, Tillman, Chmielewski, & Borsa, 2006).

PAGE 28

28 Injury to the anatomical structures around the ankle joint (muscles, ligaments and the mechanoreceptors) is a logical cons equence of an injury. Freeman et al. (1965) first put forth the pathoetiologic model for ankle instability based on the impair ment of the feedback control mechanism. In this model they proposed that instab ility that resulted after the initial ankle injury was a result of loss of proprioceptive input fr om the injured ligaments. Recent studies have demonstrated that there is a decrease in proprio ception and joint positi on sense after the first episode of injury to the ankle joint (Docherty et al., 2006; Freem an et al., 1965; Monaghan et al., 2006; Ross et al., 2004a; Wikstrom et al., 2007). Th ese deficits can demonstrate impairments in the feedback neuromuscular contro ls. Empirical data also indicate that there is alteration in the neuromuscular firing patterns in unstable ankles during normal gait (Delahunt et al., 2006; Jazayeri Shooshtari, Didehdar, & Moghtaderi Es fahani, 2007; Monaghan et al., 2006; Ross et al., 2004a; Tropp, 1986). This could help explain the theory that was put forth by Freeman and support the importance of feedback neuromuscular control mechanisms and faults in feedback neuromuscular control mechanism. However, lo ss of proprioception can only partially explain instability at the ankle joint. Recent research suggests that the feedforward neuromuscular control is also responsible for maintaining the stability at the ankle (Delahunt et al., 2006). Subsequent to the initial trau ma, the body may develop compensatory movement strategies to avoid short term effects like pain. It has been suggested that the inappropriate positioning of the ankle joint before ground contact may have import ant implications on the stability of the ankle joint (Monaghan et al., 2006). These compensatory movement patterns are thought to be the causal factors for the development of ankle inst ability. Nyska (2003) demo nstrated changes in the pattern of force transfer between the foot a nd the floor associated w ith chronically sprained ankles by measuring the peak forces and their timing under several region s of the feet during

PAGE 29

29 level walking. The authors found that patients with ankle instability showed slowing down of weight transfer from heel strike to toe off. A reduction in foot impact and lateral shift of body weight was also noted. This was in accordance with the hypothesis that there is central reorganization of movement pa tterns after ankle injury causi ng a change in feed forward mechanisms. Change in kinetics and kinematics of the ankle joint in patients with ankle instability during overground walkin g at self selected velocities was demonstrated by Monaghan et al. (2006). In this study, subj ects with chronic AI were found to have a more inverted position both before and immediately after heel strike compared with the control group. The subjects with AI showed a concentric evertor moment whereas th e controls demonstrated an eccentric invertor moment at heel strike. The limitation of this study was that no elec tromyography(EMG) data were collected. Deficits in dynamic stability of th e ankle joint can also be predictors of changes in the feedforward neuromuscular control. Differe nt models are used in literature to measure dynamic stability at the ankle joint. Ross and Guskiewicz (2004a) used single leg stance and jump landing models to examine the dynamic stability of the ankle joint. Th ey demonstrated that dynamic postural stability is affect ed in subjects having a chroni cally unstable ankle. Wikstrom and colleagues (2007) used the dynamic postural st ability index (DPSI) to measure the various components of dynamic stability in a single leg h op-stabilization model. They found that there was a significant difference in the vertical as we ll as the anteroposteri or component of dynamic stability between the control group and the subjects having a history of ankle instability. Thus, feed forward mechanisms of postural control may also contribute to AI. Dynamic Postural Stability Index W ikstrom et al. (2005) proposed a new force plat e technology measure for the measurement of dynamic postural stability. Dynamic postural stability can be define d as the ability of the person to control balance while transitioning from a dynami c to a static state (Goldie, Bach, & Evans,

PAGE 30

30 1989). Time to Stabilization (TTS) has been used as an objective measure for dynamic postural control in patients with functiona l instability of the ankle (McK inley & Pedotti, 1992; Ross et al., 2004b; Ross, Guskiewicz, & Yu, 2005). These aut hors described TTS as the time required to minimize the ground reaction force (GRF) after landi ng on the force plate to within a range of baseline (static) GRF. When a subject lande d on a force platform after a jump, TTS was measured from the forces created in 3 directions (vertical, medial-lateral and anterior-posterior). This gave researchers 3 separate measures for dynamic postural stability. The necessity of this new measure of dynamic postural control was jus tified, as the authors argued that TTS has several inherent flaws. One of the major flaws as described by Wi kstrom and colleagues was that this measure did not provide a common thread among the forces in 3 directions. Thus the global picture of dynamic postural stability remained unclear. Dynamic postural stability index (DPSI) was proposed by the authors as a measure that accounts for this shortcoming. DPSI was calculated by measuring stability indices in three principal directions: anterior-posterior (APSI), medial-lateral (MLSI) and verti cal (VSI). Stability indices were calculated by measuring the mean square deviations assessing fluctu ations around a zero point. Thus the MLSI and APSI assess the fluctuations from 0 along the frontal and sagittal axes of the force plate, respectively. The VSI assesses the fluctuation from the subjec ts body weight to standa rdize the vertical GRF along the vertical axis of the fo rce plate. The authors proposed DPSI as a composite score of APSI, MLSI and VSI and this measure was shown to be sensitive for chan ges in all directions. The authors proposed the following formula for calculation of DPSI. Formula for DPSI MLSI = [ (0x)2/number of data points] APSI = [ (0y)2 /number of data points]

PAGE 31

31 VSI = [ (body weight z)2 /number of data points] DPSI = [ (0x)2 + (0y)2 + (body weight z)2/number of data points] Wikstrom et al. (2006) used DPSI to estimate th e deficits in postural stability in subjects with self reported ankle instability. A jump protocol was used in which the participants were instructed to stand 70 cm away from the center of force plate. The pa rticipants were then instructed to jump to a height 50% of their maximum vertical leap before landing on the force plate. The authors found that there was a defici t of dynamic postural control in participants having functional instability of the ankle joint. This was demonstrated by significantly higher APSI, VSI and DPSI scores. Thus, DPSI has been used successfully as a measure to detect deficits in dynamic postural stab ility in subject with ankle instability. The application of this index to human gait could help unveil the deficits in the dynamic postural stability in patients with AI. Human Gait Introduction Nor mal human locomotion is known to be a complex interaction of various mechanical and neural strategies used by the body to move fr om one point to a desired second point. It is a rhythmic activity in which a human body is propelle d with the use of two legs alternately while maintaining the upright stability of the trunk. A typical gait cycle can be divided into three stages: gait initiation, rhythmic walking, and gait termination. Gait Initiation The biom echanics and the muscle activity during upright standing are well studied (Breniere & Do, 1986; Breniere & Do, 1991; Bren iere & Dietrich, 1992; Crenna & Frigo, 1991). During quiet stance there is a slow constant tonic cont raction of the Sol muscle. This helps to maintain the fine balance between the locations of the COM and the COP which remain within

PAGE 32

32 the base of support in quiet stance. When observed in the sagittal plan e, the COM is located anterior to the axis of the ankle joint. Gravity acts on the COM pushing it further away from the ankle joint. This creates a dorsiflexion torque at the ankle joint. The tonic muscle contractions of the Sol muscles help to counteract this torque by producing a plantarflexion torque of equal magnitude. The COP is moved forward beyond the location of the COM by the plantar flexion torque which pushes the COM back thus maintain ing equilibrium. During gait initiation there is a slight shift of the body weight on the side of th e stance foot. Pressure is created in backward and lateral direction by the foot that is about to get into the swing phase by a short burst of TA(Crenna et al., 1991). In accordance to the Newtons second law, the COM moves forward and to the side of the stance leg. This allows unloading of the swing leg making the initiation of the gait possible. Thus gait init iation involves the inhibition of th e tonic postural activity of the Sol muscle along with a burst of activity in the TA which in turn causes the shift in the weight on the side of the stance limb. The abductor muscle s on the side of the stance limb help in this weight shift (Brunt et al. 1991) Rhythmic Walking Prior to taking the first step the CO M is always kept within the base of support. Interestingly movement of COM anterior to the base of support marks the beginning of walking. It takes 2-3 steps to initiate r hythmic pattern of gait. The body uses energy economically by using the six determinants of gait that help in preventing excessive rise and fall of the center of mass. Once steady velocity of gait is achieved the body there is interaction of potential energy and kinetic ener gy based on the inverted pendulum model. The trunk is at the lowest point and has the highest kinetic ener gy during the phase of double limb support .During swing phase the trunk is lifted up by the suppor ting leg converting the kinetic energy into potential energy. During the late r half of the swing phase th e trunk is lowered and the COM

PAGE 33

33 moves in front of the supporting leg thus conve rting the kinetic energy to potential energy and this helps the body to slow down. This sinusoidal path of the center of mass continues till the point at which gait is terminated. Gait Termination The biom echanics of gait termination is less researched as compared to gait initiation and rhythmic walking. To achieve a safe and stable gait termination the forw ard acceleration of the body needs to be controlled to achieve a stable upright posture (Hase et al., 1998). A reduction in the forward push off force during th e last step and an increase in the posterior braking force are suggested to be the possible tw o reasons for gait termination (Jaeger & Vanitchatchavan, 1992). During termination, there is a large burst of So l muscle activity and a sharp reduction in TA recruitment just prior to heel strike in the lead ing limb. There is also a recruitment of Quadriceps and GM in the leading limb which helps in knee extension (Bishop, Brunt Pathare, & Patel, 2002). The trail limb shows recruitment of TA, Hamstrings (mainly Biceps Femoris ) and GM muscles which causes the body to move more poster ior (Bishop et al., 2002; Crenna et al., 1991). This helps the COM move in a posterior direction as compared to the COP which aids the body to decelerate (Stein & Hase, 1999). The study of quantifica tion of gait termination and the factors affecting it may help us understand the us efulness of this model for studying the deficits of postural control in AI. Quantifying Gait Termination Sparrow et al. (2005) d iscussed the need for an operational definition of gait termination The biomechanical definition of stopping is based on the displacement characteristics of the feet. Stopping in a gait cycle could be desc ribed as a point in time where the anterior displacement of the feet is absent and the fo rward progression has ceased (Sparrow & Tirosh, 2005). However a small difference exists between planned and unplanne d patterns of gait

PAGE 34

34 termination. Planned gait termination is a conditio n in which a person has prior knowledge of the time and location where he or she needs to stop. However in unplanned gait termination the person is totally unaware of the time and location where he or she needs to stop until receiving the cue. One important difference observed be tween the planned and unplanned stopping is the speed of feet placement. Jaeger and Vanitchatchavan (1992) observe d that the foot placement in planned gait termination takes 0.5 sec longer th an that in unplanned gait termination. This additional time is required for pl acing both feet together in pla nned gait termination. Thus while quantifying gait termination by using temporal measures, the time difference between the presentation of the stopping stimulus and the fina l heel contact is important. Bishop et al. (2003) coincided the delivery of the st opping stimulus with significan t events in gait cycle and found that there was a significant difference in the magnitude of breaking impulse which directly depended on the time point in the gait cycle at which the stopping signal was delivered. The net braking impulse that resulted when a stopping signal was given at heel strike was greater than the braking impulse when the signal was given at peak loading which was greater than that when the signal was given at midstance. They also defined a short delay at heel strike, a further delay at peak loading (braking) and the latest signal at mid-stance (the ve rtical force minimum). Stimulus delay and breaking force impulses can, therefore, be manipulated relative to key events of the gait cycle. Factors Affecting Gait Termination The term ination of gait and the characterist ic patterns of ground reaction forces and moments are governed by the following factors: stim ulus delay, stimulus pr obability and velocity of gait.

PAGE 35

35 Stimulus Delay Hase and Stein (1998) described the relationship between the tim e at which the stopping stimulus was delivered during the gait cycle and the resulting foot placemen t pattern. They stated there are two main pattern s of gait termination. The first pattern is when the subject stops with the right leg forward (lead limb forward) and the second with th e left (trail) limb forward. The pattern of stopping is governed by the pha se of gait cycle duri ng which the stopping stimulus was delivered. When the stimulus was app lied between 35 to 70% of the gait cycle (late stance phase to mid swing phase of the right limb) the rapid stopping ended with right limb forward. Whereas, when the stimulus was delivered within the first 20% of the gait cycle, the stopping occurred with the left li mb forward. When the stimulus was delivered between 20 to 35% of the gait cycle or between 70 to 85% of the gait cycle, the subject tended to stop with an extra transitional step. Very rare ly did the subjects stand with f eet together, except after having a particularly short step in the tran sition periods (Hase et al., 1998). Stimulus Probability Sparrow et al. (2005) proposed that the proba bility of stim ulus delivery might have a significant effect on different variable like the stopping time, gr ound reaction force of the stance and the swing limb etc during gait termination. In their review of the literature, these authors stated that a majority of the studies examining different aspects of gait termination used a high probability of stimulus delivery. The authors debated that higher probability of stimulus delivery could involve a pre-planned response. Thus the responses could be faster as a result of motor planning. However Tirosh et al. (2004) examined th e effect of probability of stimulus delivery on gait termination. In the experiment that was conducted they compar ed the effect of high stimulus probability (80%) and low stimulus probability (10%) on variables such as stopping time and stopping distance. They concluded that stimul us probability had only a weak effect on the

PAGE 36

36 stopping variables and stated that Stimulus proba bility was found only to affect gait termination weakly and it appeared that within the laborator y, even in the low probability condition, stopping remained relatively well anticipated. Velocity of Gait The m aintenance of the COM within the base of support is a pre requ isite for maintaining balance during gait termination. Tirosh and Sparro w (2005) developed a model to predict a safe velocity for gait. The key parameters of this model were the horizontal velocity of the center of mass normalized to body height and its position in foot lengths relative to the base of support. The fundamentals of the model are that either a forward fall or an additional step was initiated if states exceeded the regions upper boundary and a backward fall or step was required if the lower boundary were exc eeded. See Figure 2-3 Figure 2-3 Left hip anterior velocityposition plot for young (dark symbols)and older (light symbols) adults during two-step stopping. The stability region is drawn with the upper boundary for normal conditions (upper solidline) and when recalculated to accommodate a 59% reduction in strength (m iddle dashed line) to simulate the capabilities of older adults as discussedin the final section. A fall forward was anticipated if the left hiphorizontal velocityposition exceeds the upper boundary of the stabilityregion. (Reproduced from Tirosh and Sparrow [19]

PAGE 37

37 Thus, if an individual has a gait velocity above 1.1m/s at 2.4 f oot lengths anterior to foot contact, he or she would terminate gait with an additional step to maintain balance. Bishop et al. (2002) examined the relationship between the cade nce rate and the patter n of foot placement in gait termination. These authors measured the lo wer limb muscle recruitment at different cadences (50%, 100% and 150% of normal cadence ra te) and found that as the cadence increased the braking forces that were produced by the lead limb during ga it termination also increased. Also, there was an increased reliance on the l ead limb for gait termination (Bishop et al., 2002; Sparrow et al., 2005). Gait Termination as a Model for Elucid ation of the Feedforward and Feedback Mechanisms in Ankle Injury. The current literature on ankle instability sugg ests that either feedforward, or feedback control or both may be altere d in individuals who have inst ability at the ankle joint. Unfortunately, there are no studies to date that have examined the differences in feedforward and feedback control in the same se t of subjects with AI. Researcher s have studied mechanisms of AI utilizing jump landing models (Ross et al., 2004b; Wikstrom, Tillman, & Borsa, 2005a). However, this experimental model may have a limited use in revealing the mechanisms behind AI. The mechanics of landing posses great variability due to multiple degrees of freedom, i.e. control of the hip, knee, ankle, and foot, which can be used to perform the task successfully. Based on these observations, a need of an experim ental model that would reveal the underlying mechanisms that lead to AI is justified. Understanding these mechanisms will be useful to device different rehabilitative interventions to prevent AI. The present study is designed to determine if neuromuscular control alterations exist in AI su bjects. To achieve this, I propose to use a gait termination model. Gait termination involves a ra pid deceleration of the forward momentum of the body during steady gait. A safe termination of gait requires a comple x interaction of the

PAGE 38

38 neuromuscular system.(Hase et al., 1998) Furthermore, it possesse s a known and re peatable set of neuromuscular responses (Bis hop et al., 2006; Bishop et al., 2003; O'Kane et al., 2003) and gait termination experiments can be constructe d to challenge both feedforward and feedback neuromuscular control (Bishop et al., 2006). Elucidating the ne uromuscular control mechanism is of fundamental interest, because it will directly impact the direction of future investigations, and lead to effective interventions for AI.

PAGE 39

39 CHAPTER 3 METHODS This experim ent was a single session, single subject mixed model design. We analyzed the activation patterns of the muscles of lower extremity along with the propulsive force, propulsive time (time after heel strike during which ma ximum propulsion occurs), braking time and the braking force (ground reaction force) that was generated during gait termination. These data were compared between limb (involved, uninvolved) between the groups (healthy, AI), within task (planned and unplanned gait te rmination) and across different sub-phases of stance using EMG. Twenty young adults in the age group of 18 to 30 yrs and with a history of recurrent ankle sprains were recruited to part icipate in the experimental group. Twenty healthy young adults were recruited in the control group. Th is sample represented both genders. Recruitment Participants were recruited for this study from the campus of the University of Florida. The participants were informed about this study during various lectures and laboratory classes. The interested participants were further screened for the inclusion and exclusio n criteria as discussed below. The subjects satisfying the age, inclusio n and exclusion criteria were contacted and the interested participants were pr ovided with additiona l details regarding the study. Initially, the participants were screened us ing a ankle injury questionnaire and Ankle Injury Questionnaire tool (Ross & Guskiewicz, 2004c; Ross, Guskiewicz, Gross, & Yu, 2008; Rozzi, Lephart, Sterner, & Kuligowski, 1999). This questionnaire contains 6 items. The responses to each of the items were helpful in gaining insight into the injury status. They indicated which ankle was injured, the nature of treatment or immobilization, the dura tion of immobilization, the rate of recurrence of ankle sprains, whether they have returned to their initial level of activity and if they attribute the

PAGE 40

40 symptoms of instability to the first ankle inju ry. The eligible participants also signed the informed consent forms. Each testing session took about two hours includi ng the initial set up. Inclusion Criteria At leas t one grade 2 injury (partial loss of integrity of the lateral ligament complex of the ankle) which required immobilization in the past. History of recurrent ankle sp rains (at least 1 in the past 6 months) which the patient attributes to the previous ankle injury. The responses of the particip ant to specific parts of Ankl e Injury Questionnaire.(See appendix A) Exclusion Criteria Fracture or any other acute or chronic orthopedic com plication for which weight bearing is contraindicated. Systemic illnesses and conditions which could interfere with balan ce and gait (cataracts, glaucoma etc causing problems with clarity of vision, head injury, ve stibular involvement, hypertension, diabetes). Subjects with a history of bilateral ankle instability were excluded from the study. Instrumentation Force Platform Gait term ination trials were performed along an 8-m walkway contai ning force plates. Ground reaction forces were coll ected using two force plates (Type 4060-10 Bertec Corporation, Columbus, Ohio). The force plat es (0.4m x 0.6m) were mounted flush with the surface of the walkway in the laboratory so that full foot c ontact occurred on both force plates during normal walking. The force plates were oriented so that the laboratory coordinate system coincides with the right posterior corner of the force plates, with the X-axis ali gned in the direction of forward progression, Y-axis aligned in the mediolateral direction and Z-axis in the vertical direction (Figure 3-1). Forces and moment s along the 3 principle axes were sampled at 1200Hz. The force plate amplifiers were turned on at least one hour before the actual data collection.

PAGE 41

41 The Motion Capture System The m otion analysis for this study was perf ormed with a VICON Nexus 1.0 motion capture system (VICON motion systems, Calif ornia). Gait trials were digi tally recorded with 8 MX 20+ cameras mounted around the force platforms along th e perimeter of the lab (see Figure 3-1). The orientation of the cameras in the room was such that at any time instance within the calibrated volume each reflective marker was in the field of view of at least 2 cameras. The data were captured at 120Hz. Prior to the data collection the capture vo lume was calibrated using a Tshaped calibration wand (with 5 retroreflective markers). The same wand was used to set the volume origin of the capture volume which orient ed the cameras with the three axes. Within the calibrated volume, the motion capture system is accurate to < 2mm. The video cameras and force platform recordings were synchronized us ing the VICON Nexus motion capture system. Figure 3-1. Placement of the cameras and the force plates Camera 1 Camera2 Camera3 Camera4 Camera 6 Camera 7 Camera8 x y z Camera 5

PAGE 42

42 Electromyography (EMG) A KONIGS BERG T-42AL-8T (Konigsberg Instruments Inc, California) telemetric electromyography unit was used to record muscle activity for all trials. Bipolar 1-mm x 10-mm Ag/AgCl surface electrodes with an inter detection surface distance of 1.5-cm was used to detect the muscle activity. The KONIGSBERG T-42AL-8T telemetric EMG system uses a transmitter which is attached to the subject during the walking trials. This transmitter broadcasts the muscle activity of the subject to the VICON Nexus 1.0 motion capture system. The signals were band pass filtered (20-4Khz) and full wave rectified. The processed EMG and amplified force plate signals was sampled at a rate of 1200Hz. A sliding average with a window frame of 25 milliseconds was calculated to smooth the data. Stance phase of gait fo r the trail limb was divided into four sub-phases [1-h eel strike (HS) to peak loadin g (P1), 2-peak loading (P1) to mid-stance (MS), 3-mid-stance (MS) to second peak load (P2), and 4second peak load (P2) to toe-off (TO)]. See Figure 3-2. The stance phase of gait for the lead limb was divided into two sub phases (heel strike (HS) to peak loading (P1), peak loading of the lead limb (P1) to mid stance of the stance limb (MS)) based on the vertical grou nd reaction force (Figure 3-3) (Bishop et al., 2006; Bishop et al., 2003). Figure 3-2. Force curve with s ub phases of the trail limb.

PAGE 43

43 Figure 3-3. Force curve with sub phases of the of lead limb Further, average amplitude of EMG for each of the six sub phases of gait was calculated. This amplitude was normalized with the averag e amplitude of the corresponding phase for the normal walking trials Testing and Subject Preparation Data were c ollected in a single testing se ssion (an approximate dur ation of 2hrs) during which participants wore dark-col ored tight-fitting shirts shorts and walked bare foot along the 8m long walk way. Thrity-six passive retroref lective markers were placed over anatomical landmarks according to the Plug in Gait marker system (Figure 3-5). The 36 markers were used to construct a simple 15-segment model using the VICON motion captur e system. The spatial orientation of these markers was used by the VICON motion capture system to estimate the walking velocity during each trial. EMG data was collected (Konigsburg instruments T-42ALST) from the TA, Sol, and GM as representative muscle groups in both lower extremities. Prior to the electrode application th e skin was shaved and cleaned with alcohol to reduce skin impedance which allowed for a clearer signal (B rask, Lueke, & Soderberg, 1984; Hung & Gross, 1999; Isear, Jr., Erickson, & Worrell, 1997; Ninos, Irrgang, Burdett, & Weiss, 1997). The muscles were palpated for their exact anatom ical location and two bipolar surface electrodes

PAGE 44

44 with an inter detection surface distance of 1.5c m placed on the muscle bellies bilaterally. The exact anatomical location of the electrodes is described in Table 3-1 and shown in Figure 3-4 (Brask et al. 1984; Perotto A, 1994). A) B) C) Figure 3-4 Electrode placement a) Soleus b) Tibialis Anetrior c) Gluteus Medius Table-3-1. Electrode placement: Muscle Electrode location Soleus Placed one inch lateral and 3 inch inferior to the lower edge of the gastrocnemius belly on either side Gluteus Medius Spaced approximately at the midpoint of the perpendicular line joining the iliac crest and the greater trochanter. Tibialis Anterior Placed 1.5 inches lateral a nd 1 inch inferior to the tibial tuberosity Static Trial Once the participant was fitted with reflectiv e markers and EMG electrodes, one 5 second static trial was collected. In th is trial the participant stood as still as possible in the capture volume of the cameras in order to reconstruct the live model of the participant in the VICON software. All 36 markers had to be visible to the cameras and labeled in the system. Procedure for Testing Unpl anned Gait Termination Each participant was instructed to w alk along the 8 m long walk way at normal gait velocity and was instructed to stop only when cued by a buzzer. The buzzer was triggered on random trials just before the heel strike of the lead limb so that the participant was able to stop within one foot strike after the trigger. Each limb was the lead limb in 5 trials. Unplanned gait

PAGE 45

45 termination velocity was also compared to self-s elected gait velocity to ensure that gait speed was kept constant. Trials were rep eated until we recorded 10 valid trials The participant also performed normal walking trials in which no signal was given to prevent counting or anticipation of gait termination. These trials were interspersed within the sequence of unplanned stopping trials and constituted at least 60% of total number of trials recorded. During data collection the unplanned and normal walking trials were recorded before the pl anned gait termination trials. This eliminated the learning effect. Procedure for Testing Planned Gait Termination In this task the participan t was instructed to perform a planned gait termination. The participant performed 10 trials. Each limb served as the lead limb in 5 trials of planned gait termination. In each trial, the participant walk ed down the 8 meter long walkway and stopped on the second force plate. The participants were in structed to walk with his or her usual gait velocity. Procedure for calculation of the DPSI The ground reaction force data were used to ca lculate the Dynam ic Post ural Stabiltiy Index (DPSI) for a duration of 1 sec post heel strike in the lead limb. DPSI was calculated for both involved and uninvolved limbs, AI and control groups and during unplanned and planned gait termination trials. Figure 3-5. Placement of the retro reflective markers and the wireless EMG transmitter

PAGE 46

46 Statistics Statis tics for timing of stimul us delivery and gait velocity: Comparison of the kinetic and kinematics and muscle recruitment between the control and the AI groups during gait termination was reasonable only if the gait velo city remains same for both the groups and the relative timing of stimulus delivery is same for both groups (AI and control), both limbs ( involved/dominant and uninvolved/non dominant) and for each condition (planned, unplanned and catch) (Sparrow et al., 2005). A three way factorial ANOVA (2:group x 2:limb x 3:condition) was computed for velocity. A two way factorial ANOVA (2:group x 2:limb) was performed for the timing of stimulus delivery. B onferonni post hoc analysis was performed when necessary. A traditional level of significance ( =0.05) was used.. Statistics for Kinetic and Kinematics Force: A 3-way MANOVA (2:group x 2:lim b x 3:condition) with repeated measures on the last factor was computed for propulsive fo rce, propulsive time, braking force and braking time. Bonferonni post hoc analysis was perfor med when necessary. A traditional level of significance ( =0.05) was used. See Appendix B for a detailed list of dependent and independent variables. Statistics for Stability DPSI was calculated for each of the independent variab les. The effect of the independent variables on the subcomponents of DPSI i.e. A PSI (Antero posterior stability index), VSI (Vertical stability index) and MLSI (Mediola teral stability index) was also studied. The involved or uninvolved status of the limb is important in the AI group. Hence it was included as one of the independent variables. The empirical results from the prior studies have shown that for the control group the DPSI is independent of the dominance of the limb (Wikstrom et al., 2005a; Wikstrom, Tillman, Smith, & Borsa, 2005b). However for the control group the dominant

PAGE 47

47 and non dominant status of the limb were included as an independent variable to maintain statistical balance. A 3-way MANOVA (2:gr oup x 2:limb x 2:condition) was computed for APSI,MLSI, VSI and DPSI Bonferonni post hoc analyses were performed when necessary. A traditional level of significance ( =0.05) was used. See Appendix 2 for a detailed list of dependent and independent variables. Statistics for EMG MANOVA ( 2:group x 2:limb x 6:phase 2:condition) with repeated measures on the last factor were computed for 3 muscles on each limb across 6 phases. Bonferonni post hoc analysis was performed when necessary. A traditional level of significance ( =0.05) was used. See Appendix B for a detailed list of de pendent and independent variables.

PAGE 48

48 CHAPTER4 RESULTS Demographics The sam ple size consisted of 40 participants. Tw enty participants were recruited in the AI group (age 20.2.2years, height 169.8.7cm a nd mass 74.2 .2kg) and 20 participants were recruited in the control gr oup (age 20.4 1.6years, height 164.3 .9cm and mass 64.2 .6kg). Amongst the participants recruited in the AI gro up, six participants had their left limb injured whereas 14 had their right limb injured. All the pa rticipants in the control group identified the right limb as their dominant limb. The data collected from all 20 participants in each group were used to compute the force and stability statisti cs. However, the EMG statistics were computed using the data from 20 participants form the control group and 18 participants from the experimental group. These data were considered outliers created by technical problems. The probable reason for this was that the data were collected when the battery of the EMG telemetric unit was almost exhausted. Independent samp le t-tests were performed to compare anthropometric data (age, height, weight) be tween both groups. No significant statistical differences were observe d between groups (p>0.05). Velocity A 3-way ANOVA (2:group x 3:condition x 2:limb) was computed to co mpare the velocity across each of the independent variables. No significant group [F (1,199) = 1.12, p = 0.29] or limb [F (1,199) = 0.03, p = 0.84] main effects were noted. However, a significant condition [F (2,199) = 4.22, p = 0.01] main effect was observed. Bonforroni post hocs revealed that the gait velocity during unplanned gait termination was significantly higher than the gait velocity during planned gait termination (p < 0.01). No significa nt two-way interactions were revealed for condition x group [F (1,199) = 0.13, p = 0.84], cond ition x limb [F (1,19 9) = 0.08, p = 0.77] or

PAGE 49

49 group x limb [F (1,199) = 0.48, p = 0.46]. Similarly, no significant threeway interaction was detected for condition x group x limb [F (1,199) = 0.48, p = 0.48]. Gait velocity data appear in Table 4-1. Table 4-1 Velocity (m/s) (Mean SD). Indicate s a statistically significant difference between unplanned and planned trial (p 0.05). Variable Level M SD 1) Condition a)catch 1.23.01 b)planned* 1.19.11 c)unplanned* 1.24.48 2) Group a)healthy 1.21.01 b)AI 1.23.11 3) Limb a)right 1.22.11 b)left 1.22.11 Relative Time of Stimulus Delivery A two-way ANOVA (2:group x 2:limb) was comput ed to determ ine the relative difference in time between stimulus delivery and heel stri ke of the trail limb. There was no significant difference between groups [F (1,38) = 1.9, p = 0.17] or limbs [F (1,38) = 0.061, p = 0.80]. Also no group x limb [F (1,38) = 0.05, p = 0.14] interacti on was detected. Time of stimulus delivery data appear in Table 4-2. Table 4-2 Relative timing of stimulus delivery(s) (Mean SD) Variable level M SD 1)Group a)healthy 0.07.07 b)AI 0.05.08 2) Limb a)right 0.06.08 b)left 0.05.09 Force A 3-way MANOVA (2:group x 2:limb x 3:condition) with repeated m easures on the last factor was computed for propulsive force, pr opulsive time, braking force and braking time.

PAGE 50

50 Significant condition [F (8,374) = 96.28, p < 0.001] and group [F (8,187) = 6.32, p < 0.001] main effects were observed. However no significant ma in effect for limb [F (4,187) = 0.075, p = 0.99] was noted. There were no signifi cant two way interactions: c ondition x group [F (8,374) = 1.12, p = 0.34], condition x limb [F (4,187) = 0.45, p = 0.76] or group x limb [F (4,187) = 0.230, p = 0.921]. Also, no significant condition x group x limb [F (4,187) = 0.41, p = 0.79] interaction was observed. Follow-up univariate ANOVA were perfor med for propulsive force, propulsive time, braking force and braking time. The results for each of these dependent variables are provided below. Propulsive Force Subsequent ANOVA re vealed that there was a significant cond ition main effect [F (2,199) = 68.96, p < 0.001]. Bonferroni post hocs revealed th at the propulsive force during both planned and unplanned gait termination was significantly lower than that in catch trials (p < 0.001). It was also observed that the AI group had higher propulsive force than the control group [F (1,199) = 8.40, p = 0.004]. Propulsive for ce data appear in Table 4-3. Table 4-3. Propulsive Force (N) (Mean SD). *I ndicates a statistically significant difference between planned and catch trial (p 0.01). Indicates a sta tistical significant difference unplanned and catch trial (p 0.01). Indicates statistically significant difference between AI and healthy (p 0.01). Variable level M SD Propulsive Force 1) Condition a) catch* 144.61 38.16 b) planned* 83.54 20.71 c) unplanned 79.68 28.63 2) Group a) healthy 88.60.33.64 b) AI 99.82.40.79 3) Limb a) involved/dominant1 02.38 42.23 b) uninvolved/healthy 81.96 25.37

PAGE 51

51 Propulsive Time The ANOVA did not reveal significant c ondition [F (2,199) = 1.64, p = 0.20] or group [F (1,199) = 0.694, p = 0.40] m ain effects. Propulsive Ti me data are shown in the subsequent table Table 4-4. Table 4-4 Propulsive Time (s) (Mean SD) Variable level M SD Propulsive Time 1) Condition a) catch 0.62 0.14 b) planned 0.59 0.10 c) unplanned 0.57 0.14 2) Group a) healthy 0.59 0.12 b) AI 0.58 0.13 3) Limb a) involved/dominant 0.59 0.13 b) uninvolved/ healthy 0.58 0.12 Braking Force A significant condition m ain effect [F ( 2,199) = 31.40, p < 0.001] was observed. Further, Bonferroni post hocs revealed that braki ng force during unplanned gait termination was significantly higher than braking force during planned gait termina tion (p < 0.01) Braking force in the AI group was higher than in the control group [F (1,199) = 24.17, p < 0.001]. Braking force data are shown in Table 4-5. Table 4-5 Braking Force (N) (Mean SD). I ndicates a statistically significant difference between planned and unplanned (p 0.01). Indicates statistically significant difference between healthy and AI (p 0.01). Variable level M SD Braking Force 1) Condition a) planned 177.42 56.21 b) unplanned 222.89 84.39 2) Group a) healthy 161.61 62.2 b) AI 207.11 80.89 3) Limb a) involved/dominant 173.18 72.33 b) uninvolved/ healthy 201.13 77.54

PAGE 52

52 Braking Time No significant group m ain effect [F (1,199) = 0.007, p = 0.93] was noted. However, a significant condition main effect [F (2,199) = 214.34, p < 0.001] was observed. Bonferroni post hocs revealed no significant differences with the braking time during planned and unplanned gait termination. Data for Braking time are shown in Table 4-6. Table 4-6 Braking Time (s) (Mean SD) Variable level M SD Braking Time 1) Condition a) planned 0.70 0.11 b) unplanned 0.67 0.17 2) Group a) healthy 0.58 0.23 b) AI 0.58 0.26 3) Limb a) involved/dominant 0.51 0.28 b) uninvolved/ healthy 0.68 0.12 Stability A 3-way MANOVA (2:group x 2:limb x 2:conditio n) was co mputed for APSI, MLSI, VSI and DPSI. Significant main effects were not ed for condition [F (4,149) = 8.39, p < 0.001] and group [F (4,149) = 4.68, p < 0.001]. No significant limb main effect was detected [F (4,149) = 1.82, p = 0.12]. No significant condition x group [F (4,149) = 0.77, p = 0.54], condition x limb [F (4,149) = 0.27, p = 0.89], group x limb [F (4,14 9) = 0.89, p = 0.46] or condition x group x limb [F (4,149) = 0.87, p = 0.48] interactions occurre d. Subsequent ANOVA were computed for each dependent variable, the results of which are discussed below. Anteroposterior Stability Index (APSI) The APSI score for the AI group was higher than the control group [F (1,159) = 8.33, p = 0.004]. The APSI score for unplanned gait term inati on was higher than planned gait termination [F (1,159) = 26.88, p < 0.001]. APSI scor es can be found in Table 4-7.

PAGE 53

53 Table 4-7 Anteroposterior Stability Index (APSI ) (Mean SD). Indicates a statistically significant difference between unplanned and planned (p 0.01). Indicates statistically significant difference between healthy and AI (p 0.01). Variable level M SD APSI 1) Condition a) planned* 0.13 0.01 b) unplanned* 0.15 0.02 2) Group a) healthy 0.13 0.019 b) AI 0.14 0.02 3) Limb a) involved/dominant 0.14 0.02 b) uninvolved/ healthy 0.14 0.02 Mediolateral Stability Index (MLSI) The ANOVA revealed n o significant group [F (1,159) = 0.93, p = 0.33], condition [F (1,159) = 0.25, p = 0.30] or limb [F (1,159) = 1.13, p = 0.28] main effects. MLSI data appear in Table 4-8. Table 4-8. Mediolateral Stabil ity Index(MLSI) (Mean SD) Variable level M SD MLSI 1) Condition a) planned 0.11 0.69 b) unplanned 0.03 0.01 2) Group a) healthy 0.10 0.69 b) AI 0.02 0.01 3) Limb a) involved/dominant 0.11 0.69 b) uninvolved/ healthy 0.14 0.02 Vertical Stability Index(VSI) No significant condition [F (1,159) = 0.07, p = 0.79] or group [F (1,159) = 2.82, p = 0.09] m ain effects were noted for VSI. VSI data are shown in Table 4-9. Table 4-9. Vertical Stability Index (VSI) (Mean SD) Variable level M SD VSI 1) Condition a) planned 0.17 0.03 b) unplanned 0.17 0.03 2) Group a) healthy 0.17 0.03 b) AI 0.18 0.04 3) Limb a) involved/dominant 0.17 0.04 b) uninvolved/ healthy 0.17 0.03

PAGE 54

54 Dynamic Postural Stability Index (DPSI) A higher DPSI score was detected during unplanned gait term ination trials than during planned gait termination trials [F (1,159) = 7.22, p = 0.008]. A higher DPSI score was observed in AI group than in the control group [F (1,159) = 9.87, p = 0.002]. Table 4-10 shows the DPSI data. Table 4-10 Dynamic Postural Stability Index( DPSI)(Mean SD) Indicates a statistically significant difference between unplanned and planned (p 0.01) Indicates statistically significant differe nce between AI and healthy (p 0.01). Variable level M SD DPSI 1) Condition a) planned* 0.22 0.02 b) unplanned* 0.23 0.03 2) Group a) healthy 0.22 0.02 b) AI 0.24 0.04 3) Limb a) involved/dominant 0.23 0.03 b) uninvolved/ healthy 0.23 0.03 EMG Tibialis Anterior A four-way MANOVA ( 2: group x 2: limb x 6: phase x 2: condition) with repeated measures on the last factor was computed for TA. Mauchlys test indicate d that the assumptions of sphericity were violated (p < 0.001). Hence, the Greenhous e Geisser adjustment was made. Significant condition [F (1, 36) = 49.90, p < 0.001] limb [F (1, 36) = 4.47, p = 0.04] and phase [F(5,32) = 8.19, p < 0.001] main effects were observe d. No significant main effect was identified for group [F (1, 32) = 1.03, p = 0.41]. A signifi cant limb x phase [F (5,32) = 3.12, p = 0.02] interaction was noted. No signi ficant condition x group [F (1,36) = 2.19, p = 0.14], condition x limb [F (1,36) = 1.84, p = 0.18], condition x phase [F (5,32) = 1.52, p = 0.20], group x limb [F (1,36) = 0.97, p = 0.33], or phase x group [F (5,32) = 1.03, p = 0.41] interactions were detected. No significant three or four way interactions were noted: condition x group x limb [F (1,36) = 0.71 p = 0.40], condition x group x phase [F (5,32) = 0.98 p = 0.44], limb x group x phase [F

PAGE 55

55 (5,32) = 0.95 p = 0.46], condition x limb x phase [F (5,32) = 1.81 p = 0.13], and condition x group x limb x phase [F (5,32) =1.29 p = 0.29]. Subsequent ANOVA revealed that the aver age amplitude of TA during unplanned gait termination was higher than that during pla nned gait termination [F (1, 36) = 49.90, p < 0.001]. Additionally, the average amplitude of TA for the uninvolved limb was higher than the involved limb [F (1, 36) = 4.47, p = 0.04]. Also, Bonferroni post hocs revealed that the average amplitude of TA was maximal in phase 4 [F (2.11, 180) = 12.23, p < 0.001]. A significant interaction of limb x phase [F (3.84, 38.33) = 5.40, p = 0.001] was observed. However, no significant condition x group [F (1,36) = 2.19, p = 0.14], condition x limb [F (1,36) = 1.84, p = 0.18], condition x phase [F (3.29,118.74) = 2.23, p = 0.08], group x limb [F (1,36) = 0.97, p = 0.33] and phase x group [F (5,32) = 1.03, p = 0.41] interactions were observed. Also, No significant condition x group x limb [F (1,36) = 0.71, p = 0.40], conditi on x group x phase [F (3.28,32) = 0.36, p = 0.095], limb x group x phase [F (3.84,32) = 1.79, p = 0.32], condition x limb x phase [F (2.47,89.14) = 2.63, p = 0.063] and condition x group x limb x phase [F (2.47,32) = 1.24, p = 0.29] interactions were detected. Average amplitude (TA) data are shown in Table 4-11. Table 4-11 Average Amplitude (TA) (Mean SD ) Indicates a statistically significant difference between unplanned and planned (p 0.01) Indicates statistically significant difference between the involved/dominant and uninvolved/ non-dominant limb (p 0.05). Variable level M SD Average Amplitude 1) Condition a) planned* 0.99 0.03 b) unplanned* 1.20 0.05 2) Group a) healthy 1.17 0.05 b) AI 1.11 0.05 3) Limb a) involved/dominant 1.05 0.03 b) uninvolved / non dominant 1.18 0.04 4) Phase a) one 0.99 0.04 b) two 1.03 0.05 c) three 0.97 0.05 d) four 1.58 0.12 e) five 1.14 0.07 f) six 1.13 0.05

PAGE 56

56 Soleus A MANOVA (2: group x 2: lim b x 6: phase x 2: condition) with repeated measures on the last factor was computed for average amplitude of Sol. Mauchlys te st indicated that the assumptions of sphericity were violated (p < 0.001). Hence, the Greenho use Geisser adjustment was made. Significant condition [F (1, 36) = 34.99, p < 0.001], lim b [F (1, 36) = 8.65, p = 0.006] and phase [F (5, 32) = 10.50, p < 0.001] main e ffects were revealed for average amplitude. Significant limb x phase [F (5, 32) = 2.72, p = 0.03], condition x phase [F (5, 32) = 4.82, p = 0.002] and condition x group x phase [F (5, 32) = 3.35 p = 0.06] interactions were also observed. However, no significant two way interactions we re detected for condition x limb [F(1,36) = 1.92, p = 0.17] group x limb [F (1,36) = 1.86, p = 0.18] or phase x group [F (5,32) = 0.93, p = 0.47]. No significant three way interactions were id entified for condition x group x limb [F (1, 36) = 2.18 p = 0.14], limb x group x phase [F (5,32) = 0.91 p= 0.48], condition x limb x phase [F (5,32) = 0.99, p = 0.43] or condition x group x limb x phase [F (5,32) = 0.29 p = 0.91]. Subsequently, ANOVA were computed which rev ealed that the average amplitude of Sol during unplanned gait termination was greater than planned gait termination [F (1, 36) = 34.99, p < 0.001]. Average amplitude of uninvolved limb was greater than the involved limb [F (1, 36) = 8.65, p = 0.006]. There was a significant main effect for phase [F (2.93, 105.94) = 9.2, p < 0.001]. Bonferroni post hocs revealed that average amplitude of Sol was highest in phase 5 Also the AI group demonstrated higher average amplit ude of Sol than the control group [F (1, 36) = 4.18, p = 0.048]. A significant limb x phase [F (3.57, 128.72) = 4.58, p = 0.003] interaction was also observed. A significant condition x pha se [F (3.45, 124.46) = 3.37, p = 0.023] interaction was noted. The average amplitude of Sol was higher for unplanned gait termination during all six sub phases.

PAGE 57

57 No significant two way interactions were id entified for condition x limb [F (1, 36) = 2.52, p = 0.12], group x limb [F (1, 36) = 1.06, p = 0.30] or phase x group [F (5, 32) = 0.74, p = 0.52], condition x group [F (1, 36) = 3.27, p = 0.08]. Similarly, no significant condition x group x limb [F (1,36) = 2.82 p = 0.10], condition x group x phase [F (3.45, 32) = 0.99 p = 0.42], limb x group x phase [F (3.38,32) = 0.30 p = 0.84], condition x limb x phase [F (3.43, 32) = 1.48 p = 0.19], condition x group x limb x phase [F (3.43, 32) = 0.26 p = 0.86] interactions were observed. Data for average amplitude Sol for each of the independent variable appear Table 4-12. Table 4-12 Average Amplitude (Soleus) (Mean SD). Indicates a statistically significant difference from planned (p 0.01). Indicates a statistic ally significant difference from healthy (p 0.05). Indicates statistically si gnificant difference from the uninvolved/ non-dominant (p 0.01) Variable level M SD Average Amplitude 1) Condition a) planned* 1.13 0.05 b) unplanned* 1.42 0.06 2) Group a) healthy 1.13 0.07 b) AI 1.16 0.08 3) Limb a) involved/dominant 1.19 0.05 b) uninvolved/ non dominant 1.36 0.06 4) Phase a) one 1.31 0.06 b) two 1.24 0.09 c) three 0.88 0.08 d) four 1.25 0.07 e) five 1.61 0.11 f) six 1.36 0.18 Gluteus Medius A MANOVA (2:group x 2:lim b x 6:phase x 2: condition) with repeated measures on the last factor were computed for average amplitude of GM. Mauchlys test indicated that the assumptions of sphericity were violated (p < 0.001). Hence, the Greenho use Geisser adjustment was made. Significant condition [F (1, 36) = 11.49, p = 0.0002] and phase [F(5,32) = 14.77, p < 0.001] main effects were identified. However, no significant limb [F (1, 36) = 0.03, p=0.86] or

PAGE 58

58 group [F (1, 36) = 1.75, p=0.19] main effects were re vealed. No significant condition x limb [F(1, 36) = 2.13, p = 0.15], limb x phase [F (5, 32) = 1.17 p = 0.34], condition x phase [F(5, 32) = 1.88, p = 0.12], condition x group [F (1, 36) = 1.46, p = 0.23], group x limb [F(1, 36) =1.67, p = 0.20] or phase x group [F (5, 32) = 1.56, p = 0.19] interactions were detected. Also, no significant condition x group x limb [F (1, 36) = 0.001 p = 0.97], condition x group x phase [F (5, 32) = 1.77 p = 0.14], limb x group x phase [F (5, 32) = 0.818, p = 0.54], condition x limb x phase [F (5, 32) = 2.26, p = 0.07], condition x gr oup x limb x phase [F (5, 32) = 0.96, p = 0.45] interactions were detected. Subsequent ANOVA revealed that the average amplitude of GM was higher during unplanned gait termination than during planne d gait termination [F (1, 36) = 11.49, p = 0.002]. Statistical analysis also revealed that the musc le had the highest average amplitude during phase six [F (2.52, 90.75) = 12.19, p < 0.001]. A significan t condition x group x phase [F (5, 32) = 2.4, p = 0.03] interaction was detected for average am plitude of GM. No significant condition x limb [F (2.17,36) = 2.13, p = 0.15], limb x phase [F (2.67,96.30) = 1.05 p = 0.35], condition x phase [F (5,74.6) = 1.63, p = 0.20], condition x group [F (1,36) = 1.46, p = 0.23], group x limb [F (1,36) =1.67, p = 0.20] or phase x group [F (2.5,32) = 1.68, p = 0.14] interac tions were noted. No significant three way condition x group x limb [F (1, 36) = 0.001 p = 0.97], limb x group x phase [F (2.70, 32) = 1.45, p = 0.20], condition x limb x phase [F (2.28, 82.31) = 2.66, p = 0.07], condition x group x limb x phase [F (2.28,32) = 0.78, p = 0.47] interaction were observed. Table 4-13 shows the data for the average amplitude of GM.

PAGE 59

59 Table 4-13 Average Amplitude (GM) (Mean SD ). Indicates a statistically significant difference between unplanned and planned (p 0.01). Variable level M SD Average Amplitude 1) Condition a) planned* 1.19 0.05 b) unplanned* 1.40 0.07 2) Group a) healthy 1.36 0.07 b) AI 1.22 0.07 3) Limb a) involved/dominant 1.29 0.07 b) uninvolved/ healthy 1.30 0.06 4) Phase a) one 1.05 0.06 b) two 1.07 0.05 c) three 0.12 0.06 d) four 1.54 0.13 e) five 1.35 0.09 f) six 1.62 0.07

PAGE 60

60 CHAPTER 5 DISCUSSION The purpose of this study was to determ ine if differences exist between feedback and feedforward neuromuscular control mechanisms in participants with a history of AI during gait termination. We tested this by means of three major outcome measures namely force, stability and EMG. Initially gait velocity and the relati ve timing of delivery of the auditory cue with respect to the gait cycle were measured to confirm whether comparison of groups was valid. We attempted to reveal the underlying differences in force by comparing the propulsive force, propulsive time, braking force and braking time. The force results were partially supportive of the hypotheses. Dynamic postural stability in both the groups was compared on the basis of four different stability measures: APSI, VSI, MLSI and DPSI. The data suppor ted the hypothesis that the AI group would be less stab le as compared to control group. The EMG findings supported the differences in the propulsive and braking force thus contri buting to the global picture of differences in feedback and feedforward neuromuscular controls in the AI group. The EMG results supported the centra l hypothesis however only partially supported the individual hypotheses. The results and hypothe ses for individual outcome measures are examined relative to the few previously publishe d studies and discussed below. Gait Velocity It has been well documented in the litera ture that the pattern of muscle recruitment during gait termination changes with a change in gait velocity. For example, Crenna et al. (2001) suggested that with an increase in gait velocity the pattern of muscle recruitment used for gait termination shifts from distal to proximal. Hence in this study the prerequisite for a valid comparison between groups was to ensure that ga it velocity was the same. The results indicated that both the groups had very similar gait velo cities. There was no statistically significant

PAGE 61

61 difference in the gait velocity between groups. Howe ver, results indicated that the participants in both the AI and control groups walked faster during unplanned gait termination trials. The difference in the mean velocities during pl anned (1.23.01 m/s), unplanned (1.24.48 m/s) and catch (1.19.11 m/s) was practi cally negligible (~3%). Sparro w et al. (2005) proposed that the probability of stimulus delivery and an anti cipation of a stop might have an effect on different variables like the st opping time and the ground reaction force of the stance and the swing limb during gait termination. The main goal during data coll ection was to avoid anticipation of a stop during unplan ned gait termination trials. In fact, the finding that the gait velocity during unplanned gait term ination was greater indicates a minimal anticipation of a stop. Relative Timing of Stimulus Delivery It is we ll documented in literature that the patt ern of foot placement at the instant when the person terminates gait depends on the timing of s timulus delivery relative to the phase of gait cycle. Hase and Stein (1998) doc umented that when the stimulus was applied between 35 to 70% of the gait cycle (late stance phase to mid swing phase of the right limb) the rapid stopping ended with right limb forward. Whereas, when the stimulus was delivere d within the first 20% of the gait cycle, the stopping occurred wi th the left limb forward. Wh en the stimulus was delivered between 20 to 35% of the gait cycl e or between 70 to 85% of the ga it cycle, the subject tended to stop with an extra transitional step. Very rarely did the subjects st and with feet together, except after having a particularly short step in the transition periods (H ase et al., 1998). Similar findings were reported by Tirosh and Sparrow (2004) who studied the effect of stimulus delay on gait termination. A visual stopping signa l was presented 10ms before the h eel strike of the trail limb and 450ms post heel strike (just before heel off of the trail limb). The au thors concluded that the pattern of final foot placement is directly related to the relati ve timing of stimulus delivery (Bishop et al., 2003; Delahunt et al ., 2006; Tirosh & Sparrow, 2004). Hence, we chose to keep

PAGE 62

62 the relative timing of stimulus delivery constant in both groups across al l conditions. The results indicate that there was no difference in timi ng delivery between groups or across conditions. Forces Propulsive Force Hase and Stein (1998) analyzed the changes in f orces that oc curred during gait termination. They documented that propulsive force in the trail limb is sp ecifically reduced during gait termination. Crenna and colleagues (2001) descri bed the different motor programs used during gait termination. These authors have also identif ied a reduction in the propulsive force as a strategy used in the trail limb during gait termination. Interestingl y, recent studies have reported that there are significant force sense deficits in functionally un stable ankles (Bishop et al., 2003; Docherty & Arnold, 2008; Hase et al., 1998; Jaeg er et al., 1992). Doch erty and Arnold (2008) measured the ability of subject s with history of AI to produc e 10 %, 20% and 30% of maximum voluntary isometric and isokinetic torques and compared it with c ontrols. The authors noted that the AI group lacked the ability to accurately reproduce a given force. On the other hand it has been documented that in the AI popu lation the ankle has an increased joint velocity at heel strike during normal gait (Bishop et al., 2003; Monaghan et al., 2006). This is indirect evidence of inability of the AI group to modulate propulsive forces even during normal walking. In the current study we tried to compare the propulsive forces produced by th e AI population with the propulsive forces produced by healthy adults du ring planned and unplanned gait termination and during catch trials. The major finding was that the propulsive force during both planned and unplanned gait termination was significantly less than the catch trials for both AI and the control group. The first hypothesis made was that propul sive force during unplanned gait termination would be higher than during planned gait te rmination. This hypothesis was based on the assumption that unplanned gait termination was an unanticipated stop which may not provide

PAGE 63

63 enough time to reduce the propulsive force. Inte restingly the results did not support this hypothesis: the propulsive force during unplanned gait termination was in fact less than during planned gait termination, alt hough the differences were statistic ally not significant. Unplanned gait termination involves a sudden stop and hence w ould involve a greater reliance on feedback neuromuscular control for gait termination. It co uld be argued that the feedback mechanisms were equally effective in reducing the propul sive force as the feedforward mechanisms. The second hypothesis was that the propulsive force in the AI group would be lower than in the control group. This hypothesis was based on findings of previous studies which report lower average amplitude of the muscles respons ible for producing propulsion during normal gait in AI (Delahunt et al., 2006). The results of th e current study failed to support the hypothesis. Interestingly propulsive force duri ng gait termination in AI group was greater than in controls. In our experiment the AI group was unable to reduce propulsive force both during unplanned and planned gait termination when compared to th e propulsive force during catch trials. Bishop and colleagues (2003) have demonstrated that in healthy adults during unpl anned gait termination, the trail limb strategy is used to reduce propulsive fo rce when the stimulus is given at heel strike. Empirical data suggesting sensory proprioceptive deficits and alte red neuromuscular feedback in AI could explain the finding for unplanned gait termination (Dela hunt et al., 2006; Docherty et al., 2006; Freeman et al., 1965; Monaghan et al ., 2006; Ross & Guskiewicz, 2004d; Wikstrom et al., 2007). The propulsive force during unplanned ga it termination in AI group could be larger due to decreased feedback from the ankle joint. The inability of the AI group to reduce the propulsive force during planned ga it termination when compared to the propulsive force during catch trials suggests a potential reorganiza tion of feedforward neuromuscular control.

PAGE 64

64 The third hypothesis was that propulsive for ce in the involved limb would be lower than the propulsive force in the uninvolved limb. This hypothesis was again based on the finding of decreased EMG amplitude in the muscles responsible for producing propulsion during normal gait in AI (Delahunt et al., 2006) In addition, recent EMG studies regarding H:M ratios (H reflex is the electrical equivalent of a monosynaptic stretch refl ex, M wave signifies a maximal response that can be obtained on stimulation of the muscle. M wave values are used of normalization of H reflex values and the H/M rati os are used commonly in reflex studies) report that post ankle injury there is a significant ar throgenic inhibition in muscles like the Sol which are responsible for producing propulsive force (B ishop et al., 2003; Hase et al., 1998; Jaeger et al., 1992; McVey, Palmieri, Docherty, Zinder, & I ngersoll, 2005). However the present results failed to support this hypothesi s. There was no significant sta tistical difference between the involved/dominant and the uninvolved/nondominant limb. A possible reason for this finding is the method of statistical analys is. In our analysis we grouped the dominant limb in the control group with the injured limb in the AI group and the nondominant limb in the control group with the uninjured limb of the AI gr oup. Possibly this comparison failed to reveal the potential main effect of limb on the propulsive force as the differences between the involved and the uninvolved limbs of the AI group were masked by the inclusion of the control group (Fig 5-1). Propulsive Force0 20 40 60 80 100 120 uninv/noninj inj/dom LimbForce (N) Control AI Figure 5-1. Comparison of pr opulsive force between limbs and across groups. uninv/nondominant

PAGE 65

65 Braking Force The first hypothesis stated was that the br aking f orce during unplanned gait termination would be higher than that seen during planned gait terminatio n. The results of this study supported this hypothesis. Jaeger and colleagues (1992) were th e first to document that a reduction in propulsive and an increase in brak ing force were the two main characteristic findings during gait termination. During normal gait termination there is both reduction in propulsive force in the trail limb along with an in crease in braking force in the lead limb (Bishop et al., 2003; Crenna, Cuong, & Breniere, 2001; Hase et al., 1998; Jaeger et al., 1992). Hence it could be argued that during unplanned gait termin ation an increased braking force was required to arrest the forward momentum of the body gene rated by a higher propulsive force for a safe gait termination. The second hypothesis stated that the braking force in AI gr oup would be higher than that seen in the control group. This hypothesis was su pported by the results of this study. It is well documented in the literature that both reduction in propulsive force of the trail limb and an increase in the braking force of the lead limb are required for a safe gait termination (Hase et al., 1998; Jaeger et al., 1992). The requirement of a higher braking force for a safe gait termination in the AI group could be explained as a cons equence of a higher propulsive force. A higher propulsive force in AI group would make it necessary for the muscles to produce a higher braking force for gait termination. In contrast, th e controls demonstrated less braking force. It was interesting to note that in control group there was a greater reduction in propulsive force in the trail limb. Thus the control gr oup was able to effectively modul ate the trail lim b and the lead limb strategies during gait termin ation. Alternatively, the AI group depended heavily on the lead limb strategy, possibly due to feedbackback defic its during unplanned gait termination and feed forward deficits during planned gait termination.

PAGE 66

66 The third hypothesis stated that the braking force in the invol ved limb would be lower than that in the uninvolved limb. Recent studies report that there are definite deficits in force production in the injured limb (Docherty et al., 2008; Fox, Docherty, Schrader, & Applegate, 2008). Docherty et al. (2008) have also reported force sense deficits a nd the inability of the injured limb in the AI group to consistently repro duce a specific level of force. In another study, Fox and colleagues (2008) have reported that the injured limb shows deficits in eccentric plantarflexion torque when compared to the norm al limb. However, in spite of the obvious group differences in braking force the results of this study failed to support the first hypothesis. In our analysis for comparing the differences between limbs we grouped the dominant limb in the control group with the injured limb in the AI group and th e nondominant limb in the control group with the uninjured limb of the AI group. Th is comparison may have failed to reveal the statistical main effect of limb on the braking force as the differences between the involved and the uninvolved limbs of the AI group were ma sked by the inclusion of the control group. Braking Force0 50 100 150 200 250 uninv/noninj inj/dom LimbForce (N) Control AI Figure 5-2 Comparison of braking fo rce between limbs and across groups Uninv/nondominant

PAGE 67

67 Propulsive Time and Braking Time Crenna and colleagues (2001) described the m otor programs that are used during gait termination. In their study they described that the lead limb motor prog ram is quite robust. The onset latencies of the individual muscle recruited in this strategy (Quadriceps and Sol) showed a close correlation, and the spatio -temporal parameters were always scaled in parallel However they reported that trail limb mo tor program is flexible and velocity dependent. The trail limb showed a delay of 150ms in the muscle recruitm ent after the braking s timulus was applied. In contrast, the limb showed a delay of 330 ms after the stimulus was applied. Even the lead limb strategy showed a distal to proximal muscle recruitment mainly in Sol and Quadriceps. Increase in walking speed shifted the pattern of muscle recruitment in the trail limb from distal to proximal. However, in our study there was no prac tically significant diffe rence in the velocity across groups, conditions and limb. Hence it could be argued that the same lead limb and trail limb motor programs were executed while termin ating gait in both the conditions, for both limbs and in both groups. Thus if the basic sequence of the pattern of muscle re cruitment remained the same it could be argued that th e propulsive time and the braki ng time remained the same across groups, condition and limb. Stability The first hypothesis stated that the AI group would be less stab le than the c ontrol group. In this study, dynam ic postural stability was evalua ted based on scores for APSI, MLSI, VSI and DPSI. The results supported this hypothesis. Ross et al. (2004) theorized that people with FAI take longer to decelerate their center-of-mass osci llations because they allo w their center of mass to approach the limits of stabil ity which lead to large external moments that act on the joint to destabilize the body. Previous researchers have documented differences in dynamic postural stability between AI and control groups based on the APSI, MLSI, VSI and DPSI scores

PAGE 68

68 (Wikstrom et al., 2005b; Wikstrom, Tillman, K line, & Borsa, 2006; Wikstrom et al., 2007). Specifically dynamic postural stabil ity deficits in the AI group were observed. Wiksstrom et al (2005) reported that the DPSI (0.85.17), APSI (0.36.09) and VSI (0.73.17), scores in their study were higher/worse in the AI gr oup as compared to DPSI (0.73.17), APSI (0.30.06) and VSI (0.61.13) scores in contro ls indicating deficits in dynamic postural stability in AI. In the presen t study the APSI and the DPSI scor es for the AI group were higher than the control group indicating that the AI gr oup was less stable. These findings correlate to those documented in literature and in turn re inforce the finding of feedback neuromuscular deficits that exist in the AI gr oup. However, in the present study the values of DPSI and all its subcomponents are lower than those reported by Wi kstrom and colleagues (2006) in the studies done on jump landing. The likely reason for this fi nding is that the present study used the gait termination model for measuring the DPSI and its subcomponents. The task of gait termination involves a lesser impact than the task of jump landing hence the values of all subcomponents are lower. It was also interesting to note that in this study we did not find any significant differences in the VSI sub component of DPSI. The possible explanation for this fi nding is that the study used the gait termination model. Previous studi es which have documente d significant differences in VSI between the AI and the control groups ha ve used the jump landing model. The vertical destabilizing component in this m odel is expected to be greater than the gait termination model. On similar lines the APSI and the DPSI scores in unplanned gait termination were higher than the score in planned gait termin ation. Unplanned gait terminati on involved an unanticipated and sudden stop. Such gait termina tion is known to pose a greater challenge to maintenance of balance and postural stab ility (Bishop et al., 2003). The findings of this study reinforce this finding and thus indicate that the unplanned gait termination posed a greater challenge to

PAGE 69

69 maintenance of the postural stability. Planned gait termination involved an anticipated stop. Studies have documented that during planned gait termination subjects decrease propulsive force and increase braking force at least two steps prior to stopping (Wearing, Urry, Smeathers, & Battistutta, 1999) with as much as 90% of th e deceleration occurring in the final step (Jian, 1993). Thus it could be argued that this type of gait termination involves feedforward planning. The stability indices failed to show any si gnificant differences be tween dominant/involved and non dominant/uninvolved limbs. Thus the re sults did not support the hypothesis that the involved limb would be less stable than the uninvolved limb. This fi nding seems to contradict the earlier finding of group differences in the stab ility indices. However this finding could be explained on the basis of findi ngs of other studies. Wikstrom et al. (2007) reported that for healthy individuals the DPSI scor es in the dominant limb were not significantly different than those for the non dominant limb. In this study the dominant and the involved limb were together statistically compared to the non dominant and uninvolved limb. Since we had 20 healthy subjects the actual differences in the involved an d the uninvolved limb could be masked due to the design of the statistical analysis of this study. EMG Distal Muscles (TA and Soleus) The first hypothesis stated was that the aver age amplitude of the TA and Sol in AI group would be less than the control group. This hypothe sis was supported by the results of this study. The average am plitude of Sol was significantly less in the AI group as co mpared to the control group across all six phases. This finding is however contradictory to that reported in the literature. Delahunt and colleagues (2006) compar ed the integral EMG of TA and Sol between the AI and controls. They indicated that the Sol muscle in the AI group had a higher average EMG as compared to the controls. However the av erage integral EMG of TA was reported to be

PAGE 70

70 less than the controls. The likely reasons for the results of our study to be different from those reported are two. First the Delahunt study investigated the muscle activity of TA and Sol during normal walking. Conversely, our study measured th e average amplitude of these muscles during the task of planned and unplanned gait termina tion. Second the results reported by the study are calculated for the time frame of 200ms before heel strike to 80 milliseconds after heel strike. In our study we measured the average amplitude of the muscles during the terminal swing phase and whole of the stance phase of each limb. The group data were not statistically different for the TA. However the mean of the average amplitude of TA was AI group was lower than that of the control group. This finding supports the data i ndicating that the propulsi ve force for AI group was higher than in the control group. During gait te rmination, the TA is responsible for reducing the propulsive force of the tr ail limb (Bishop et al., 2002; Cre nna et al., 1991). During gait termination, TA is known to be most efficient at reducing the anterior tib ial advancement in the trail limb between heel strike to midstan ce phases (Crenna et al., 2001). The lower average amplitude of TA in the AI group may explain the finding of increased propul sive force in the AI group. The significant limb x phase interaction supports this finding mo re specifically. The average amplitude of TA in the involved limb was less than the uninvolved limb during the first four phases of gait te rmination (Fig 5-3). Average Amplitude of Soleus (Limb x Phase)0 0.5 1 1.5 2 123456 Limb x PhaseAverage EMG Ampitude involved uninvolved Figure 5-3 Average amplitude of TA in limb x phase interaction

PAGE 71

71 Also other plantar flexors like the gastrocn emius could possibly be contributing toward propulsion. However this justification is only a su rmise as our study did not measure the muscle activity of either gastrocnenius or peroneal muscle group. A burst of Sol muscle activity in the lead limb is responsible for increasing the br aking force during gait term ination (Crenna et al., 1991). The results of this study indicate that the br aking force in the AI group was more than the control group. However the averag e amplitude of the Sol muscle in the AI group was lower than the control group. These contradict ory findings may indica te that the AI group relied on muscles other than Sol for generating the braking force. The second hypothesis stated that the average amplitude of TA and Sol would be higher for unplanned gait termination as compared to th e planned gait termination. This hypothesis was supported. The average amplitudes of TA and Sol were significantly more during unplanned gait termination as compared to planned gait terminat ion. These results support the finding that the propulsive force during unplanned gait termin ation was lower (though not statis tically significant) than during pl anned gait termination. As stated earlier, an increased TA activity in the trail limb is responsible for reduction of propulsive force of the trail lim b. The results of our study demonstrated higher average amplitude of TA during unplanned gait termination than during planned gait termination. The average am plitude of Sol was higher during unplanned gait termination than during planned ga it termination. As mentioned earlier, the Sol is the primary muscle responsible for producing braking force in the lead limb during ga it termination (Crenna et al., 1991; Delahunt et al., 2006; Hase et al., 1998). Hence the hi gher average amplitude of Sol during unplanned gait termination justified the higher braking force that was evident during unplanned gait termination. More specifically a significant condition x phase interaction

PAGE 72

72 indicated that the Sloeus musc le had higher average amplitude during all six sub phases of the limb during gait termination (Fig 5-4). The third hypothesis stated that the average amplitude of TA and Sol for the involved limb would be lower than the uninvolved limb. Th is hypothesis was supported. Lower average amplitude of TA in the uninvolved limb and co rresponds to the higher propulsive force in the uninvolved limb. Average Amplitude of Soleus (condition x phase)0 0.5 1 1.5 2 2.5 123456 condition x phaseAverage EMG Amplitude planned unplanned Figure 5-4 Average amplitude of Soleus in condition x phase interaction Likewise lower TA activity in the injured limb justifies the higher propulsive force in the AI group. The finding of lower average amplitude of the Sol in the involved limb does not support the finding that the braking force in the AI group was higher than the control group. These two contradictory findings indicate th at the involved limb may rely on a muscle other than Sol for producing the braking force fo r planned gait termination Proximal Muscle (Gluteus Medius) To date researchers have focused on the c ontribution of GM duri ng ga it termination in healthy adults (Delahunt et al ., 2006; Hase et al., 1998; Crenna et al., 2001). The current study was the first to measure the average amplitude of GM in AI group during gait termination. The

PAGE 73

73 GM along with Quadriceps and Sol is responsible for the producing the braking force in the lead limb (Crenna et al., 2001). Three hypotheses were made. The first hypothesis stated that the average amplitude of GM during unplanned gait termination would be greater than during planned gait termination. The second hypothesis st ated that the average amplitude of GM for involved limb would be higher than that in the uninvolved limb. The thir d hypothesis stated was that the average amplitude for the AI group would be higher than that of the control group. The average amplitude of GM for unplanned gait termin ation was higher than that seen in planned gait termination which supported the first hypothesis made for GM. This finding also supported the result of higher braking for ce during unplanned gait terminati on that was discussed earlier. Bullock et al (1994) investigated the activation pattern s of proximal muscles like Errector spinae, Gluteus Maximus and Hamstrings. The authors reporte d there is a definite delay in the onset time of the proximal muscles on the injured side (Bullock-Saxton, Janda, & Bullock, 1994) .The present study however failed to show any diffe rence between the involved and the uninvolved limb or between the AI and the control group thus failing to support the second and the third hypotheses. Poor signal to noise ra tio due to the deep anatomical location of the muscle could be a possible cause. Synopsis of the Results It has been well documented in th e litera ture that ankle joint is more susceptible to injury when there is a sudden shift or decelerati on of COM of the body (Johansson et al., 1991; Wikstrom et al., 2006). Gait termination involves a rapid deceleration of the forward momentum of the body during steady gait. A safe termination of gait requires a comp lex interaction of the neuromuscular system (Hase et al., 1998) a nd involves a known and repeatable set of neuromuscular responses (Bishop et al., 2006; Bishop et al., 2003; O'Kane et al., 2003). Accordingly, gait termination experiments can be constructed to challenge both feedforward and

PAGE 74

74 feedback neuromuscular control (Bishop et al., 2006). Crenna et al. (2001) investigated the strategies and patterns of muscle recruitment during gait termination in healthy individuals. Other researchers have also investigated the deficits of force and changes in the muscle recruitment patterns in AI group during normal gait (Delahunt et al., 2006; Monaghan et al., 2006). Interestingly, the current study is the first to examine the forces, stability and the patterns of muscle recruitment in AI popul ation during gait term ination. In this study we attempted to unveil the feedback and feed forward deficits by analyzing forces, stab ility and the average amplitude of the muscle recruitment during unpl anned gait termination and feedback deficits during planned gait termination. The results sugges t that both feedback and feedforward deficits coexist in AI. This finding is similar to the recent work of Hertel and colleagues (2008). In this study we found significant feedb ack deficits in the AI group when we analyzed both the propulsive force and braking force. Both propulsive force and braking force in the AI group were higher than that seen for the control group dur ing unplanned gait terminat ion. More specifically the results demonstrated that the AI group relied more on lead limb strategy than the trail limb strategy during gait termination. This statement can be supported by the higher propulsive force and the braking force that can be seen in th e AI group as compared to controls. The EMG findings for TA and Sol also revealed some in teresting differences between groups. The average amplitude of TA for the involved limb was less than the uninvolved lim b. Further a significant limb x phase interaction revealed that the TA in the involved fired less than the uninvolved limb during all four subphases of stance (Fig 5-3). These two findings suggest that the involved limb was less ef fective in executing the trail limb strategy of reducing the propulsive force and also support the obser vation that the AI group relied less on the trail limb strategy of reduci ng the propulsive force during gait termination. The

PAGE 75

75 average amplitude of TA during unplanned gait termination was larger than the planned gait termination which helps explain the decrea sed propulsive force during unplanned gait termination. This finding also suggests that the feedback mechanisms of neuromuscular control are equally effective in reducing the propulsive fo rce when the stimulus to stop is given just before heel strike and is consistent with th e findings reported by Bishop and colleagues that the trail limb strategy is effective during unplanned gait termination even when the trigger is given at heel strike (Bishop et al., 2003; Delahunt et al ., 2006). The increased average amplitude of Sol during unplanned gait termination supported the re sult that braking force during unplanned gait termination was higher than during planned gait term ination. Interestingly, it was also noted that the average amplitude of Sol in the AI group was less than that in the control group. This finding failed to explain the re sult that the AI group generates a higher braking force than the control group during gait termination. However another musc le might be responsible for producing the braking force in the AI group during gait terminat ion. Greater average amplitude of the GM was evident during unplanned gait termination as co mpared to planned gait termination which supports the higher braking force that is evident during unplanned gait termination. However, the present study failed to demonstrate any group or limb differences in the average amplitude of GM. The result of increased propulsive forces in AI group as compared to controls during planned gait termination suggests that the AI group was not able to effectively reduce the propulsive force even during planned gait terminati on. Thus feedforward deficits coexist with the feedback deficits in AI. This study was the first designe d to specifically investigat e dynamic postural stability deficits in an AI group by using DPSI and its components during gait te rmination. Compromised dynamic postural stability was not ed in the AI group as compared to control and during

PAGE 76

76 unplanned gait termination than during planned gait termination. An interes ting finding of this study was that there were significa nt deficits in the anteroposteri or stability indicated by a higher APSI score in AI population and in the involved limb. Also the DPSI score for the involved limb was higher than the uninvolved limb. All these fi nding suggest that there are significant dynamic postural stability deficits in the AI population even during a simple and functional task like unplanned gait termination. Limitations This study included an AI population across only a very sm all age range. Although the inclusion criterion allowed the age range of 18 to 30 y ears, we were only able to collect data on participants between 2022 years for both the gro ups. Presumably, AI affects individuals of all ages. Further there were no specific criteria based on the functional deficits that determined the inclusion criterion as no standa rdized screening ques tionnaire (AJFAT) was used. In addition, the basic screening of the partic ipants was done on the basis of the information provided that was subjective and retrospective. A nother significant limitation was that this study used surface EMG to measure the average amplitude of the musc les during gait termina tion. Although practically difficult, indwelling EMG would be able to give mo re accurate results especi ally for muscles like GM which have an anatomically deeper locati on. In this study we investigated only three representative muscles for studyi ng the strategies of gait termination. Furthermore only average amplitude of the muscle was measured. Further investigation of muscle recruitment based on the variables like onset time, onset duration, and peak amplitude could give us valuable information regarding the pattern of muscle recruitment during gait termin ation. Also investigation of muscles (peroneal group and the Hamstrings) could help us further understand the strategies used by the subjects with AI during gait termination.

PAGE 77

77 Conclusion From the results of this investigation it appears that the control strategi es used by the AI group differ from that of the control group. In the current study we investigated fo rce, stability and the muscle recruitment in AI group and made compar isons to the control group both during planned and unplanned gait termination. The results for the fo rce and the stability m easures revealed that there are alterations in both feedback and feed forward neuromuscular control mechanisms in persons who suffer from AI. Higher DPSI and A PSI scores in the AI group along with the finding of a definite difference in the average amplitude of TA and Sol between the AI and control group and between planned and unpla nned gait termination s upported the central hypothesis that feedback and feedforward defici ts in neuromuscular control exist for AI individuals and that the gait term ination model was effective in re vealing both the those deficits. Future research should be direct ed at understanding the underlying reasons behind feedforward deficits and coul d be beneficial for clinicia ns who evaluate and develop rehabilitation protocols for individuals with AI More specifically, the time of onset of feed forward deficits may be important. Also, a possible relationship between feedback and feedforward neuromuscular deficits and whether they are reversible is of concern. Study of variables like the onset time, onset duration and RM S of Sol and TA would be helpful to broaden our understanding of the problem of AI. Further research targ eting the peroneal and the hamstring groups in the AI populat ion could help us understand the global picture of the changes in the pattern of muscle recr uitment during gait termination in this population. Elucidation of these changes could help us devi ce specific rehabilitation strategies to treat and prevent one of the most commonly occurring and functionally debi litating musculoskeleta l injury.

PAGE 78

78 APPENDIX A ANKLE INJURY QUESTIONNAIRE Ankle Injury Question naire 1. Which ankle have you sprained in the past? If neither, please turn in this questionnaire R L Neither 2. Did the initial injury to your ankle requir e crutches, immobilization, or both, of any form (cast, braces, etc.)? Y N How long were you on crutches or immobilized? ______ days 3. How many times have you sprained your ankle since the initial injury? ______ 4. How many times have you had episodes of your ankle giving way or rolling over during daily activity (athletic or otherwise)? ______ 5. Do you have pain, instability, or weakness in your involved ankle? Y N 6. Do you attribute these signs and symptoms to the previous injuries associated youre your involved ankle? Y N To qualify as an AI individual their responses will have to match a specific format : Question 1R or L but no hi story of bilateral trauma Question 2Yes but no requirements were made on the number of days Question 3must be >1 Question 4must be >1 Question 5must answer Yes Question 6Must answer Yes

PAGE 79

79 APPENDIX B STATISTICAL TABLE Table B-1. S tatistics for force Type of statistic Independent va riable( IV) Dependent variable(DV) MANOVA 1)Condition(planned,unplanned, catch) 2) Group (AI, control) 3) Limb (involved,uninvolved) 1) Propulsive Force 2) Propulsive time 3) Braking force 4) Braking Time 3 way factorial ANOVA (2x2x3) 1)Condition(planned,unplanned, catch) 2) Group (AI, control) 3) Limb (involved, uninvolved) 1) Propulsive Force 3 way factorial ANOVA (2x2x3) 1)Condition(planned,unplanned, catch) 2) Group (AI, control) 3) Limb (involved, uninvolved) 1) Propulsive time 3 way factorial ANOVA (2x2x3) 1)Condition(planned,unplanned, catch) 2) Group (AI, control) 3) Limb (involved,uninvolved) 1) Braking force 3 way factorial ANOVA (2x2x2) 1)Condition(planned,unplanned, catch) 2) Group (AI, control) 3) Limb (involved,uninvolved) 1) Breaking time

PAGE 80

80 Table B-2. Statistics for stability Type of statistic Independent Variable( IV) Dependent Variable(DV) MANOVA 1)Condition-(planned,unplanned) 2) Group (AI, control) 3) Limb (involved,uninvolved) 1) DPSI 2) APSI 3) MLSI 4) VSI 3 way factorial ANOVA (2x2x2) 1)Condition-(planned ,unplanned) 2) Group (AI, control) 3) Limb (involved,uninvolved) 1) DPSI 3 way factorial ANOVA (2x2x2) 1)Condition-(planned,unplanned) 2) Group (AI, control) 3) Limb (involved,uninvolved) 1) APSI 3 way factorial ANOVA (2x2x2) 1)Condition-(planned,unplanned) 2) Group (AI, control) 3) Limb (involved,uninvolved) 1) VSI 3 way factorial ANOVA (2x2x2) 1)Condition-(planned,unplanned) 2) Group (AI, control) 3) Limb (involved,uninvolved) 2) MLSI Table B-3. Statistics for EMG(Soleus) Type of statistic Independent Variable( IV) Dependent Variable(DV) 3 way factorial ANOVA (2x2x2) 1)Condition(planned,unplanned) 2) Group (AI, control) 3)Limb(involved, uninvolved) 1) Average amplitude (Soleus) Table B-4. Statistics for EMG (TA) Type of statistic Independent Variable( IV) Dependent Variable(DV) 3 way factorial ANOVA (2x2x2) 1)Condition-(planned ,unplanned) 2) Group (AI, control) 3) Limb (involved, uninvolved) 1) Average amplitude (TA)

PAGE 81

81 Table B-5. Statistics for EMG(GM) Type of statistic Independent Variable( IV) Dependent Variable(DV) 3 way factorial ANOVA (2x2x2) 1)Condition-(planned ,unplanned) 2) Group (AI, control) 3) Limb (involved,uninvolved) 1) GM Average Amplitude

PAGE 82

82 LIST OF REFERENCES Baum hauer, J. F., Alosa, D. M., Renstr om, A. F., Trevino, S., & Beynnon, B. (1995). A prospective study of ankle injury risk factors. Am.J.Sports Med., 23, 564-570. Beumer, A., Valstar, E. R., Garling, E. H., Niesi ng, R., Ginai, A. Z., Ranstam, J. et al. (2006). Effects of ligament sectioning on the kinematic s of the distal tibiofibular syndesmosis: a radiostereometric study of 10 cadaveric specimens based on presumed trauma mechanisms with suggestions for treatment. Acta Orthop., 77, 531-540. Bishop, M., Brunt, D., & Marjama-Lyons, J. (2006) Do people with Parkinson's disease change strategy during unplanned gait termination? Neurosci.Lett., 397, 240-244. Bishop, M. D., Brunt, D., Kukulka, C., Tillman, M. D., & Pathare, N. (2003). Braking impulse and muscle activation during unplanned gait termination in human subjects with parkinsonism. Neurosci.Lett., 348, 89-92. Bishop, M. D., Brunt, D., Pathare, N., & Patel, B. (2002). The interaction between leading and trailing limbs during stopping in humans. Neurosci.Lett., 323, 1-4. Brask, B., Lueke, R. H., & Soderberg, G. L. (1984). Electromyographic an alysis of selected muscles during the lateral step-up exercise. Phys.Ther., 64, 324-329. Breniere, Y. & Dietrich, G. (1992). Heel-off pe rturbation during gait init iation: biomechanical analysis using triaxial accelerometry and a force plate. J.Biomech., 25, 121-127. Breniere, Y. & Do, M. C. (1986). When and how does steady state gait movement induced from upright posture begin? J.Biomech., 19, 1035-1040. Breniere, Y. & Do, M. C. (1991). Control of gait initiation. J.Mot.Behav., 23, 235-240. Brunt, D., Lafferty, M. J., Mckeon, A., Goode, B ., Mulhausen, C., & Polk, P. (1991). Invariant characteristics of gait initiation. Am.J.Phys.Med.Rehabil., 70, 206-212. Bullock-Saxton, J. E., Janda, V., & Bullock, M. I. (1994). The influence of ankle sprain injury on muscle activation du ring hip extension. Int.J.Sports Med., 15, 330-334. Crenna, P., Cuong, D. M., & Breniere, Y. (2001). Motor programmes for the termination of gait in humans: organisation and velocity-dependent adaptation. J.Physiol, 537, 1059-1072. Crenna, P. & Frigo, C. (1991). A motor progra mme for the initiation of forward-oriented movements in humans. J.Physiol, 437, 635-653. Delahunt, E., Monaghan, K., & Caulfield, B. (200 6). Altered neuromuscular control and ankle joint kinematics during walking in subjects with functional instability of the ankle joint. Am.J.Sports Med., 34, 1970-1976.

PAGE 83

83 Docherty, C. L. & Arnold, B. L. (2008). Force se nse deficits in functionally unstable ankles. J.Orthop.Res. Docherty, C. L., Valovich McLeod, T. C., & Shultz S. J. (2006). Postural control deficits in participants with functional ankle instability as measured by the balance error scoring system. Clin.J.Sport Med., 16, 203-208. Ergen, E. & Ulkar, B. (2008). Propriocep tion and ankle inju ries in soccer. Clin.Sports Med., 27, 195-217, x. Evans, T., Hertel, J., & Sebastianelli, W. (2004) Bilateral deficits in postural control following lateral ankle sprain. Foot Ankle Int., 25, 833-839. Fallat, L., Grimm, D. J., & Saracco, J. A. ( 1998). Sprained ankle syndrome: prevalence and analysis of 639 acute injuries. J.Foot Ankle Surg., 37, 280-285. Fox, J., Docherty, C. L., Schrader, J., & Applegat e, T. (2008). Eccentric plantar-flexor torque deficits in participants with functional ankle instability. J.Athl.Train., 43, 51-54. Freeman, M. A., Dean, M. R., & Hanham, I. W. (1965). The etiol ogy and prevention of functional instability of the foot. J.Bone Joint Surg.Br., 47, 678-685. Gerber, J. P., Williams, G. N., Scoville, C. R., Ar ciero, R. A., & Taylor, D. C. (1998). Persistent disability associated with ankle sprains: a prospective examination of an athletic population. Foot Ankle Int., 19, 653-660. Goldie, P. A., Bach, T. M., & Evans, O. M. (1989). Force platform measures for evaluating postural control: reliability and validity. Arch.Phys.Med.Rehabil., 70, 510-517. Gross, P. & Marti, B. (1999). Risk of degenerative ankle joint disease in volleyball players: study of former elite athletes. Int.J.Sports Med., 20, 58-63. Harrington, K. D. (1979). Degenerativ e arthritis of the ankle seconda ry to long-standing lateral ligament instability. J.Bone Joint Surg.Am., 61, 354-361. Hase, K. & Stein, R. B. (1998). Analysis of rapid stopping during human walking. J.Neurophysiol., 80, 255-261. Hertel, J. (2002). Functional An atomy, Pathomechanics, and Pathophysiology of Lateral Ankle Instability. J.Athl.Train., 37, 364-375. Hertel, J. (2008). Sensorimotor deficits with ankle sprains and chr onic ankle instability. Clin.Sports Med., 27, 353-70, vii. Hertel, J., Denegar, C. R., Monroe, M. M., & Stoke s, W. L. (1999). Talocrur al and subtalar joint instability after lateral ankle sprain. Med.Sci.Sports Exerc., 31, 1501-1508.

PAGE 84

84 Hollis, J. M., Blasier, R. D., & Flahiff, C. M. (1995). Simulated lateral ankle ligamentous injury. Change in ankle stability. Am.J.Sports Med., 23, 672-677. Hoppenfeld S. (1976). Physical Examination of the Spine and Extremities Appleton & Lange, Norwalk. Hoppenfeld, S. (1976). Physical Examination of th e Spine and Extremities. Appleton & Lange, Norwalk. Hung, Y. J. & Gross, M. T. (1999) Effect of foot pos ition on electromyogr aphic activity of the vastus medialis oblique and vastus late ralis during lower-extremity weight-bearing activities. J.Orthop.Sports Phys.Ther., 29, 93-102. Isear, J. A., Jr., Erickson, J. C., & Worrell, T. W. (1997). EMG analysis of lower extremity muscle recruitment patterns during an unloaded squat. Med.Sci.Sports Exerc., 29, 532539. Jaeger, R. J. & Vanitchatchavan, P. (1992). Gr ound reaction forces during termination of human gait. J.Biomech., 25, 1233-1236. Jazayeri Shooshtari, S. M., Didehdar, D., & Moghtaderi Esfahani, A. R. (2007). Tibial and peroneal nerve conduction st udies in ankle sprain. Electromyogr.Clin.Neurophysiol., 47, 301-304. Jian, Y. W. D. A. I. M. G. a. G. L. (1993). Trajectory of the body GOG and COP during initiation and termination of gait. Gait.Posture. 1, 9-22. Johansson, R. & Magnusson, M. (1 991). Human postural dynamics. Crit Rev.Biomed.Eng, 18, 413-437. Kannus, P. & Renstrom, P. (1991). Treatment for acute tears of the lateral ligaments of the ankle. Operation, cast, or earl y controlled mobilization. J.Bone Joint Surg.Am., 73, 305-312. Kapandji, I. A. (1987). The Physiology of the Joints. In (2 ed., pp. 148-164). Kapandji.I.A. (1987). The Physiology of the Joints. 2[fifth edition], 148-164. Lephart, S. M., Pincivero, D. M., & Rozzi, S. L. (1998). Proprioception of the ankle and knee. Sports Med., 25, 149-155. Loudon, J. B. S. J. J. (1998). The Clinical Orthopedic Assessment Guide. Human Kinetics, Champaign. Lynch, S. A. (2002). Assessment of th e Injured Ankle in the Athlete. J.Athl.Train., 37, 406-412. McKay, G. D., Goldie, P. A., Payne, W. R., & Oake s, B. W. (2001). Ankle in juries in basketball: injury rate and risk factors. Br.J.Sports Med., 35, 103-108.

PAGE 85

85 McKinley, P. & Pedotti, A. (1992). Motor strategies in landing from a jump : the role of skill in task execution. Exp.Brain Res., 90, 427-440. McVey, E. D., Palmieri, R. M., Docherty, C. L ., Zinder, S. M., & Ingersoll, C. D. (2005). Arthrogenic muscle inhibition in the leg mu scles of subjects exhibiting functional ankle instability. Foot Ankle Int., 26, 1055-1061. Meir Nyska. (2002). The Unstable Ankle. Human Kinetic publisher Inc 2002. Monaghan, K., Delahunt, E., & Caulfield, B. (2006) Ankle function during gait in patients with chronic ankle instability compared to controls. Clin.Biomech.(Bristol., Avon.), 21, 168174. Mulligan BR. (1995). Manual Therapy: ''NAGS'', ''SNAGS'', ''MWMS'', Etc. 3. Wellington, New Zealand: Plane View Services LTD. Ninos, J. C., Irrgang, J. J., Burdett, R., & Weiss, J. R. (1997). Electromyographic analysis of the squat performed in self-selected lower extrem ity neutral rotation and 30 degrees of lower extremity turn-out from the self-selected neutral position. J.Orthop.Sports Phys.Ther., 25, 307-315. Nyska, M. (2002). The Unstable Ankle. Human Kinetic publisher Inc 2002. O'Kane, F. W., McGibbon, C. A., & Krebs, D. E. (2003). Kinetic analysis of planned gait termination in healthy subjects and patients with balance disorders. Gait.Posture., 17, 170-179. Osborne, M. D. & Rizzo, T. D., Jr. (2003). Preventi on and treatment of ankl e sprain in athletes. Sports Med., 33, 1145-1150. Perotto A, D. E. I. J. M. D. (1994). Anatomical guide for the electromyographer. [3rd edition]. Springfield, MO, Springfie ld, MO: Charles Thompson. Rasmussen, O. & Tovborg-Jensen, I. (1982). Mobility of the ankle joint: recording of rotatory movements in the talocrural joint in vitro wi th and without the lateral collateral ligaments of the ankle. Acta Orthop.Scand., 53, 155-160. Riemann, B. L. & Lephart, S. M. (2002). The Se nsorimotor System, Part I: The Physiologic Basis of Functional Joint Stability. J.Athl.Train., 37, 71-79. Ross, S. E. & Guskiewicz, K. M. (2004c). Examinat ion of static and dynamic postural stability in individuals with f unctionally stable and unstable ankles. Clin.J.Sport Med., 14, 332-338. Ross, S. E. & Guskiewicz, K. M. (2004a). Examinat ion of static and dynamic postural stability in individuals with f unctionally stable and unstable ankles. Clin.J.Sport Med., 14, 332-338.

PAGE 86

86 Ross, S. E. & Guskiewicz, K. M. (2004b). Examin ation of static and dynamic postural stability in individuals with functionally stable and unstable ankles. Clin.J.Sport Med., 14, 332338. Ross, S. E. & Guskiewicz, K. M. (2004d). Examin ation of static and dynamic postural stability in individuals with functionally stable and unstable ankles. Clin.J.Sport Med., 14, 332338. Ross, S. E., Guskiewicz, K. M., Gross, M. T., & Yu, B. (2008). Assessment tools for identifying functional limitations associated with functional a nkle instability. J.Athl.Train., 43, 4450. Ross, S. E., Guskiewicz, K. M., & Yu, B. (2005). Single-leg jump-landing stabilization times in subjects with functionally unstable ankles. J.Athl.Train., 40, 298-304. Rozzi, S. L., Lephart, S. M., St erner, R., & Kuligowski, L. ( 1999). Balance training for persons with functionally unstable ankles. J.Orthop.Sports Phys.Ther., 29, 478-486. Smith, R. W. & Reischl, S. F. (1986). Tr eatment of ankle sprains in young athletes. Am.J.Sports Med., 14, 465-471. Sparrow, W. A. & Tirosh, O. (2005). Gait termina tion: a review of experimental methods and the effects of ageing and gait pathologies. Gait.Posture., 22, 362-371. Stein, R. B. & Hase, K. (1999). Stop ping and turning during human walking. Prog.Brain Res., 123, 445-453. Sugimoto, K., Samoto, N., Takakura, Y., & Tamai, S. (1997). Varus tilt of the tibial plafond as a factor in chronic ligament instability of the ankle. Foot Ankle Int., 18, 402-405. Teramoto, A., Kura, H., Uchiyama, E., Suzuki, D., & Yamashita, T. (2008). Three-dimensional analysis of ankle instability after tibiofi bular syndesmosis injuries: a biomechanical experimental study. Am.J.Sports Med., 36, 348-352. Tirosh, O. & Sparrow, W. A. (2004). Gait term ination in young and olde r adults: effects of stopping stimulus probability and stimulus delay. Gait.Posture., 19, 243-251. Tropp, H. (1986). Pronator muscle weakness in f unctional instability of the ankle joint. Int.J.Sports Med., 7, 291-294. Tropp, H. (2002). Commentary: Functi onal Ankle Instability Revisited. J.Athl.Train., 37, 512515. Verhagen, R. A., de Keizer, G., & van Dijk, C. N. (1995). Long-term follow-up of inversion trauma of the ankle. Arch.Orthop.Trauma Surg., 114, 92-96.

PAGE 87

87 Wearing, S. C., Urry, S., Smeathers, J. E., & Battistutta, D. (1999). A comparison of gait initiation and termination methods for obtaining plantar foot pressures. Gait.Posture., 10, 255-263. Wikstrom, E. A., Tillman, M. D., & Borsa, P. A. (2005a). Detection of dynamic stability deficits in subjects with functional ankle instability. Med.Sci.Sports Exerc., 37, 169-175. Wikstrom, E. A., Tillman, M. D., Chmielewski, T. L., & Borsa, P. A. (2006). Measurement and evaluation of dynamic joint stability of the knee and ankle after injury. Sports Med., 36, 393-410. Wikstrom, E. A., Tillman, M. D., Chmielewski, T. L., Cauraugh, J. H., & Borsa, P. A. (2007). Dynamic postural stability deficits in subj ects with self-report ed ankle instability. Med.Sci.Sports Exerc., 39, 397-402. Wikstrom, E. A., Tillman, M. D., Kline, K. J., & Borsa, P. A. (2006). Gender and limb differences in dynamic postural stability during landing. Clin.J.Sport Med., 16, 311-315. Wikstrom, E. A., Tillman, M. D., Smith, A. N ., & Borsa, P. A. (2005b). A new force-plate technology measure of dynamic postural stabil ity: the dynamic postural stability index. J.Athl.Train., 40, 305-309. Yeung, M. S., Chan, K. M., So, C. H., & Yuan, W. Y. (1994). An epidemiological survey on ankle sprain. Br.J.Sports Med., 28, 112-116.

PAGE 88

88 BIOGRAPHICAL SKETCH I was born in 1981 in a city called Pune, in In dia, to Mr. Sanjay Suresh Hatttangadi and Mrs. Geetan jali Sanjay Hattanga di. I completed my primary and high school education in Pune and moved to the city of Mumbai to comple te my professional education. I completed by bachelors degree in physical therapy from Lo kmanya Tilak Municipal Medical College in Mumbai. These four years gave me the opportunity to receive the highest standard of education in my field from one of the prestigious and hi ghly regarded colleges in the India for physical therapy education. These four year s were the most memorable year s of my life. Becoming one of the best clinicians has been my goal ever since. I decided to pursue my graduate education here in the United States of America. I was lucky to be able to complete my masters education in the field of biomechanics at the University of Florid a. This unique experience as helped me grow at an individual and professional level. It was i ndeed a great academic and cultural experience.