Proprioception Acuity in the Lower Extremity Between Young and Older Adults. Adriana Menting de Alba European School of Physiotherapy, Amsterdam University of Applied Sciences, Tafelbergweg 51, Amsterdam, The Netherlands Submitted: December 7, 2017

Abstract Introduction: Approximately 646,000 individuals die from falls each year and roughly 37.3 million falls including those that are none fatal, require medical attention each year (WHO 2017). With the growing elderly population and the raise in fall injuries, the medical costs are expected to increase as well reaching US$ 240 billion by the year 2040 (Centers 2017). The aim of this review is to look at recent evidence investigating the extent of proprioception decline with aging in different joints of the lower extremity, as it has been linked to balance deficits and risk for falling. Methods: The literature search resulted in 10 potential articles. Results: In total there were 7 studies in total included in this review with a sample of 316 subjects and an age group distribution of 147 younger and 171 older adults. The main outcome measure was JPS in angle reproduction by ipsilateral or contralateral matching tasks used in the hip, knee, and ankle. 4 out of the 7 studies with high methodological quality showed significant results (P<.05) in the difference of angular error scores between the older and the younger groups. Discussion: In all the 4 articles which found significant difference the absolute error score of the older groups increased to almost double compared to those of the younger groups. No proprioception decline threshold scores were set by any of the studies to indicate risk for falling, and should be further investigated. A relationship between proprioception decline and history of falls and dynamic balance deficit was also found. Conclusion: The quantitative scores provided can help physiotherapists and other clinicians to identify patients with proprioception deficit and motivate them to include programs for dynamic balance training for midlife and older age groups to attenuate its decrease and reduce the risk of falls. Keywords: ‘age related changes’ ‘proprioception’ ‘joint position sense’ ‘assessment’ ‘balance’ ‘sensorimotor changes’ ‘aging’ ‘lower extremity’

Introduction Falls are the second leading cause of accidental death worldwide, and the leading cause of death by injury in adults over the age of 65 (WHO 2017). Approximately 646,000 individuals die from falls each year and roughly 37.3 million falls including those that are none fatal, require medical attention each year (WHO 2017). The risk for falling increases with age with individuals aged over 60 accounting for 50% of their overall injury- related hospitalizations. Some major underlying causes for fall-related hospital admission are hip fracture, traumatic brain injuries and upper limb injuries (Centers 2017). For example, in the United States, 20–30% of older people who fall suffer moderate to severe injuries such as bruises, hip fractures, or head trauma (Centers 2017). There is a huge financial burden that comes from fall related injuries. They are amongst the 20 most expensive medical conditions with the overall

medical cost being $31 billion annually (Burns et al. 2016). This cost is only expected to grow in the future in response to the growing elderly population. Individuals aged over 60 are fastest growing age group, which is expected to reach 2 billion by 2050 (WHO 2017). Therefore, a pronounced number of elderly individuals will not only trigger substantial increase of falls and fall injury but also an alarming increase in its medical costs, which is expected to reach US$ 240 billion by the year 2040 (Centers 2017). Viewing these statistics makes it evident that aging is major risk for falls but why? A study by Hawk et al 2006 investigated balance and risk for falls in a sample of community-dwelling adults aged 65 and older and found that about one third (32%) of participants had a Burg Balance Scale (BBS) score of below 46, the cut off point for predicting high risk of falling (Hawks 2006). This and many other like the study by Donaghy 2016 have linked balance decline with high risk for falling among the elderly population. This is in part due to the natural decline of the

sensorimotor components which work together to maintaining functional joint stability. The sensorimotor system involves the following senses vision, vestibular, and most importantly proprioception to maintain balance and coordination. But what is proprioception? In 1906 it was first defined by neurophysiologist Sir Charles Sherrington as “the perception of joint and body movement as well as position of the body, or body segments, in space” (Sherington 1906). In other words, it is the awareness of body position, movement and sensation of force. Proprioception incorporates the afferent input arising from internal peripheral areas of the body to the central nervous system for processing that contribute to postural control, joint stability and several other conscious sensations. These sensations are detected by different nerve endings called mechanoreceptors, each specialize in detecting distinct stimulus such as force, vibration, stretch and so on. They can be found in and around joints such as in muscle, tendon, skin, fascia, joint capsules, and ligaments (Anat 1988). The stimuli perceived by these receptors reaches the central nervous system, where it is processed and organized, via afferent nerve tracks. Motor control signals originating from the central nervous system are sent to specific muscles around a joint via efferent nerve tracks to monitor posture, fine tune movement and therefore help maintain balance (Riemann 2002). A review written by Goble et al 2009 examined proprioceptive acuity in joints of the upper and lower extremity, finding a deterioration with aging (Goble 2009). The association between the physiological changes that occur in the natural process of aging and proprioception decline has been investigated. Among the most prominent changes seen are the fallowing; a decrease in total number of muscle fibers per muscle spindle, an increase in the spindle capsule thickness, expanded motor end plates, larger and slower motor units, myelin abnormalities, and a decrease in the number of neurons and receptors in the brain (Ribeiro et al. 2007). These central and peripheral deterioration changes hinder the continuous monitoring of coordination and balance (Franco et al. 2015). The natural physiological changes just discussed can contribute to the decline of proprioception acuity and therefore influence the risk of falling. A loss in joint proprioception sensibility may also lead to abnormal joint biomechanics during functional activities which over a period of time, can cause degenerative joint disease (Skinner 1993). So how can proprioception be objectify assessed to determine differences? Measuring proprioception is very difficult clinically since it is a phenomenon both occurring consciously and unconsciously within the body (Benjaminse 2004). Researchers and clinicians measure the four submodalities of proprioception known as: JPS - the ability to reproduce the same joint position, TTDPM -

ability to detect the initiation of passive joint movement, velocity sense (VS) and FS - ability to reproduce the same force (Benjaminse 2004). Joint position sense is the modality most frequently assessed due to its accessibility and accuracy (Globe 2010). It is assessed in two ways; by ipsilateral matching or contralateral matching tasks. When using the ipsilateral matching test the joint assessed is placed in a given position and held for several seconds before the limb is returned to the starting position and the subject is asked to match the previous position. In the contralateral matching test the joint examined is passively moved to a specific angle by an experimenter and then the participant matches the same angle with the contralateral limb. Some of the instruments used to objectively measure JPS include goniometers, isokinetic dynamometers, electromagnetic tracking devices, or custom-made jigs (You 2005, Riemann 2002, Warner 1996). With continued medical advancements which expand life expectancy, the elderly population is only expected to keep growing. This will result in an increase in falls and fall related injuries in the future, not to mention the huge economic burden that will come from treating them. More focus therefore should be put on ways in which they can be prevented. This brings health professional especially physical therapists into the picture, as they can help to assess for and treat balance deficits in the elderly which is directly related to risk of falls. As proprioception is one of the most important mechanisms in maintaining balance and coordination more investigation should be put on ways to assess it and treat it. This review aims to investigate just how much decline in proprioception there is with aging by answering the following research question: Is there an age-related deterioration in joint position sense of the lower extremity between individuals aged 15 to 40 and 65 and older? It would be interesting to investigate if proprioception decline causes more of an effect on static postural stability or dynamic balance. Evaluating the different assessment set ups currently used can also be looked at and whether they involve static or dynamic evaluation of proprioception. Other factors that could be examined, are the joints and planes of movements which have more of an influence in maintaining postural balance and the compensatory responses used by the elderly to maintain balance. Professionals, can come to quantitatively understand the decline of proprioception acuity and evaluate its impact on balance and risk of falling if threshold scores are provided by each lower extremity joint. My partner’s review aims to evaluate on the most effective exercises to help prevent or improve proprioception in the elderly. Together we hope to not only provide information on proprioception deterioration with aging and the impact it has in risk of falls but also provide information on whether it can be

improved or attenuated with specific exercises. In this review review will look at threshold measurements to see at what point proprioception deficit is considered to be a factor for risk of falling. New interest into the magnitude of proprioception deterioration in older adults will hopefully foster interest in advancements for treatment/prevention on proprioception deterioration.

Methods Search engines and databases The following four databases were used for article searching: PubMed, PEDro Cochrane and Google Scholar. The PEDro database was chosen for its recognition in providing randomized trials, systematic reviews and clinical practice guidelines focused on physiotherapy practice. Since this review aims to inform physical therapists primarily, it was perfect to include the PEDro database in this search. Cochrane was specifically chosen to find systematic reviews focused on human health care and health policy as it is its specialty. Though PEDro and Cochrane are both excellent to include for the topic of this research, the database most relied upon was PubMed. PubMed provides a huge selection of academic journals covering the fields of medicine, nursing, dentistry, and veterinary medicine, therefore most of the scientific articles found in other databases are already included in PubMed. Finally, Google Scholar was used to find those extra articles not provided in PubMed or the other databases since it allows for a full text search that would yield different results than PubMed which reviews only the abstracts of articles. To help retrieve the full text article of any of the databases the HvA library website was used.

Keywords The keywords used for this research were: ‘age related changes’ ‘proprioception’ ‘joint position sense’ ‘assessment’ ‘balance’ ‘sensorimotor changes’ ‘sensorimotor function in the elderly’ ‘aging’ ‘proprioception decline’ ‘lower extremity’ Search strategy The full PubMed search string can be viewed in Appendix 1. In the databases, the ‘Advanced’ search option was used to find MeSH terms and combine keywords with Boolean operators. The keywords; ‘aging’, ‘aged’, ‘age factors’, ‘proprioception’, ‘joint proprioception’ ‘joint position sense’ and ‘lower extremity’ were added as MeSH terms while ‘aging’, ‘elderly’, ‘age factor’, ‘aging effect’, ‘influence of aging’, ‘lifespan’, ‘elderly’, ‘older adults’ ‘joint position’ ‘lower limb’ and ‘propriocepti*’ were added as text words. The Boolean operator ‘AND’ was used to find articles that combine the synonyms and MeSH terms ‘aged’

and ‘aging factors’ with ‘proprioception’ ‘joint position sense’ and ‘lower extremity’ together. The Boolean operator ‘OR’ was used to add synonyms for the MeSH terms ‘aging’, and ‘proprioception’ ‘joint position sense’ and ‘lower extremity’ to the search. The keywords ‘age factors’, ‘aging effect’ and ‘propriocepti’ were combined with the search operator ‘*’, to include all the terms developed around their root word’. This search strategy gave a total of 191 articles in PubMed, 0 in PEDro, 7 in Cochrane and 5 articles in Google Scholar. Figure 1. shows the PRISMA flowchart mapping out the number of articles identified, included and excluded and the reason for their exclusion. Most of the articles found in Cochrane and Google Scholar were already included in PubMed therefore after the duplicates were removed 193 were left. No articles relating to the research question were found in PEDro. Screening The 193 articles found through database searching and additional searching were screened for inclusion and exclusion criteria on their title and abstract yielding a total of 13 and 180 of them being rejected. The following inclusion criteria was employed: healthy human subjects, “young” population age range: 15-40y (control group) and “old” population age range: 65y and older (case group), community dwelling elderly, a quantified measure of joint position sense in the lower extremity, and articles written in English. The reason for including the specific age of 15 to represent one end range of the “young group” (control) was because studies provide evidence that developmental changes in proprioceptive ability are relatively stable by approximately 8-12yrs (Goble 2005). I felt that including participants with an age of 12 would be too young and they might not collaborate adequately in the examinations, therefore I included an older age of 15 to also leave room to include those studies investigating in adolescent participants. The exclusion criteria included: articles older than 2002, studies which involved participation of fewer than 8 test subjects, articles written in languages other than English, and participants younger than 15yrs.

confounding and it is measured by one of the following codes; ++ All or most of the checklist criteria have been fulfilled, + Some of the criteria have been fulfilled, those criteria that have not been fulfilled or not adequately described are thought unlikely to alter the conclusions – Few or no criteria have been fulfilled. The conclusions of the study are thought likely or very likely to alter.

Figure 1. PRISMA 2009 Diagram for Study Inclusion of Articles Investigating Lower Extremity Proprioception Acuity between Younger and Older Adults.

Quality check Another 3 articles were unqualified due to unavailable retrieval of their full text leaving a total of 10 articles for methodological quality assessment. The NICE 2009 checklist can be used for methodological grading of systematic reviews, RCT’s, case control studies and cohort studies. In this review, it was used to grade case-control studies. The checklist is broken up into two sections evaluating the overall internal and external validity of a study. The questions in section 1 are aimed at establishing the internal validity and regard the following 4 subjects; selection of subjects, assessment set up, confounding factors, and statistical analysis (NICE 2009). Each question covers an aspect of methodology that has been shown to make a significant difference to the conclusions of a study and can receive one of the following marcs; ‘well covered’, ‘adequately addressed’, ‘poorly addressed’, ‘not addressed’, ‘not reported’, ‘not applicable’ (NICE 2009). Section 2 looks at the overall external validity of the study by evaluating the following; impact of risk of bias or confounding, whether the overall effect was due to the exposure being investigated, and whether the results of the study are directly applicable to the patient group targeted. The overall quality assessment of the article is based on how well the study minimized the risk of bias or

As addressed by the NICE guidelines, the complex design of case–control studies often require strict grading since so many poor- quality studies are conducted. There are few criteria that should, alone be unsupported, leading to rejection of a study. Both my partner and I graded each article, recognizing it as having good methodological quality if in Section 2 it received an overall marc of ++ and all the questions in Section 1 had at least an ‘adequately addressed’ response. Articles that failed to address one or more of the questions were automatically rejected as recommended by the NICE guidelines. Table 1. gives a detailed overview of the grading received per question by each article using the NICE 2009 checklist. Only articles with good methodological quality were used in this review, resulting in 7 eligible and 3 rejected studies. Those rejected were by Pickard et al. (2003), Verschueren et al (2002), and Franco et al. (2015) which failed to or gave a poor report on one of the following questions; not comparable case and control groups, a lack of information to distinguish the groups similarities and differences or there was uncertainty of whether the overall effect of the assessment was due to the exposure being investigated. If there were discrepancies between the scores my partner and I gave a certain article, we came together to discuss about it and if necessary took the lower score out of the two for a given question.

Results The 7 studies all assessed joint position sense acuity in the lower extremity and involved a total of 316 subjects with an age group distribution of 147 younger and 171 older adults. There were two articles by Wingert et al. (2014) and by Low Choy et al. (2007) that included only women in their study while the rest included a relatively equal distribution of female and male subjects. Table 2. gives a deeper overview on

the studies’ joint segment involved, assessment set up, instrumentation used, and results which will be discussed in this section. The subjects in a given study were comparable at baseline between age groups in the following factors; activity level, no neurological or vascular pathologies, no cognitive deficit, no

musculoskeletal injuries and number of subjects between groups were relatively the same. Participants with other diseases that might interfere with outcome measures apart from age were also excluded. All subjects were blindfolded to the matching tasks and used either the dominant or non-dominant limb in both groups. Though this review looks at the specific outcome measure of JPS angular error between the young and older age groups, it is also interesting to mention the diverse range of other outcome measures included in the studies. For example; Tegner activity levels, postural sway, clinical balance and fear of falling assessed with the mini-Balance Evaluation System Test and Activities-specific Balance Confidence Scale. The study of Low et al. 2007 also measured strength in the quadriceps, hip abductors and adductors and tactile acuity and vibration threshold in the lower extremity was also tested. The study by Craig et al. 2016 examined muscle activity and proprioception acuity by recording surface electromyography from the bilateral tibialis anterior and gastrocnemius medialis muscles in postural assessment on sway references and the study of Boisgontier et al. 2012, measured proprioception control while performing a secondary cognitive task to see if there was a decreased in matching performance in older adults. There was a total of 3 studies evaluating JPS at the ankle by; Boisgontier et al. (2012), You (2005), and Craig et al. (2016), 3 evaluating the knee by; Ribeiro and Oliveira (2010), Relph and Harrington (2016), and Low Choy et al. (2007) and 1 evaluating the hip by; Wingert et al. (2014). All subjects performed the matching tasks actively and blindfolded regardless of whether the matching was performed in the ipsilateral remembered side like in the studies of Ribeiro and Oliveira (2010), Wingert et al. (2014), You (2005), Relph and Harrington (2016), and Low Choy et al. (2007) or on the contralateral remembered side like in the studies of Craig et al. (2016) and Boisgontier et al. (2012). In most studies, the matching tasks were done in a non-weight bearing open chain position while the only two that involved full weight baring in a standing position were by Low et al. (2007), who evaluated knee flexion and You et al. (2005) who evaluated ankle plantarflexion. These comparisons can be seen clearly by looking at Table 3 below. Table 3. Comparison of JPS assessment between studies.

It is interesting to note that the two studies that chose to assess joint position sense in a contralateral matching task both assessed the ankle and the only two studies that included weight baring chose not to assess it by the contralateral matching tasks. The instruments used to measure angular displacement varied in complexity from study to study. Some involved relatively simple custom-made gadgets to evaluate JPS such as the semi-goniometer used in the study by Wingert et al. (2014). The large goniometer was placed the transverse plane to assess internal and external rotation of the hip. The replicated angles were manually recorded by the examiner on the goniometer and error scores were calculated by taking the difference between the target and matched angles. More elaborate instrumentation was used when assessing the ankle, which used foot plates attached to linear potentiometers. These potentiometers were connected to different computer software systems, namely the LabView computer software. Once the participants felt they had reached the target position they pressed a button to record the matched angle in the LabView system which converted footplate voltage outputs to joint angles. When evaluating the knee, complexity of instrumentation used varied. To determine angular error in the studies evaluating JPS of the knee, a camera was used to either take pictures or videotape the target and matched positions. In these studies, the following anatomical points were marked with markers; greater trochanter, lateral epicondyle, and lateral malleolus were common landmarks on the legs to allow the camera could easily track the joint positions. The camera was connected to a computer software such as a two-dimensional manual distinguishing software or the Ariel Performance Analysis System to calculated angular error scores. Table 4. demonstrates one or two proprioception assessment method(s) used per joint by the studies reviewed. The specific methods seen were chosen based on good reliability and validity though the ones for the hip were not found. The hip was given two assessment methods to show a variety of different ranges of motion that can be examined in the hip, therefore the study by Wingert et al (2014) which looks at IR and ER was included and another study by Pickard et al. (2003) (not analyzed in this review) which looks at abduction and adduction, was also chosen. This information can proof useful for physical therapists, as they can become aware of different current methods to assess the ankle, knee and hip. The reliability and validity scores were not found for all the instrumentations shown, but they were for the studies of Relph and Harrington (2016) and You (2005) as demonstrated in Table 4.

Table 4. Description of recommended instrumentation per joint analyzed.

Boisgontier et al. (2012) gave10 trials. The other studies had a number in between such as the one from You (2005) and Low Choy et al. (2007) which gave 2 trials, or by Ribeiro and Oliveira (2010) which gave 3 trials, and finally those by Relph and Harrington (2016) and Craig et al. (2016) which gave 5 trials. The studies of Ribeiro and Oliveira (2010), and Craig et al. (2016) were the only two studies which tested the nondominant leg while the rest tested the dominant leg. Statistical significance for all the studies was set at (P<.05) with a total of 4 showing a significant difference in joint position sense error between the older and younger groups. Those which saw significant differences were by Wingert et al. (2014) who examined IR and ER of the hip (P<0.001), by Ribeiro and Oliveira (2010) who examined knee flexion in open chain, with error scores of 4.74˚±2.67˚ for the younger group and 9.35˚±4.34˚ for the older group (P<0.001), the one by You et al. (2005) which examined knee flexion in weight bearing had error scores of 1.37˚±.56˚ for the younger group and 2.32˚±.70˚ for the older group (P<.05), and finally the one by Low et al. (2007) analyzing ankle neutral, IV, EV, PF, DF positions had an error score of 3.7˚ for the younger group and 6.6˚ for the older group (P<.01).

As see in Table 4. most of the instrumentation used is quite elaborate especially for the ankle and the hip which assessed abd/add. These studies needing motion analysis systems or potentiometers a well as the LabView computer system to help determine angular error scores. For descriptive statistics SPSS 10.0 was used in most studies to determine population variance. The repeated measure ANOVA was also frequently used to view significant differences in JPS error scores between the age groups passive and active tasks. In some studies, the relationship of physical activity level, walking velocity, static and dynamic balance on JPS was measured using the Pearson’s correlation coefficient and the criterion of significant difference was set at level p<.05. Several assessment factors fluctuated between the studies such as; target angle amplitude, number of trials performed per range of motion, whether dominant or non-dominant leg was tested, how many seconds to hold the actively matched position, randomization or standardization of target angles chosen, range of motion analyzed, and practice trials given. The number of trials given to perform between the studies ranged from 1 to 10. The one by Wingert et al. (2014) gave just one trial while

The studies showing no statistical significance were by Boisgontier et al. (2012) which measured ankle DF and had error scores of 3.0˚±.2˚ for the younger group and 3.8˚±.5˚ for the older group (P=.954), by Craig et al. (2016) which also measured DF of the ankle that showed a difference between groups of P=.65, and by Relph and Harrington (2016) testing knee flexion and extension in open chain which showed FLEX: 3.6˚, EXT: 2.7˚ error scores for the younger group and FLEX: 3.7˚, EXT: 3.7˚ for the older group (FLEX: p=.603, EXT: p=.536).

Discussion The 25-year gap difference between groups gave an almost double absolute error increase when comparing the young to old groups. These findings support those of the review written by Goble et al. 2009 which assessed proprioception in the upper and lower extremity, demonstrating a proprioception decline with aging. There was no meta-analysis of Goble’s findings and therefore was not included in this review. It was interesting to see that in the three studies assessing DF of the ankle, the only one that had significant results was the one evaluating JPS in standing by Low Choy et al. (2007). This finding seems to support those by Bullock-saxton et al. 2001. which found that position matching ability improves with weight bearing.

Limitations There are several limitations in the set up of the studies and design of the studies that could be seen as confounding factors to the results. One limitation regards the use of the contralateral matching tasks when assessing JPS as evaluated in the review by Goble (2010). Goble suggested that since the reference limb provides input on the desired joint angle, it is difficult to distinguish if the errors come from the reference limb, the matching limb, or both (Goble, 2010). In addition, the transfer of information between limbs requires the information to pass through the cerebral hemispheres which may cause interhemispheric communication issues between errors caused by poor proprioception and errors caused by central processing (Goble, 2010). The fact that 5 studies using IR matching had significance findings while the 2 studies that used CC matching both found no significance may be due to the facts just mentioned. No standardization of trials for JPS in the ankle, knee or hip have been set which could have been another limitation. The two studies assessing DF of the ankle by Boisgontier et al. (2012) and Craig et al. (2016) resulting in no significant results, gave their subjects more practice trials than the other studies. The one by Boisgontier et al. (2012) gave subjects 10 practice trials while the one by Craig et al. (2016) gave practice trials until the error match was 2˚ away from target. These extra trials might have caused a learning curve between subjects and therefore no significance between the group scores was found. The study by You (2005) assessed the DF of the ankle both in an elderly group of mean age 73.12y without history of falls and in an elderly group with the same mean age with a history of falls and found that the group with history of falls had a significantly higher error score than the other elderly group. Not all the studies in this review assessed if the individuals in the elderly groups had a history of falls which could have caused big influence in the results. To appropriately measure proprioception, all its submodalities need to be assessed; velocity, position sense, force and threshold to detect motion. All these components work together in different degrees to maintain static stability and dynamic balance therefore the results of this review only tell part of the story. This can be seen as a limiting factor since only one submodality was assessed and to perform a thorough investigation into the proprioception deficits in an elderly subject, all submodalities must be assessed. Finally, it is also important to note that none of the studies gave threshold marcs to determine high risk of falling and none examined dynamic position sense. Dynamic position sense is the ability to monitor position during motion and plays a role in functional stability (Medhavan and Shields 2005). Medhavan and Shields 2005 assessed dynamic position sense at the ankle. The results showed that acuity of dynamic position

sense was significantly influenced by the speed of joint movement. Further research The study by Wingert et al. (2014) investigated the correlation between increased proprioception error and increase in postural sway in older adults, finding no relationship. Perhaps it was due to the fact that only IR and ER of the hip was evaluated which might provide less information than other planes for postural control in stance. It is my believe that distal joints could provide more useful proprioception information for stance and postural sway. In contrast, there was a relationship between hip proprioception and dynamic balance in this study. Older adults with higher hip proprioception acuity scored significantly higher on the mini-BESTest which suggests that hip proprioception is important in maintaining balance during motor activities such as single-leg stance, gait and multitasking. Results also revealed an increase in sway displacement and velocity in older adults with increase proprioception errors. Further research should focus on examining the impact in proprioception deficiency certain joint of the lower extremity have in postural stability and dynamic balance. It was also noted that older adults rely on vision more for maintenance of balance which could be a compensatory strategy for proprioception deficit. It would be interesting to investigate how much the elderly rely on vision to compensate for loss of proprioception. The study by Craig et al. (2016) examined the correlation between muscle activity, tested with an EMG on the tibialis anterior and gastrocnemius, and proprioceptive acuity. It was found that better proprioception acuity is associated with more cocontraction during unstable posture. This may be useful information for physiotherapists to develop training programs involving lower limb strength to improve balance and coordination. The findings of Craig et al. 2016 also saw that the elderly employed a muscle co-contraction strategy to prevent a further increase in postural sway by increasing the stiffness of the ankle joint. I believe this stiffening or muscle contraction could be a compensatory strategy response for degraded proprioception input and would also be an interesting subject to further analyze. Research could compare young and older adults and the degree of muscle activation when put in an unstable situation, therefore forced to maintain balance. Then one could see if the elderly do activate more muscles to compensate for loss of proprioception. Regarding the fact that not all studies evaluated history of falls in their subjects, research in the future could break up the elderly group into two groups, one with history of falls and the other with no history of falls and investigate JPS differences in diverse joints. This also relates to the study by Wingert et al. (2014), proprioception acuity was also assessed

in mid aged individuals from 40-64y which also had a significant increase in error scores compared to the younger group aged 19-37y (P=.006). These results imply that significant reduction of proprioception decline may start even earlier than 65y and should be further investigated to know at what age it starts to significantly deteriorate. I believe It is important in clinical practice to assess for all the submodalities in proprioception, and the development of new more clinically feasible instrumentation to do so, especially for JPS, is needed. Threshold scores also need to me put in place for all the joints of the lower extremity to assess for high risk of falls since they all work together and have an impact in maintaining functional balance. JPS measurements in movement might have more of a functional significance than measuring it in rest and therefore could also be a researched to see its impact in keeping dynamic balance in older adults

• •

•

•

•

Conclusion The findings of joint proprioception deficit in older adults observed in this review are consistent with prior findings of age related proprioception joint loss. There is almost a double decline in the proprioception acuity with adults aged 65 and above. These finding have been correlated with deficiency in dynamic balance and increase in fall risk therefore becoming aware of this deterioration is important not only for physical therapists but patients as well. Threshold angular error scores have not been set for any of the lower extremity joints to marc for risk of falling and they need to be established. These threshold scores can help clinicians determine those at higher risk for falling and provide necessary treatment. In order to increase the frequency in assessing and treating proprioception, easier examination methods need to be developed and implemented. Making physical therapists and patients more aware on age related proprioception decline will hopefully foster more interest in its assessment and treatment, especially since fall cases are expected to continue to grow in the future.

Acknowledgements I would like to thank my coach, Jan-Jaap Voigt, for taking the time to answer my questions and giving instructive advice on the evolution of my review as well as Abby Daly, who also offer her support and helped to grade my articles. I also would like to thank Esther Verloop for guiding me throughout the multiple research possibilities offered by the HVA library, and making this process easier.

References

•

•

•

•

•

•

Anat A. Mechanoreceptors in articular tissue. US National Library of Medicine. 1988;182(1):16-32. Boisgontier MP, Olivier I, Chenu O, Nougier V. Presbypropria: the effects of physio- logical ageing on proprioceptive control. American Aging Association. 2012;34:1179–94. Bullock S, Wong W, Hogan N, The influence of age on weight-bearing joint reposition sense of the knee. US National Library of Medicine. 2001;136(3):400-6. Burns EB, Stevens JA, Lee RL. The direct costs of fatal and non-fatal falls among older adults—United States. J Safety Res 2016:58. Craig CE, Goble DJ, Doumas M. Proprioception acuity predicts muscle co-contraction of the tibialis anterior and gastrocnemius medialis in older adults’ dynamic postural control. Neuroscience. 2016;322:251-61. Donaghy A, DeMott T, Allet L, Kim H, Ashton-Miller J, Richardson J. Accurancy of clinical techniques for evaluating lower limb sensorymotor functions associated with increased fall risk. US National Library of Medicine. 2016;8(4):331-339.

Franco P, Santos K, Rodacki A. Joint positioning sense, perceived force level and two-point discrimination tests of young and active elderly adults. Brazilian Journal of Physical Therapy. 2015;19(4):304-310. Goble DJ, Hurvitz EA, Brown SH. Deficits in the ability to use propriocetive feedback in children with hemiplegic central palsy. Elsevier. 2009;32:267269. Goble DJ, Lewis CA, Hurvitz EA, Brown SH. Development of upper limb proprioceptive accuracy in children and adolescents. Hum Mov Sci 2005; 24:155-170. Goble DJ, Noble BC, Brown SH. Where was my arm again? Memory-based matching of proprioceptive targets is enhanced by increased target presentation time. Neurosci Lett. 2010;481:54-58.

Injury prevention & control: data & statistics. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2013. [cited 2017 October 17].URL: https://www.cdc.gov/injury/wisqars/

•

•

•

•

•

•

•

Low Choy N, Brauer S, Nitz J. Age-related changes in stregth an somatosensation during midlife. Rationale for Trgeted Preventive Intervention Programs. Ann. N.Y. Acad. Sci. 2007;1114:180193. Madhavan S, Shields RK. Influence of age on dynamic position sense: evidence using a sequential movement task. Exp Brain Res. 2005;164:18-28.

•

•

• Magnusson M, Enbom H, Johansson R, Pyykko I. Significance of pressor input from the human feet in anterior-posterior postural control: the effect of hypothermia on vibration-induced body-sway. Acta Otolaryngol: 1990 [cited 2017 October 7]. URL: http://www.tandfonline.com/doi/abs/.

•

Melzer I, Benjuya N, Kaplanski J. Postural stability in the elderly: a comparison between fallers and non fallers. Age Ageing. 2004;33:602-607.

•

NICE style guide. National institute of health and care excellence [online]. 2009. [Cited 2017 October 13]URL:https://www.nice.org.uk/corporate/ecd1/ch apter/referencing-and-citations

•

Pickard C.M, Sullivan P.E, Allison G.T, Singer K.P. Is there a difference in hip joint position sense between young and older groups? Gerentol. Med. Sci. 2003;58(7):631-635.

•

Verschueren SMP, Brumagne S, Swinnen SP, Cordo PJ. The effect of aging on dynamic position sense at the ankle. Elsevier. 2002;136:593-603.

Warner J, Lephart S Fu FH. Role of propriocpetion in pathoetiology of shoulder instability. US National Library of Medicine. 1996;330:35-9. Wingert J, Welder C, Foo P. Age-related hip proprioception declines: effect on postural sway and dynamic balance. Archives of Physical Medicine and Rehabilitation. 2014;94:253-61.

Woollacott MH, Tang P-F, Balance control during walking in the older adult: research and its implications. Phys Ther 1997;77:646-60. World Health Organization. Falls 2017. [cited 2017 November 20] URL: http://www.who.int/mediacentre/factsheets/fs344/e n/ You SH. Joint position sense in elderly fallers: a preliminery investigation of the validity and reliability of the SENSERite measure. Arch Phys Med Rehabil. 2005;86:346-352. Hawk C, Hyland J, Rupert R, Colonvega M and Hall S. Assessment of balance and risk for falls in a sample of community-dwelling adults aged 65 and older. BioMed Central. 2006;14:3.

Relph N and Harrington L. The effects of knee direction, physical activity and age on knee joint position sense. Elsevier. 2016;23:393-398. .

•

•

•

•

• •

Ribeiro, F., Mota, J., Oliveira, J. Effect of exerciseinduced fatigue on position sense of the knee in the elderly. J. Appl. Physiol. 2007;99:379-385. Ribeiro F, Oliveira J. Effect of physical therapy exercise and age on knee joint position sense. Elsevier. 2010;51:64-67. Riemann BL, Lephart SM. The sensorimotor system, part II: the role of propriocep-tion in motor control and functional joint stability. J Athl Train 2002;37:80–4. Sharrington C.S. on the proprioceptive system, especially on its reflex aspects. Brain. 1906;29:467482. Skinner H.B. Pathokinesiology and total joint arthroplasty. Clin. Orthop. 1993;288:78-86. Swash M, Fox KP. The effect of age on human skeletal muscle studies of the morphology and innervation of muscle spindles. J Neurol Sci. 1972;16:417–32.

Appendix 1 PubMed search string: (((("Aging"[Mesh:NoExp]) OR "Aged"[Mesh])) OR (aging[Text Word] OR elderly[Text Word]))) OR (("age factors"[MeSH Terms]) OR (age factor*[Text Word] OR aging effect*[Text Word] OR influence of aging[Text Word])))) OR ((lifespan[Text Word]) OR (("aged"[MeSH Terms]) OR (elderly[Text Word] OR older adults[Text Word])))))) AND (((("Proprioception"[Mesh:NoExp] AND English[lang])) OR propriocepti*[Text Word]) AND English[lang])))) AND lower extremity[MeSH Terms]