ITEM METADATA RECORD
Title: Proprioception. Associations with low back pain and physical activity.
Other Titles: Proprioceptie. Associaties met lage rugpijn en fysieke activiteit.
Authors: Kiers, Henderikus; S0166884
Issue Date: 21-Nov-2014
Abstract: General overviewThis thesis describes a series of studies into the relationship between low back pain (LBP), proprioceptive postural control and physical activity. Two central elements in this thesis are the UP-LIFT cohort (Utrecht Police Lifestyle Intervention Fitness and Training study) and the experimental set-up. The UP-LIFT cohort consist of employees of the Utrecht police department in the Netherlands (n=1723), aged between 18 and 62 years, who visited the health research department at the University of Applied Sciences Utrecht for a fitness and lifestyle evaluation. In this evaluation a broad range of parameters were tested, concerning cardio vascular risk factors, motivational and social-cognitive determinants regarding a healthy lifestyle, physical fitness (peak VO2, muscle strength and endurance, flexibility) and physical activity (kind, intensity and duration).Random samples out of this cohort wereenrolled in the experimental set up used in the studies in this thesis.To quantify proprioceptive postural control (the way proprioception is used in postural control) we used a series of 7 trials of 60 seconds. Inall trials participants stood barefoot on a force plate. In all but thefirst and the last trial, vision was occluded by means of taped safety glasses. In trials 1 to 3 subjects stood relaxed without perturbations, while in trial 3 a foam pad was place over the force plate. The Centre of Pressure (CoP), which is the location of the ground reaction force acting on the body, was measured 200 times per second. Human beings never stand completely still. With the use of the CoP these movements, known aspostural sway, can be measured. In trials 4 to 8, muscle vibration was applied between the 15th and the 30th second, in all combinations of m. Triceps Surea (TSM) and Lumbar Paravertebral Muscles (LPM) vibration andstanding on the force plate (rigid surface) or on a foam pad placed over the force plate (unstable surface). Muscle vibration induces an afferent signal from the spindles within the muscle that corresponds to a lengthening of the muscle. Signals from all muscles spindles surrounding a joint are considered to be the main source of information about the position of the joint. Therefore changes in CoP location during muscle vibration provide an indication of the impact the proprioceptive signals from the vibrated muscle spindles exhibit on the integration of all sources of information (vestibular, visual, proprioceptive) in maintaining postural control. Part I. Introduction and methodologyChapter 1 provides a broad introduction into LBP, proprioception and proprioceptivepostural control, and the influence of physical activity on those entities. Chapter 2 & 3 describe two studies into the characteristics and measurement properties of the experimental set-up used in this thesis. In Chapter 2 we studied the influence of standing on an unstable surface, i.e. foam, on muscle vibration effects. The first 100 subjects from the UP-LIFT cohort performing the proprioceptive postural control tests were enrolled. We compared the effect TSM and LPM vibration had on CoP position and velocity when standing on a rigid surface with that when standing on a foam pad. The results showed that on foam the effect of TSM vibration was significantly smaller than on a solid surface, bothfor CoP velocity as for change in CoP position under vibration, while for LPM vibration the effects were reversed to that of TSM vibration. We hypothesized that this is caused by a decreased weighting of proprioceptive signals originating from the ankle musculature, and an up weighting of proprioceptive signals originating from lumbar spinal musculature when standing on foam. Therefore we concluded that exercises on unstable surfaces do not appear to target peripheral ankle proprioception, but may challenge the capacity of the central nervous system to shift the weighting of sources of proprioceptive signals between body regions.In Chapter 3, the results of a reliability study of a broad range of possible outcome variables of muscle vibration effects are presented. This study was conducted among 20 students, staff and family members of the students. Change in mean position of the CoP (’displacement’) during vibration showed good reliability (ICC’s > 0.6). Ratios of displacement between LPM and TSM vibration (‘proprioceptive weighting’) showed fair to good reliability (0.52–0.73). Change in CoP velocity under influence of vibration appeared not to be reliable. Balance recovery, when calculated based CoP position a short period after cessation of vibration, showed good reliability (ICC’s >0.6). Agreement measures were poor, with most CV’s ranging between 18% and 36%. Expressing variables relative to the limits of stability did not improve reliability or agreement. According to this study, displacement during vibration, proprioceptive weighting and selected recovery variables are the most reliable indicatorsof the response to muscle vibration. In the present form these variables do not seem suitable for use in clinical practice.Part II. Low back pain and proprioceptive postural controlIn Part II, the findings in the UPLIFT study among 215 policemen and women regarding proprioceptive postural control variables and their association with LBP are presented (Chapter 4). In this cross-sectional study, proprioceptive postural control and its relationship to LBP were investigated by means of muscle vibration effects and by postural sway. Postural sway was expressed as the structure, range and velocity of the CoP. To reduce the large amount of possible variables to quantify muscle vibration responses and posturalsway under different conditions, we first performed three factor analyses, in which we only entered variables that showed at least a fair reliability in Chapter 3. A biologically plausible construct could be appointed to every one of the in total 10 factors. For postural sway these werefrequency and irregularity on rigid surface (1), velocity and range on rigid (2), frequency and irregularity on foam (3), and velocity and range on foam (4). For muscle vibration effects these were response to TSM vibration on rigid and foam (1), response to LPM vibration on foam (2), and response to LPM vibration on a rigid surface (3). For recovery after vibration cessation these were recovery standing on a solid surface (1),recovery standing on a foam surface (2), and peak recovery, the maximumanterior CoP position after cessation of vibration (3).The analysisof possible associations demonstrated that subjects with LBP sway with comparable amplitudes as subjects without LBP, but the structure of their sway pattern was less regular with higher frequency content. Subjects with LBP also showed a smaller response to TSM vibration, a non significant smaller response to LPM vibration, and a slower balance recovery after cessation of vibration when standing on a solid surface. There was a weak but significant association between smaller responses to TSM vibration and an irregular, high frequency sway pattern, independent from LBP. Based on these findings, we proposed a model for control of postural sway. This model suggests that subjects with LBP use more co-contraction and less cognitive control, to maintain a standing balance when compared to subjects without LBP. Cognitive attention may, in subjects with LBP under challenging conditions, be directed towards the lower back. In addition, a generally reduced weighting of proprioceptive signals in subjects with LBP is suggested as an explanation for the findings in muscle vibration. Part III. The influence of physical activityPart III describes two studies on the influence of physical activity on proprioceptive control. The first study is a systematic review into the relationship between physical activity and postural sway (Chapter 5). The search of the literature retrieved 39 studies, 37 with a comparativedesign, one designed as a cohort study, and one as a randomized controlled trial. The main conclusion in this study was that in general sport practitioners sway less than controls, and high-level athletes sway less than low-level athletes. Additionally, we identified specific effects dependent on the use of vision, sport specific postures, and frequency andduration of the (sports) activity. From Chapter 5 we conclude that postural sway in unperturbed bipedal stance does not seem suitable to detect subtle differences between groups of healthy people. The effects of sports activity on postural sway are specific to the characteristics of the sport, and become more manifest under challenging conditions. This is in line with our own studies, in which we also found the clearest differences between groups when standing on foam or during muscle vibration. This could have clinical implications, as clinical examinations often take place in standardized, non-challenging conditions. Perhaps it would be better to evaluate movement behaviour of subjects with LBP under more stressful conditions. In Chapter 6 we were interested inthe influence of physical activity and aerobic fitness on the relationships found in Part III. For this study we used data of the same 215 subjects enrolled in the UP-LIFT study. We asked subjects for kind, level and duration of physical activity by means of two questionnaires, the SQUASH and a custom made questionnaire. We used peak oxygen uptake (VO2 per kilogram), resulting from the ergometer test in the UP-LIFT study as measure for aerobic fitness. There were two major findings: a trend towardsa higher response to muscle vibration in subjects with a higher level of aerobic fitness (p= 0.06), and a quadratic association between aerobicfitness and sway irregularity (low and high level of fitness associatedwith a more regular sway pattern with lower frequency content). With respect to physical activity a trend towards the same associations as for physical fitness could be seen, but these associations never reached statistical significance. The reason for this could well be the methodological shortcomings of self-reported physical activity measures. Apossible explanation for the increased sensitivity to muscle vibration in subjects with high levels of peak VO2, could be an improved muscle oxygenation. The subjects with high levels of peak VO2 also showed a more regular sway pattern. As it is likely that these subjects are more competitive, we suggest this could be caused by a supraspinal control strategy, which implies an exploratory behaviour. At the other end of the peak VO2 spectrum, the more regular sway pattern withlower frequency content found in subjects with low levels of peak VO2 could be due to more attention to the postural task than other subjects. It has been shown that adequate responses to changes in the environment during a postural task depend on the amount of physical activity subjetcs exhibit in daily life. The changes in control strategies in the previous studies seen in subjects with LBP were in our study independent from changes in peak VO2 and physical activity level. PartIV. Discussion and summariesIn Part IV all results of the aforementioned studies are discussed as a whole. We noticed some issues that could not be fully explained by contemporary views on proprioceptive weighting. The first issue is that the initial velocity of the CoP towards the final CoP position during vibration was similar for both the condition on a rigid surface as on foam. A down weighting of proprioceptive signals from the ankle muscles on foam, as suggested, would lead to a smaller resulting moment, thus to a smaller CoP velocity in the initial vibration epoch towards the final CoP position. This is even more emphasized by the mechanical properties of foam. On foam a comparable neural drive to alpha motor neurons would lead to a lower moment around the ankle due to the deformation of the foam pad. The second issue is that in some studies where fingertip contact without mechanical support was allowed during vibration, fingertip contact decreased the effect of muscle vibration when standing on a rigid surface, but increased the muscle vibration effect when standing on an unstable surface. The latter cannot be explained by a down weighting of proprioceptive signals based on their reliability. Fingertip touch would add extra information to the proprioceptive system, which will make even clearer that signals originating from the ankles are not reliable when standing on an unstable surface.The third issue is the common opinion that the CNS is able to gate sensory input in accordance to their reliability. However, the brain never has direct access to the true parameter values but only has access tothe data from which it makes inferences. The explanation that signals from the ankle region are less informative when standing on foam is teleological in nature, but does not supply an explanation of how this weighting is achieved. If the CNS is not a contemporary form of the homunculus, who is the one who decides? A possible solution to these problems is based upon the model for postural control presented in Part II (Chapter 3). This model includes 3 control strategies, co-contraction, spinal proprioceptive feedback and supraspinal control. We propose a further refinement of this model in which we integrate the original model with aspects of the Equilibrium Point (EP) theory and of Bayesian probability statistics. In short, the CNS can set joint stiffness by co-contraction and sensitivity of the proprioceptive reflex. Decreasing thethresholds of all muscles, leaving the vector summation of all spindle thresholds surrounding the joint unchanged, sets the sensitivity of the proprioceptive reflex.In an independent process, the CNS sets the planned joint angle by determining the ratio of the muscle spindle activation thresholds of all muscles surrounding a joint. Which joint angle is most appropriate is the resultant of a central weighing process with other sources of sensory information in relation to the desired task. The relative weight of every sensory source depends on the signal noise ratio. When discussing Part III, we looked deeper into the differences between subjects with and subjects without LBP when under vibration. The initial response to vibration is equal in both groups, but the initial response to cessation of LPM vibration is much larger in subjects with LBP when standing on foam, as is the initial response to cessation of TSM vibration when standing on a rigid surface. We explain this according to the refined model as a higher peripheral reflex sensitivity in subjects with LBP, but a smaller weighting in the centrally mediated controlof position. The smaller weighting of proprioceptive signals in subjects with LBP is explained by a lower signal noise ratio from peripheral afference due to a comprised proprioceptive signal or due to higher peripheral stiffness. With regard to Chapter 6, we suggest that a smaller microcirculation in subjects with lower levels of aerobic fitness (peak VO2) causes more variance in the afferent peripheral signal, with consequently a down weighting of the signal in the centrally mediated position control. The lack of differences between groups in response to cessation of vibration could be due to higher levels of co-contraction and reflex sensitivity in both high and low levels of aerobic fitness. This ideais strengthened by the findings in postural sway structure, which is for both ends of the spectrum more regular and with a lower frequency content than for subjects with average levels of peak VO2.This refined model has been developed retrospectively. Obviously it has to be tested prospectively. Recommendations for further research are made based upon the predictions of this model about stretch reflexes and proprioceptive weighting based on variation of the peripheral signal. /* Font Definitions */@font-face {font-family:Arial; panose-1:2 11 6 4 2 2 2 2 2 4; mso-font-charset:0; mso-generic-font-family:auto; mso-font-pitch:variable; mso-font-signature:3 0 0 0 1 0;}@font-face {font-family:宋体; panose-1:0 0 0 0 0 0 0 0 0 0; mso-font-charset:134; mso-generic-font-family:auto; mso-font-format:other; mso-font-pitch:variable; mso-font-signature:1 135135232 16 0 262144 0;}@font-face {font-family:"Cambria Math"; panose-1:2 4 5 3 5 4 6 3 2 4; mso-font-charset:0; mso-generic-font-family:auto; mso-font-pitch:variable; mso-font-signature:3 0 0 0 1 0;} /* Style Definitions */p.MsoNormal, li.MsoNormal, div.MsoNormal {mso-style-unhide:no; mso-style-qformat:yes; mso-style-parent:""; margin:0cm; margin-bottom:.0001pt; text-align:justify; text-justify:inter-ideograph; line-height:150%; mso-pagination:widow-orphan; font-size:10.0pt; font-family:Helvetica; mso-fareast-font-family:宋体; mso-fareast-theme-font:minor-fareast; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi; mso-ansi-language:NL-BE;}h3 {mso-style-priority:9; mso-style-qformat:yes; mso-style-link:"Heading 3 Char"; mso-style-next:Normal; margin-top:12.0pt; margin-right:0cm; margin-bottom:0cm; margin-left:0cm; margin-bottom:.0001pt; mso-add-space:auto; text-align:justify; text-justify:inter-ideograph; line-height:150%; mso-pagination:widow-orphan; mso-outline-level:3; font-size:12.0pt; font-family:Helvetica; font-variant:small-caps; letter-spacing:.25pt; mso-ansi-language:NL-BE; font-weight:normal;}h3.CxSpFirst {mso-style-priority:9; mso-style-qformat:yes; mso-style-link:"Heading 3Char"; mso-style-next:Normal; mso-style-type:export-only; margin-top:12.0pt; margin-right:0cm; margin-bottom:0cm; margin-left:0cm; margin-bottom:.0001pt; mso-add-space:auto; text-align:justify; text-justify:inter-ideograph; line-height:150%; mso-pagination:widow-orphan; mso-outline-level:3; font-size:12.0pt; font-family:Helvetica; font-variant:small-caps; letter-spacing:.25pt; mso-ansi-language:NL-BE; font-weight:normal;}h3.CxSpMiddle {mso-style-priority:9; mso-style-qformat:yes; mso-style-link:"Heading 3 Char"; mso-style-next:Normal; mso-style-type:export-only; margin:0cm; margin-bottom:.0001pt; mso-add-space:auto; text-align:justify; text-justify:inter-ideograph; line-height:150%; mso-pagination:widow-orphan; mso-outline-level:3; font-size:12.0pt; font-family:Helvetica; font-variant:small-caps; letter-spacing:.25pt; mso-ansi-language:NL-BE; font-weight:normal;}h3.CxSpLast {mso-style-priority:9; mso-style-qformat:yes; mso-style-link:"Heading 3 Char"; mso-style-next:Normal; mso-style-type:export-only; margin:0cm; margin-bottom:.0001pt; mso-add-space:auto; text-align:justify; text-justify:inter-ideograph; line-height:150%; mso-pagination:widow-orphan; mso-outline-level:3; font-size:12.0pt; font-family:Helvetica; font-variant:small-caps; letter-spacing:.25pt; mso-ansi-language:NL-BE; font-weight:normal;}span.Heading3Char {mso-style-name:"Heading 3 Char"; mso-style-priority:9; mso-style-unhide:no; mso-style-locked:yes; mso-style-link:"Heading 3"; mso-ansi-font-size:12.0pt; mso-bidi-font-size:12.0pt; font-family:Helvetica; mso-ascii-font-family:Helvetica; mso-hansi-font-family:Helvetica; font-variant:small-caps; letter-spacing:.25pt; mso-ansi-language:NL-BE;}.MsoChpDefault {mso-style-type:export-only; mso-default-props:yes; font-size:10.0pt; mso-ansi-font-size:10.0pt; mso-bidi-font-size:10.0pt; font-family:Calibri; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-fareast-font-family:宋体; mso-fareast-theme-font:minor-fareast; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi; mso-ansi-language:NL;}.MsoPapDefault {mso-style-type:export-only; margin-bottom:10.0pt; text-align:justify; text-justify:inter-ideograph; line-height:115%;}@page WordSection1 {size:612.0pt 792.0pt; margin:72.0pt 90.0pt72.0pt 90.0pt; mso-header-margin:36.0pt; mso-footer-margin:36.0pt; mso-paper-source:0;}div.WordSection1 {page:WordSection1;}-->General overviewThis thesis describes a series of studiesinto the relationship between low back pain (LBP), proprioceptive posturalcontrol and physical activity. Two central elements in this thesis are the UP-LIFTcohort (UtrechtPolice Lifestyle Intervention Fitness and Training study) andthe experimental set-up. The UP-LIFT cohort consist of employees of the Utrechtpolice department in the Netherlands (n=1723), aged between 18 and 62 years,who visited the health research department at the University of Applied SciencesUtrecht for a fitness and lifestyle evaluation.In this evaluation a broadrange of parameters were tested, concerning cardio vascular risk factors,motivational and social-cognitive determinants regarding a healthy lifestyle,physical fitness (peak Arial">VO2, muscle strength andendurance, flexibility) and physical activity(kind, intensity and duration).Random samples out of this cohort wereenrolled in the experimental set up used in the studies in this thesis. Toquantify proprioceptive postural control (the way proprioception is used inpostural control) we used a series of 7 trials of60 seconds. In all trialsparticipants stood barefoot on a force plate. In all but the first and the lasttrial, vision was occluded by means of taped safety glasses. In trials 1 to 3subjects stood relaxedwithout perturbations, while in trial 3 a foam pad wasplace over the force plate. The Centre of Pressure (CoP), which is the locationofthe ground reaction force acting on the body, was measured 200 times persecond. Human beings never stand completely still. With the use of the CoPthese movements, known as postural sway, can be measured. In trials 4 to 8,muscle vibration was applied between the 15th and the 30th second, in allcombinations of m. Triceps Surea (TSM) and LumbarParavertebral Muscles (LPM)vibration and standing on the force plate (rigid surface) or on a foam padplaced over the force plate (unstable surface). Muscle vibration induces anafferent signal from the spindles within the muscle that corresponds to alengthening of the muscle. Signals from all muscles spindles surrounding ajoint are considered to be the main source of information about the position ofthe joint. Therefore changes in CoP location during muscle vibration provide anindication of the impact the proprioceptive signals from the vibrated musclespindles exhibit on the integration of all sources of information (vestibular,visual, proprioceptive) in maintaining postural control. Part I. Introduction and methodologyChapter 1 provides a broad introductioninto LBP, proprioception and proprioceptive postural control, and the influenceof physical activity on those entities. Chapter 2 & 3 describe two studiesinto the characteristics and measurement properties of the experimental set-upused in this thesis. In Chapter 2 we studied the influence ofstanding on an unstable surface, i.e. foam, on muscle vibration effects. Thefirst 100 subjects from the UP-LIFT cohort performing the proprioceptivepostural control tests were enrolled. We compared the effect TSM and LPMvibration had on CoP position and velocity when standing on a rigid surfacewith that when standing on a foam pad. The results showed that on foam theeffect of TSM vibration was significantly smaller than on a solid surface, bothfor CoP velocity as for change in CoP position under vibration, while for LPMvibration the effects were reversed to that of TSM vibration. We hypothesizedthat this is caused by a decreased weighting of proprioceptive signalsoriginating from the ankle musculature, and an up weighting of proprioceptivesignals originating from lumbar spinal musculature when standing on foam.Therefore we concluded that exercises on unstable surfaces do not appear totarget peripheral ankle proprioception, but may challenge the capacity of thecentral nervous system to shift the weighting of sources of proprioceptivesignals between body regions.In Chapter 3, the results of a reliabilitystudy of a broad range of possibleoutcome variables of muscle vibration effectsare presented. This study was conducted among 20 students, staff and familymembers of the students. Change in mean position of the CoP(’displacement’) during vibration showed good reliability (ICC’s > 0.6).Ratios of displacement between LPM and TSM vibration (‘proprioceptiveweighting’) showed fair to good reliability (0.52–0.73). Change in CoP velocityunder influence of vibration appeared not to be reliable. Balance recovery,when calculated based CoP position a short period aftercessation of vibration,showed good reliability (ICC’s >0.6). Agreement measures were poor, withmost CV’s ranging between 18% and 36%. Expressing variables relative to thelimits of stability did not improve reliability or agreement. According to thisstudy, displacement during vibration, proprioceptive weighting and selectedrecovery variables are the most reliable indicators of the response to musclevibration. In the present form these variables do not seem suitable for use inclinical practice.Part II. Low back pain andproprioceptive postural controlIn Part II, the findings in the UPLIFTstudy among 215 policemen and women regarding proprioceptive postural controlvariables and their association with LBP are presented (Chapter 4). In thiscross-sectional study, proprioceptive postural controland its relationship toLBP were investigated by means of muscle vibration effects and by posturalsway. Postural sway was expressed as the structure, range and velocity of theCoP. To reduce the large amount of possible variables to quantify musclevibration responses and postural sway under different conditions, we firstperformed three factor analyses, in which we only entered variables that showedat leasta fair reliability in Chapter 3. A biologically plausible constructcould be appointed to every one of the in total 10 factors. For posturalswaythese were frequency and irregularity on rigid surface (1), velocity and rangeon rigid (2), frequency and irregularity on foam (3),and velocity and range onfoam (4). For muscle vibration effects these were response to TSM vibration onrigid and foam (1), response to LPM vibration on foam (2), and response to LPMvibration on a rigid surface (3). For recovery after vibration cessation thesewere recovery standing on a solid surface (1), recovery standing on a foamsurface (2), and peak recovery, the maximum anterior CoP position aftercessation of vibration (3).The analysis of possible associationsdemonstrated that subjects with LBP sway with comparable amplitudes as subjectswithout LBP, but the structure of their sway pattern was less regular withhigher frequency content. Subjects with LBP also showed a smaller response toTSM vibration, a non significant smaller response to LPM vibration, and aslower balance recovery after cessation of vibration when standing on a solidsurface. There was a weak but significant association between smaller responsesto TSM vibration andan irregular, high frequency sway pattern, independentfrom LBP. Based on these findings, we proposed a modelfor control ofpostural sway. This model suggests that subjects with LBP usemore co-contraction and less cognitive control, to maintain a standing balancewhen compared to subjects without LBP. Cognitive attention may, in subjectswith LBP under challenging conditions, be directed towards the lower back. Inaddition, a generally reduced weighting of proprioceptive signals in subjectswith LBP is suggested as an explanation forthe findings in muscle vibration. Part III. The influence ofphysical activityPart III describes two studies on theinfluence of physical activity on proprioceptive control. The first study is asystematic review into the relationship between physical activity and posturalsway (Chapter 5). The search of the literature retrieved 39 studies, 37 with acomparative design, one designed as a cohort study, and one as a randomizedcontrolled trial. The main conclusion in this study was that in general sportpractitioners sway less than controls, and high-level athletes sway less than low-levelathletes. Additionally, we identified specific effects dependent on the use ofvision, sport specific postures, and frequency and duration of the (sports)activity. From Chapter 5 we conclude that posturalsway in unpertur
bed bipedal stance does not seem suitable to detectsubtledifferences between groups of healthy people. The effects of sports activity onpostural sway are specific to the characteristics of the sport, and become moremanifest under challenging conditions. This is in line with our own studies, inwhich we also found the clearest differences between groups when standing onfoam or during muscle vibration. This could have clinical implications, asclinical examinations often take place in standardized, non-challengingconditions.Perhaps it would be better to evaluate movement behaviour ofsubjects with LBP under more stressful conditions.In Chapter 6 we were interested in theinfluence of physical activity and aerobic fitness on the relationships foundin Part III. For this study we used data of the same 215 subjects enrolled inthe UP-LIFT study. We asked subjects for kind, level and duration of physicalactivity by means of twoquestionnaires, the SQUASH and a custom madequestionnaire. We used peak oxygen uptake (VO2 per kilogram), resulting from the ergometer testin the UP-LIFTstudy as measure for aerobic fitness. There were two major findings: a trendtowards a higher response to muscle vibrationin subjects with a higher levelof aerobic fitness (p= 0.06), and a quadratic association between aerobicfitness and sway irregularity (low and high level of fitness associated with amore regular sway pattern with lower frequency content). With respect tophysical activitya trend towards the same associations as for physical fitnesscould be seen, but these associations never reached statistical significance.The reason for this could well be the methodological shortcomings of self-reportedphysical activity measures. A possible explanation for the increasedsensitivity to muscle vibration in subjects with high levels of peak VO2, could be an improved muscle oxygenation. The subjects with highlevels of peak VO2 also showed a more regularsway pattern. As it is likely that these subjects are morecompetitive, wesuggest this could be caused by a supraspinal control strategy, which impliesan exploratory behaviour. At the other end of the peak VO2 spectrum, the more regular sway pattern with lower frequencycontent found in subjects with low levels of peak VO2 could be due to more attention to the postural task than othersubjects. It has been shown that adequate responses to changes intheenvironment during a postural task depend on the amount of physical activitysubjetcs exhibit in daily life. The changes in control strategies in the previousstudies seen in subjects with LBP were in our study independent from changes inpeak VO2 and physical activity level. Part IV. Discussion and summariesIn PartIV all results of theaforementioned studies are discussed as a whole. We noticed some issues thatcould not be fully explained by contemporary views on proprioceptive weighting.The first issue is that the initialvelocity of the CoP towards the final CoP position during vibration was similarfor both the condition on a rigid surfaceas on foam. A down weighting ofproprioceptive signals from the ankle muscles on foam, as suggested, would leadto a smaller resulting moment, thus to a smaller CoP velocity in the initialvibration epoch towards the final CoP position. This is even more emphasized bythe mechanical properties of foam. On foam a comparable neural drive to alphamotor neurons would lead to a lower moment around the ankle due to thedeformation of the foam pad. The second issue is that in some studieswhere fingertip contact without mechanical support was allowed duringvibration, fingertip contact decreased the effect ofmuscle vibration whenstanding on a rigid surface,but increased themuscle vibration effect whenstanding on an unstable surface. The latter cannot be explained by a downweighting of proprioceptive signals based on their reliability. Fingertip touchwould add extra information to the proprioceptive system, which will make evenclearer that signals originating from the ankles are not reliable when standingon an unstable surface.The third issue is the common opinion thatthe CNS is able to gate sensory input in accordance to their reliability.However, the brain never has direct access to the true parameter values butonly has access to the data from which it makes inferences.The explanationthat signals from the ankle region are less informative when standing on foamis teleological in nature, but does not supply an explanation of how this weightingis achieved. If the CNS is not a contemporary form of the homunculus, who isthe one who decides? A possible solution to these problems isbased upon themodel for postural control presented in Part II (Chapter 3).This model includes 3 control strategies, co-contraction, spinal proprioceptivefeedback and supraspinal control. We propose a further refinement ofthis modelin which we integrate the original model with aspects of the Equilibrium Point(EP) theory and of Bayesian probability statistics. In short, the CNS can set joint stiffnessby co-contraction and sensitivity of the proprioceptive reflex. Decreasing thethresholds of all muscles, leaving the vector summation of all spindlethresholds surrounding the joint unchanged, sets the sensitivity of theproprioceptive reflex.In an independent process, the CNS sets theplanned joint angle by determining the ratio of the muscle spindle activationthresholds of all muscles surrounding a joint. Whichjoint angle is mostappropriate is the resultant of a central weighing process with other sourcesof sensory information in relation to the desired task. The relative weight ofevery sensory source depends on the signal noise ratio. When discussing Part III, we looked deeperinto the differences between subjects with and subjects without LBP when undervibration. The initial response to vibration is equal in both groups, but theinitial response to cessation of LPM vibration is much larger in subjects withLBP when standing on foam, asis the initial response to cessation of TSMvibration when standing on a rigid surface. We explain this according to therefined model asa higher peripheral reflex sensitivity in subjects with LBP,but a smaller weighting in the centrally mediated control of position. Thesmaller weighting of proprioceptive signals in subjects with LBP is explainedby a lower signal noise ratio from peripheral afference due to acomprisedproprioceptive signal or due to higher peripheral stiffness. With regard toChapter 6, we suggest that a smaller microcirculation in subjects with lowerlevels of aerobic fitness (peak Arial">VO2) causes more variancein the afferent peripheral signal, with consequently a down weighting of thesignal in the centrally mediated position control. The lack of differencesbetween groups in response to cessation of vibration could be due to higherlevels of co-contraction and reflex sensitivity in both high and low levels ofaerobic fitness. This idea is strengthened by the findings in postural swaystructure, which is for both ends of the spectrum more regular and with a lowerfrequency content than for subjects with average levels of peak VO2.This refined model has been developedretrospectively. Obviously it has to be tested prospectively. Recommendationsfor further research are made based upon the predictions of this model aboutstretch reflexes and proprioceptive weighting based on variation of theperipheral signal. /* Font Definitions */@font-face {font-family:Arial; panose-1:2 11 6 4 2 2 2 2 2 4; mso-font-charset:0; mso-generic-font-family:auto; mso-font-pitch:variable; mso-font-signature:3 0 0 0 1 0;}@font-face {font-family:宋体; panose-1:0 0 0 0 0 0 0 0 0 0; mso-font-charset:134; mso-generic-font-family:auto; mso-font-format:other; mso-font-pitch:variable; mso-font-signature:1 135135232 16 0 262144 0;}@font-face {font-family:宋体; panose-1:0 0 0 0 0 0 0 0 0 0; mso-font-charset:134; mso-generic-font-family:auto; mso-font-format:other; mso-font-pitch:variable; mso-font-signature:1 135135232 16 0 262144 0;} /* Style Definitions */p.MsoNormal, li.MsoNormal, div.MsoNormal {mso-style-unhide:no; mso-style-qformat:yes; mso-style-parent:""; margin:0cm; margin-bottom:.0001pt; text-align:justify; text-justify:inter-ideograph; line-height:150%; mso-pagination:widow-orphan; font-size:10.0pt; font-family:Helvetica; mso-fareast-font-family:宋体; mso-fareast-theme-font:minor-fareast; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi; mso-ansi-language:NL-BE;}h3 {mso-style-priority:9; mso-style-qformat:yes; mso-style-link:"Heading 3 Char"; mso-style-next:Normal; margin-top:12.0pt; margin-right:0cm; margin-bottom:0cm; margin-left:0cm; margin-bottom:.0001pt; mso-add-space:auto; text-align:justify; text-justify:inter-ideograph; line-height:150%; mso-pagination:widow-orphan; mso-outline-level:3; font-size:12.0pt; font-family:Helvetica; font-variant:small-caps; letter-spacing:.25pt; mso-ansi-language:NL-BE; font-weight:normal;}h3.CxSpFirst {mso-style-priority:9; mso-style-qformat:yes; mso-style-link:"Heading 3 Char"; mso-style-next:Normal; mso-style-type:export-only; margin-top:12.0pt; margin-right:0cm; margin-bottom:0cm; margin-left:0cm; margin-bottom:.0001pt; mso-add-space:auto; text-align:justify; text-justify:inter-ideograph; line-height:150%; mso-pagination:widow-orphan; mso-outline-level:3; font-size:12.0pt; font-family:Helvetica; font-variant:small-caps; letter-spacing:.25pt; mso-ansi-language:NL-BE; font-weight:normal;}h3.CxSpMiddle {mso-style-priority:9; mso-style-qformat:yes; mso-style-link:"Heading 3 Char"; mso-style-next:Normal; mso-style-type:export-only; margin:0cm; margin-bottom:.0001pt; mso-add-space:auto; text-align:justify; text-justify:inter-ideograph; line-height:150%; mso-pagination:widow-orphan; mso-outline-level:3; font-size:12.0pt; font-family:Helvetica; font-variant:small-caps; letter-spacing:.25pt; mso-ansi-language:NL-BE; font-weight:normal;}h3.CxSpLast {mso-style-priority:9; mso-style-qformat:yes; mso-style-link:"Heading 3 Char"; mso-style-next:Normal; mso-style-type:export-only; margin:0cm; margin-bottom:.0001pt; mso-add-space:auto; text-align:justify; text-justify:inter-ideograph; line-height:150%; mso-pagination:widow-orphan; mso-outline-level:3; font-size:12.0pt; font-family:Helvetica; font-variant:small-caps; letter-spacing:.25pt; mso-ansi-language:NL-BE; font-weight:normal;}span.Heading3Char {mso-style-name:"Heading 3 Char"; mso-style-priority:9; mso-style-unhide:no; mso-style-locked:yes; mso-style-link:"Heading 3"; mso-ansi-font-size:12.0pt; mso-bidi-font-size:12.0pt; font-family:Helvetica; mso-ascii-font-family:Helvetica; mso-hansi-font-family:Helvetica; font-variant:small-caps; letter-spacing:.25pt; mso-ansi-language:NL-BE;}.MsoChpDefault {mso-style-type:export-only; mso-default-props:yes; font-size:10.0pt; mso-ansi-font-size:10.0pt; mso-bidi-font-size:10.0pt; font-family:Calibri; mso-ascii-font-family:Calibri; mso-ascii-theme-font:minor-latin; mso-fareast-font-family:宋体; mso-fareast-theme-font:minor-fareast; mso-hansi-font-family:Calibri; mso-hansi-theme-font:minor-latin; mso-bidi-font-family:"Times New Roman"; mso-bidi-theme-font:minor-bidi; mso-ansi-language:NL;}.MsoPapDefault {mso-style-type:export-only; margin-bottom:10.0pt; text-align:justify; text-justify:inter-ideograph; line-height:115%;}@page WordSection1 {size:612.0pt 792.0pt; margin:72.0pt 90.0pt 72.0pt 90.0pt; mso-header-margin:36.0pt; mso-footer-margin:36.0pt; mso-paper-source:0;}div.WordSection1 {page:WordSection1;}-->General overviewThis thesis describes a series of studiesinto the relationship between low back pain (LBP), proprioceptive posturalcontrol and physical activity. Two central elements in this thesis are the UP-LIFTcohort (Utrecht Police Lifestyle Intervention Fitness and Training study) andthe experimental set-up. The UP-LIFT cohort consist of employees of the Utrechtpolice department in the Netherlands (n=1723), agedbetween 18 and 62 years,who visited the health research department at the University of Applied SciencesUtrecht for a fitness and lifestyle evaluation. In this evaluation a broadrange of parameters were tested, concerning cardio vascular risk factors,motivational and social-cognitive determinants regarding a healthy lifestyle,physical fitness (peak Arial">VO2, muscle strength andendurance, flexibility) and physical activity (kind, intensity and duration).Random samples out of this cohort wereenrolled in the experimental set up used in the studies in this thesis. Toquantify proprioceptive postural control (the way proprioception is used inpostural control) we used a series of 7 trials of 60 seconds. In all trialsparticipants stood barefoot on a force plate. In all but the first and the lasttrial, vision was occluded by means of taped safety glasses. In trials 1 to 3subjects stood relaxed without perturbations, while in trial 3 a foam pad wasplace over the force plate. The Centre of Pressure (CoP), which is the locationof the ground reaction force acting on the body, was measured 200 times persecond. Human beings never stand completely still.With the use of the CoPthese movements, known as postural sway, canbe measured. In trials 4 to 8,muscle vibration was applied between the 15th and the 30th second, in allcombinations of m. Triceps Surea(TSM) and Lumbar Paravertebral Muscles (LPM)vibration and standing on the force plate (rigid surface) or on a foam padplaced over the force plate (unstable surface). Muscle vibration induces anafferent signal from the spindles within the muscle that corresponds to alengthening of the muscle. Signals from all muscles spindles surrounding ajoint are considered to be the main source of information about the position ofthe joint. Therefore changes in CoP location during muscle vibration provide anindication of the impact the proprioceptive signals from the vibrated musclespindles exhibit on the integration of all sources of information (vestibular,visual, proprioceptive) in maintaining postural control. Part I. Introduction and methodologyChapter 1 provides a broad introductioninto LBP, proprioception and proprioceptive postural control, and the influenceof physical activity on those entities. Chapter 2 & 3 describe two studiesinto the characteristics and measurement properties of the experimental set-upused in this thesis. In Chapter 2 we studied the influence ofstanding on an unstable surface, i.e. foam, on muscle vibration effects. Thefirst 100 subjects from the UP-LIFT cohort performing the proprioceptivepostural control tests were enrolled. Wecompared the effect TSM and LPMvibration had onCoP position and velocity when standing on a rigid surfacewith that when standing on a foam pad. The results showed that on foam theeffect of TSM vibration was significantly smaller than on a solid surface, bothfor CoP velocity as for change in CoP position under vibration, while for LPMvibration the effects were reversed to that of TSM vibration. We hypothesizedthat this is caused by a decreased weighting of proprioceptive signalsoriginating from the ankle musculature, and an up weighting of proprioceptivesignals originating from lumbar spinal musculature when standing on foam.Therefore weconcluded that exercises on unstable surfaces do not appear totarget peripheral ankle proprioception, but may challenge the capacity of thecentral nervous system to shift the weighting of sources of proprioceptivesignals between body regions.In Chapter 3, the results of a reliabilitystudy of a broad range of possible outcome variables of muscle vibration effectsare presented. This study was conducted among 20 students, staff and familymembers of the students. Change in mean position of the CoP(’displacement’) during vibration showed good reliability (ICC’s > 0.6).Ratios of displacement between LPM and TSM vibration (‘proprioceptiveweighting’) showed fair to good reliability (0.52–0.73). Change in CoP velocityunder influence of vibration appeared not to bereliable. Balance recovery,when calculated based CoP position a short period after cessation of vibration,showed good reliability (ICC’s >0.6). Agreement measures were poor, withmost CV’s ranging between 18% and 36%. Expressing variables relative to thelimits of stability did not improve reliability or agreement. According to thisstudy, displacement during vibration, proprioceptive weighting and selectedrecovery variables are the most reliable indicators of the responseto musclevibration. In the present form these variables do not seemsuitable for use inclinical practice.Part II. Low back painandproprioceptive postural controlIn Part II, the findings in the UPLIFTstudy among 215 policemen and women regarding proprioceptive postural controlvariables and their association with LBP are presented (Chapter 4). In thiscross-sectional study, proprioceptive postural control and its relationship toLBP were investigated by means of muscle vibration effects and by posturalsway. Postural sway wasexpressed as the structure, range and velocity of theCoP. To reducethe large amount of possible variables to quantify musclevibration responses and postural sway under different conditions, we firstperformed three factor analyses, in which we only entered variables that showedat least a fair reliability in Chapter 3. A biologically plausible constructcould be appointed to every one of the in total 10 factors. For postural swaythese were frequency and irregularity on rigid surface (1), velocity and rangeon rigid (2), frequency and irregularity on foam (3), and velocity and range onfoam (4). For muscle vibration effects these were response to TSM vibration onrigid and foam (1), response to LPM vibration on foam (2), and response to LPMvibration on a rigid surface (3). For recovery after vibration cessation thesewere recovery standing on a solid surface (1), recovery standing on afoamsurface (2), and peak recovery, the maximum anterior CoP position aftercessation of vibration (3).The analysis of possible associationsdemonstrated that subjects with LBP sway with comparableamplitudes as subjectswithout LBP, but the structure of their sway pattern was less regular withhigher frequency content. Subjects withLBP also showed a smaller response toTSM vibration, a non significant smaller response to LPM vibration, and aslower balance recovery after cessation of vibration when standing on a solidsurface. There was a weak but significant association between smaller responsesto TSM vibration and an irregular, high frequency sway pattern, independentfrom LBP. Based on these findings, we proposed a modelfor control of postural sway. This model suggests that subjects with LBP usemore co-contraction and less cognitive control, to maintain a standing balancewhen compared to subjects without LBP. Cognitive attention may, in subjectswith LBP under challenging conditions, be directed towards the lower back. Inaddition, a generally reduced weighting of proprioceptive signals in subjectswith LBP is suggested as an explanation for the findings in muscle vibration. Part III.The influence ofphysical activityPart III describes two studies on theinfluence of physical activity on proprioceptive control.The first study is asystematic review into the relationship betweenphysical activity and posturalsway (Chapter 5). The search of the literature retrieved 39 studies, 37 with acomparative design, one designed as a cohort study, and one as a randomizedcontrolled trial. The main conclusion in this study was that in general sportpractitioners sway less than controls, and high-level athletes sway less than low-levelathletes. Additionally, we identified specific effects dependenton the use ofvision, sport specific postures, and frequency and duration of the (sports)activity. From Chapter 5 we conclude that posturalsway in unperturbed bipedal stance does not seem suitable to detect subtledifferences between groups of healthy people. The effects of sports activity onpostural sway are specific to the characteristics of the sport, and become moremanifest under challenging conditions. This is in line with our own studies, inwhich we alsofound the clearest differences between groups when standing onfoam or during muscle vibration. This could have clinical implications, asclinical examinations often take place in standardized, non-challengingconditions. Perhaps it would be better to evaluate movement behaviour ofsubjects with LBP under more stressful conditions. In Chapter 6 we were interested in theinfluence of physical activityand aerobic fitness on the relationships foundin Part III. For thisstudy we used data of the same 215 subjects enrolled inthe UP-LIFT study. We asked subjects for kind, level and duration of physicalactivity by means of two questionnaires, the SQUASH and a custom madequestionnaire. We used peak oxygen uptake (VO2 per kilogram), resulting from the ergometer test in the UP-LIFTstudy as measure for aerobic fitness. There were two major findings: a trendtowards a higher response to muscle vibration in subjects with a higher levelof aerobic fitness (p= 0.06), and a quadratic association between aerobicfitness and sway irregularity (low and high level of fitness associated with amore regular sway pattern with lower frequency content). With respect tophysical activity a trend towards the same associations as for physical fitnesscould be seen, but these associations never reached statistical significance.The reason for this could well be the methodological shortcomings of self-reportedphysical activity measures. A possible explanation for the increasedsensitivity to musclevibration in subjects with high levels of peak VO2, could be an improved muscle oxygenation. The subjects with highlevels of peak VO2 also showed a more regularsway pattern. As it is likely that these subjects are more competitive, wesuggest this could be caused bya supraspinal control strategy, which impliesan exploratory behaviour. At the other end of the peak VO2 spectrum, the more regular sway pattern with lower frequencycontent found in subjectswith low levels of peak VO2 could be due to more attention to the postural task than othersubjects. It has been shown that adequate responses to changes in theenvironment during a postural task depend on the amount of physical activitysubjetcs exhibit in daily life. Thechanges in control strategies in the previousstudies seen in subjects with LBP were in our study independent from changes inpeak VO2 and physical activity level. Part IV. Discussion and summariesIn Part IV all results of theaforementioned studies are discussed as a whole. We noticed some issues thatcould not be fully explained by contemporary views on proprioceptive weighting.The first issue is that the initialvelocity of the CoP towards thefinal CoP position during vibration was similarfor both the condition on a rigid surface as on foam. A down weighting ofproprioceptive signals from the ankle muscles on foam, as suggested, would leadto asmaller resulting moment, thus to a smaller CoP velocity in the initialvibration epoch towards the final CoP position. This is even more emphasized bythe mechanical properties of foam. On foam a comparable neural drive to alphamotor neurons would lead toa lower moment aroundthe ankle due to thedeformation of the foam pad. The second issue is that in some studieswhere fingertip contact without mechanical support was allowed duringvibration, fingertip contact decreased the effect of muscle vibration whenstanding on a rigid surface, but increased the muscle vibration effect whenstanding on an unstable surface. The latter cannot be explained by a downweighting of proprioceptive signals based on their reliability. Fingertip touchwould add extra information to the proprioceptive system, which will make evenclearer that signals originating from the ankles are not reliable when standingon an unstable surface.The third issue is the common opinion thatthe CNS is able to gate sensory input in accordance to their reliability.However, the brain never has direct access tothe true parameter values butonly has access to the data from whichit makes inferences. The explanationthat signals from the ankle region are less informative when standing on foamis teleological in nature, but does not supply an explanation of how this weightingis achieved. If the CNS is not a contemporary form of the homunculus, who isthe one who decides? A possible solution to these problemsisbased upon the model for postural control presented in Part II (Chapter 3).This model includes 3 control strategies, co-contraction, spinal proprioceptivefeedback and supraspinal control. We propose a further refinement of this modelin which we integrate the original model with aspects of the Equilibrium Point(EP) theory and of Bayesian probability statistics. In short, the CNS can set joint stiffnessby co-contraction and sensitivity of the proprioceptive reflex. Decreasing thethresholds of all muscles, leaving the vector summation of all spindlethresholds surrounding the joint unchanged, setsthe sensitivity of theproprioceptive reflex.In an independent process, the CNS sets theplanned joint angle by determining the ratio of the muscle spindle activationthresholds of all muscles surrounding a joint. Which joint angle is mostappropriate is the resultant of a central weighing process with other sourcesof sensory information in relation to the desired task. The relative weight ofevery sensory source depends on the signal noise ratio. When discussing Part III, we looked deeperinto the differences between subjectswith and subjects without LBP when undervibration. The initial response to vibration is equal in both groups, but theinitial response to cessation of LPM vibration is much larger in subjects withLBP whenstanding on foam, as is the initial response to cessation of TSMvibration when standing on a rigid surface. We explain this according to therefined model as a higher peripheral reflex sensitivity in subjectswith LBP,but a smaller weighting in the centrally mediated control of position. Thesmaller weighting of proprioceptive signals in subjects with LBP is explainedby a lower signal noise ratio from peripheral afference due to a comprisedproprioceptive signal or due to higher peripheral stiffness. With regard toChapter 6, we suggest that a smaller microcirculation in subjects with lowerlevels of aerobic fitness (peak Arial">VO2) causes more variancein the afferent peripheral signal, with consequently a down weighting of thesignal in the centrally mediated position control. Thelack of differencesbetween groups in response to cessation of vibration could be due to higherlevels ofco-contraction and reflex sensitivity in both high and low levels ofaerobic fitness. This idea is strengthened by the findings in postural swaystructure, which is for both ends of the spectrum more regular and with a lowerfrequency content than for subjects with average levels of peak VO2.This refined model has been developedretrospectively. Obviously it has to be tested prospectively. Recommendationsfor further research are made based upon the predictions of this model aboutstretch reflexes and proprioceptive weighting based on variation of theperipheral signal.<br
>
Publication status: published
KU Leuven publication type: TH
Appears in Collections:Research Group for Cardiovascular and Respiratory Rehabilitation
Research Group for Musculoskeletal Rehabilitation

Files in This Item:

There are no files associated with this item.

Request a copy

 




All items in Lirias are protected by copyright, with all rights reserved.