Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Aug 1.
Published in final edited form as: Clin Biomech (Bristol). 2019 Jun 14;68:197–204. doi: 10.1016/j.clinbiomech.2019.06.006

Comparing single and multi-joint methods to detect knee joint proprioception deficits post primary unilateral total knee arthroplasty

Abderrahman Ouattas a,*, Elizabeth Wellsandt b, Nathaniel H Hunt a, C Kent Boese c, Brian A Knarr a
PMCID: PMC7197211  NIHMSID: NIHMS1582420  PMID: 31238189

Abstract

Background

The use of various single-joint proprioception measurements has resulted in contradictory findings after knee arthroplasty. The use of balance as a surrogate measure to assess knee proprioception post-operation has resulted in further confusion. The aim of this study was to measure single joint knee proprioception in participants after unilateral knee arthroplasty, and compares it to multi-joint balance.

Methods

Eleven participants at 1 year after unilateral total knee arthroplasty and twelve age-matched controls were enrolled. The threshold to detect passive motion and the sensory organization test were used to measure single joint knee proprioception and multi-joint balance respectively. Two-way ANOVA and independent t-tests were used to measure differences between and within groups. Regression analysis was used to measure the association between proprioception and balance measurements.

Findings

Surgical knees demonstrated significantly more deficient proprioception compared to the non-surgical knees and both knees of the control groups during flexion (P < 0.01) and extension (P < 0.05). Non-surgical knees showed similar proprioception to both knees of the control group during flexion and extension. Within the knee arthroplasty group, only deficiencies during flexion showed significant correlation with Sensory Organization Test visual ratio. No additional differences between both groups during balance measurements, nor any correlations between local joint proprioception and balance were seen.

Interpretation

These findings indicate deficient surgical knee proprioception in participants one year after unilateral total knee arthroplasty. Limited associations between measurements indicate that balance may be a poor measure of single-joint proprioception.

Keywords: Primary unilateral total knee arthroplasty, Knee proprioception, Balance, Risk of falls

1. Introduction

Nearly all modern total knee arthroplasty (TKA) designs sacrifice the anterior cruciate ligament (ACL), while approximately half sacrifice both cruciate ligaments (American Joint Replacement Registry, 2017). It is believed that both ACL and posterior cruciate ligament (PCL) have a major contribution in providing proprioceptive information throughout all knee range of motion (Amis et al., 2006; Johansson et al., 1991). Thus, removal of cruciate ligaments during TKA may lead to knee proprioception deficits post-surgery when compared to healthy knees. This reduced knee proprioception may lead to deficient balance, and subsequently a greater risk of falls (Horlings et al., 2008). It has been suggested that deficient knee proprioception, alongside weak knee muscle extensors, may be the result of falls 4 months post-TKA (Levinger et al., 2011), despite the improvement in balance confidence in individuals who did not experience falls before surgery (Swinkels et al., 2008). However, contradictory results have been reported in literature thus far on whether surgical knee proprioception is deficient or not post-surgery, mainly due to the use of different methods to assess single-joint knee proprioception (Barrett et al., 1991; Fuchs et al., 1999; Isaac et al., 2007; Pap et al., 2000; Skinner et al., 1984; Swanik et al., 2004).

To date, two single-joint proprioception methods have been used to assess knee proprioception post-TKA; joint position reproduction (JPR) and threshold to detect passive motion (TDPM). JPR is a static test and it examines knee proprioception through passively moving the participant’s limb to a certain angle and later requires the participant to actively reproduce the angle correctly (Goble, 2010). The test is believed to be more dependent on muscle sensors, interhemispheric communication, and memory rather than ligament sensors (Isaac et al., 2007; Wikstrom et al., 2006). TDPM is a dynamic test that measures knee proprioception by testing one’s ability to indicate the knee position movement in space when passively moved in either flexion or extension directions. The test is widely used in neurophysiological studies to test the contribution of afferent neurons on proprioception through emphasizing the role of ligaments (Waddington and Adams, 1999; Wikstrom et al., 2006). Using TDPM, proprioceptive deficiencies on the surgical limb were reported in one study (Pap et al., 2000), but did not show interlimb deficiencies in another study (Skinner et al., 1984), although both show deficiencies compared to a healthy older population. Using JPR, studies have reported improved proprioception compared to pre-TKA but failed to show proprioceptive deficiencies in the surgical knees compared to the non-surgical (Barrett et al., 1991; Fuchs et al., 1999; Isaac et al., 2007). Despite the improvements, these studies do not consider pain, impaired excitation of motor units, and increased knee laxity as major contributors of decreased knee proprioception. These deficiencies are the main components of end-stage knee OA (Fortier and Basset, 2012; Ju et al., 2010; Kaya, 2015). Hence, comparing post-TKA to end stage OA conditions may not be ideal to test the effect of surgery on knee proprioception. All previously mentioned studies that used JPR showed deficient surgical knee proprioception compared to knees of healthy aged matched controls (Barrett et al., 1991; Fuchs et al., 1999; Isaac et al., 2007; Skinner et al., 1984).

Additionally, multi-joint static balance has been used as a surrogate to measure knee proprioception post-TKA (Baumann et al., 2017; Gauchard et al., 2010; Isaac et al., 2007; Vandekerckhove et al., 2015). Moreover, although different methods and equipment were used to measure balance post-TKA, three common outcome variables were linearly investigated; center of pressure (COP) sway, area, and velocity (Baumann et al., 2017; Gauchard et al., 2010; Isaac et al., 2007; Swanik et al., 2004; Vandekerckhove et al., 2015). When participants post-TKA were compared to healthy controls using the sensory organization test (SOT), visual, vestibular, and somatosensory information associated with balance were deficient 3 weeks post-op compared to healthy adults, but show no significant deficiencies 5 weeks post-op (Gauchard et al., 2010). SOT measures balance through manipulation of each of the three systems associated with balance (visual, vestibular, and somatosensory) with the intention to perturb the system and induce adaptive sensory recalibration processes. Despite the importance of somatosensory information on postural control, balance is multi-joint and multivariate activity that requires proprioception, visual, vestibular, reaction time, and muscle strength to provide feedback of the body position in space and produce the appropriate motor control to maintain stability (Akram et al., 2008; Fransson et al., 2000; Keshner et al., 1987). Its value for assessing single joint proprioception remains unclear.

Due to the specificity of each proprioceptive test, involvement of different TKA types, and the use of different methods and techniques in each test, we still lack evidence to whether surgical knee proprioception is deficient in population with both cruciate ligaments sacrificed after primary unilateral TKA. However, determining this answer to whether knee proprioception is deficient in individuals with cruciate ligament sacrificed post-TKA remains clinically relevant. This information can help researchers in the future to investigate the differences in retaining or sacrificing cruciate ligaments, and may guide surgeons to modify their surgical techniques to ultimately improve function and quality of life. As a first step, the primary aim of this study was to determine if single-joint proprioception differed in patients after primary unilateral TKA with both cruciate ligaments sacrificed when compared to that of the non-surgical and those of the age matched controls using measurement. We hypothesized that the surgical limb will show deficient proprioception compared to the age matched healthy and non-surgical knees. Second, we aimed to determine if balance is an appropriate measure of single joint knee proprioception. We hypothesized that patients with proprioceptive deficits will not show balance deficits, and that no correlation will be present between knee proprioception and balance measures.

2. Methods

2.1. Participants

Eleven individuals after primary unilateral TKA due to osteoarthritis (OA) and twelve age-matched adults participated in the current study (Table 1). Only unilateral post-TKA participants and healthy controls were recruited for this study to minimize the limitations associated with end-stage OA population that limit knee proprioception including knee pain, impaired excitation of motor units, and joint laxity. Participants were excluded if they had any lower limb joint pain > 3 in a scale from 0 to 10 (0 being absence of pain and 10 being extreme pain) in the contralateral limb in the TKA group and in both limbs in the control group. In both groups, participants were excluded if they; 1) have been diagnosed with a neurological disorder including stroke, traumatic brain injury, or any other neurological condition that affects their decision-making or ability to move normally, 2) require an assistive device to walk, or 3) have doctor-diagnosed lower extremity OA in non-surgical joint in the TKA group and all joints in the healthy group.

Table 1.

Participant demographics.

Participants Months post-TKA Age (years) Height (m) Weight (kg) BMI (kg/m2) Gender
Post-TKA (11) 11.50 (6.26) 65 (7.36) 1.7 (0.10) *86.20 (16.88) *29.79 (5.67) 6F 5M
Control (12) – 65.25 (8.68) 1.68 (0.10) 71.81 (12.62) 25.82 (3.36) 9F 3M
Open in a new tab

Values are presented as mean (standard deviation).

*

Statistical differences between groups. P < 0.05.

TKA participants were recruited from a private orthopedic clinic. A single board-certified orthopedic surgeon completed all arthroplasties. A DePuy Attune posterior stabilized cemented implant type was used. This implant uses a cam and post mechanism to substitute for the sacrificed posterior cruciate ligament. Patients had both their ACL and PCL sacrificed and the lateral and medial collateral ligaments retained. Patients underwent physical therapy sessions according to the clinic’s guidelines, which focused on improving functional outcomes, knee flexor/extensor muscle strength to achieve symmetrical strength, and achieving functional range of motion (ROM) (0–120° or better). Healthy controls were recruited from a university-based wellness center and university employees.

Participants were informed of the study procedures prior to signing the informed consent. The study was approved by the University of Nebraska Medical Center’s Institutional Review Board.

2.2. Single joint proprioception measurements

Using a Biodex dynamometer (Biodex Medical Inc., Shirley, NY, USA), the threshold to detect passive motion (TDPM) was used in this study as a measurement of knee proprioception. TDPM has been previously described and validated as being a test of dynamic proprioception to identify kinesthesia ability which mainly tests neural information delivered to the brain from joint mechanoreceptors (afferent neurons) (Wikstrom et al., 2006). Participants were seated on the chair, and the distal end of the movement arm was attached in reference to the lateral and medial malleolus using an inflated splint to minimize cutaneous pressure. The researcher manually measured each participant’s 90° knee angle using a goniometer to calibrate the system. Participants were asked to perform two flexion and extension TDPM trials for each knee. Before each trial, a practice calibration trial was given to reduce any learning effect or nervousness and anxiousness. During flexion trials, participants performed two trials while seated with their knees positioned at 50° of knee flexion (0° being maximum knee extension) and the device arm initiated passive flexion with a 1°/s angular velocity. During extension trials, participants performed two trials while seated with their knee positioned at 90° of knee flexion and the device arm initiated passive extension at the same angular velocity. The average threshold of all trials was calculated as the final score. Flexion and extension starting angles were chosen to provide optimal knee sensations compared to any angle < 30° of full extension (Pincivero et al., 2001).

Participants were given a handheld push button and were specifically instructed to press the button and stop the movement arm as soon as they felt any movement sensation. The detection angle was then subtracted from the initial angle to measure the detection angle error for each limb (Eq. (1)).

TDPM(°)=|StopAngle(°)−InitialAngle(°)| (1)

Participants were asked to keep their eyes closed while wearing a blindfold, earplugs and headsets to eliminate any auditory or visual inputs. Participants were asked not to perform any vigorous activity and were asked to rest for at least 10 min before starting the test to help control for fatigue.

2.3. Balance measurement

Participants performed the sensory organization test (SOT) using the Balance Master System 8.4 (NeuroCom International Clackamas, OR, USA) (Fig. 1). The surface carries two 22.9 × 45.7 cm force plates used to collect center of pressure (COP) data and is used to estimate center of mass (COM) sway based on individuals’ height. COM is calculated perpendicular to the middle point between the two feet when placed on equal distance to the center of the force plates as described by the manufacturer (Natus Medical Incorporated, 2016). The SOT was used according to previous research studies that used linear measures of COP sway to assess balance in population after TKA (Baumann et al., 2017; Gauchard et al., 2010; Isaac et al., 2007; Swanik et al., 2004). The SOT consists of 6 conditions, each involves 3 trials that lasts 20 s each.The smart balance master calculates the equilibrium score in each trial (Eq. (2))

EquilibriumScore=[12.5°−(GreatestCOMsway−LowestCOMsway)12.5°]×100 (2)

Fig. 1.

Fig. 1.

Open in a new tab

Balance Master System (NeuroCom International Clackamas, OR, USA). A. Standing platform: surface carries two AMTI force plates used to collect COP. The platforms rotate with 6-degrees of freedom used to perturb the somatosensory system. B. Movable surround: rotates in the AP direction to perturb the vestibular system. C. Harness. COM sway angle (°) is calculated as the angle between a vertical line projecting upward from the center of the base of support (midline between feet) and a second line projecting from the same point to the participant’s calculated COM.

12.5° represents the anterior/posterior stability limits before falling; 8.5° anteriorly and 4° posteriorly. COM sway is limited to 12.5° in both A and P directions. Equilibrium scores of 0 would indicate a fall while scores of 100 would indicate no COM sway. The composite score (CS) is then calculated dividing the sum of all equilibrium scores by the number of trials. CS above 68% represent above average scores based on the Balance Master database average results of the same age group. “Sensory Ratios” were also included and calculated based on the manufacturer’s calculations (Natus Medical Incorporated, 2016) (Table 2).

Table 2.

Sensory organization test (SOT) conditions and sensory ratios.

Variables Definition Perturbed sensory systems
SOT conditions Condition 1 (Cl) Eyes open, fixed support None
Condition 2 (C2) Eyes closed, fixed support Inhibited visual
Condition 3 (C3) Sway-referenced visual surround, fixed support Vestibular and visual
Condition 4 (C4) Eyes open, sway-referenced support Somatosensory
Condition 5 (C5) Eyes dosed, sway-referenced support Inhibited visual & perturbed somatosensory
Condition 6 (C6) Sway-referenced visual surround, sway-referenced visual support Visual, somatosensory, vestibular
Variables Outcome Conditions involved
Sensory ratios Somatosensory (RSOM) Poor use of somatosensory sense C2C1
Visual (RVIS) Poor use of visual sense C4C1
Vestibular (RVEST) Poor use of available vestibular sense C5C1
Visual preference (RPREF) Use of inaccurate visual senses C3+C6C5+C2
Altered proprioceptive information management (RPMAN) Poor compensation of lost somatosensory senses C4+C5+C6C1+C2+C3
Open in a new tab

Foot placement during the SOT was adjusted based on participants’ height according to the NeuroCom’s guidelines. During the test, participants were wearing a harness providing no body weight support which was attached to the system to insure the safety of each participant. Participants were informed that the researchers would not be interacting with them during each trial. They were also asked to maintain balance as best as they could.

2.4. Patient reported balance measurements

The Activities-specific Balance Confidence (ABC) scale has been used as a clinical tool to assess subjective balance (Powell and Myers, 1995). The ABC is a valid and reliable tool used in neurological and vestibular deficient population, as well in community dwelling elderly and fallers (Peretz et al., 2006; Powell and Myers, 1995). The ABC scale measures confidence in performing several different activities of daily living with complete steadiness and postural control. The ABC consists of 16 questions and the overall score is presented in percentage. Low (scores below 50%) and medium (scores between 50 and 80%) scores indicate low and moderate functional levels respectively (Peretz et al., 2006) and are both associated with increased risk of falls (Peretz et al., 2006). High scores (higher than 80%) indicate high functional levels and are believed to be associated with reduced risk of falls (Peretz et al., 2006). Participants completed the self-administered ABC scale.

2.5. Data analysis

Independent t-test was used to measure the statistical demographic differences between the two groups. Shapiro-Wilk test was used to analyze all data for normality. Mann-Whitney U test was used to compare variables that are not normally distributed. Levene’s test was used to test homoscedasticity of all variables. Welch’s test was used to test variables with unequal variance. Two-way repeated measures ANOVAs [2 × 2; Knee side TDPM (Surgical vs Non-surgical vs right vs left) × Group (TKA vs control)] were performed to analyze the differences in knee proprioception between all knees (Surgical and non-surgical knees of the TKA group, and right and left knees of the control group). Independent t-test was used to assess the differences between groups with normally distributed variables while Independent Mann-Whitney U test was used to test the differences between groups in variables that did not show normal distribution.

The TDPM values were standardized as a percentage of proprioceptive deficiencies between limbs in each group to measure the relationship between TDPM and Balance (Eq. (3)). The percentage of proprioceptive deficiencies was calculated as the differences between the surgical (Sx) and non-surgical (NSx) raw TDPM divided by Sx TDPM (Eq. (3)). Higher percentages indicate larger proprioceptive deficits in the surgical knee when compared to the non-surgical knee.

TDPMDeficiency(%)=SxTDPM−NSxTDPMSxTDPM×100 (3)

Regression analyses were performed to determine the relationship between TDPM deficiencies and SOT’s conditions and sensory ratios in each group. Values were presented as coefficient of determination. Custom MATLAB codes (Mathworks, Natick, MA) and SPSS (IBM, v.23, Somers, NY) were used for statistical analyses, and significance level was set at 0.05.

3. Results

The TKA group showed significantly higher BMI and weight compared to the control group (Table 1). No additional demographic differences were seen. Shapiro-Wilk test showed normal distribution in all variables except the following (P < 0.05); ABC scale (both groups), SOT C2 (TKA), C4 (Control), C4 (Control), C5 (Control), C6 (TKA), RSOM (TKA), RVIS (Control). RVEST (Control), and RPMAN (Control). Based on Levene’s test, all variables showed equal error variance across groups except for SOT C6 (P < 0.05). Pairwise comparisons following a two-way ANOVA showed significantly higher TDPM in the surgical knee compared to the non-surgical knee as well as both knees of those of the control group during flexion (P < 0.01) and extension (P < 0.05) TDPM (Table 3). No significant differences were found between the non-surgical knee and healthy control knees during flexion and extension trials (Table 3). TDPM showed mean angle difference of 2.9° (SD 3.2°) between the Sx and NSx knees during flexion and 1.6° (SD 2.5°) during extension trials in the TKA group. The control group showed mean angle difference of 0.5° (SD 0.3°) between the right and left knees during flexion and 0.4° (SD 0.3°) during extension trials.

Table 3.

TDPM findings & power analysis.

Variables Knee side Flexion THPM Extension THPM
Participants Post-TKA Sx **4.67 (3.94) *3.58 (3.2)
NSx 1.80 (0.88) 1.96 (0.90)
Control Right 1.70 (0.68) 1.78 (0.58)
Left 1.78 (0.83) 1.85 (0.88)
Open in a new tab

Values are presented as mean (standard deviation).

**

Two-way ANOVA P < 0.001

*

Two-way ANOVA P < 0.05.

Two-way ANOVA showed differences between the surgical knee and all other knees.

Both groups scored above average composite scores (CS) based on the Balance Master database of the same age groups; TKA group scored 73% SD 3% while the control group scored 73% SD 6% (Fig. 2). The TKA group showed significantly more stable balance during condition 6 of the SOT (P < 0.05) (Fig. 2). No additional significant differences were found between the TKA and control groups during all conditions and sensory ratios of the SOT (Figs. 2 & 3).

Fig. 2.

Fig. 2.

Open in a new tab

Sensory organization test (SOT) scores between the TKA and control groups. C1–C6 represent scores from condition 1 through 6 between healthy & TKA groups. CS represents the overall composite score between healthy & TKA groups. Dashed red line represent the average scores based on the Balance Master database for the corresponding age groups. Error bars represent standard deviation. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 3.

Fig. 3.

Open in a new tab

SOT sensory ratios between TKA and control groups. Somatosensory ratio (RSOM). Visual ratio (RVIS). Vestibular ratio (RVIST). Visual preference ratio (RPREF). Altered proprioception information management ratio (RPMAN). Error bars represent standard deviation.

In the TKA cohort, no correlations were recorded between TDPM deficiencies (flexion and extension) and all SOT conditions and ratios. In addition, no correlations were recorded between TDPM and SOT conditions and ratios in the control group.

Similar to SOT conditions and sensory ratio scores, patient reported outcomes (ABC scale) did not show any significant differences between the TKA and control groups. In addition, both groups demonstrated high scores; TKA scored 94.22% SD 5.20% while the age matched healthy adults scored 94.41% SD 5.52%. No significant correlations were found between TDPM and ABC scores.

4. Discussion

This primary aim of this study was to determine if participants after unilateral TKA with sacrificed cruciate ligaments show reduced knee proprioception using single-joint proprioception measurement. Our main finding indicates proprioceptive deficits in the surgical knees compared to the non-surgical knees and those of the healthy controls. The second aim investigated the validity of using multi-joint balance as a surrogate for single joint proprioception measurement. Unlike single joint proprioception measurement, TKA participants did not show deficient balance on the SOT, but instead results showed significantly better balance during the most somatosensory demanding condition, which supports our first hypothesis. However, we did report one association between TDPM deficiency into flexion and SOT visual ratio, which do not completely support our second hypothesis, although no associations were seen with any ratios or conditions perturbing somatosensory input.

Across studies that used the TDPM method, our findings mostly closely reflects those reported by Pap (Pap et al., 2000) as they showed similar trend of deficient surgical knee proprioception compared to the non-surgical and healthy controls. However, our findings contradict the data presented by Skinner (Skinner et al., 1984), as they reported no proprioceptive deficiencies in the surgical knee when compared to the non-surgical. Our results also contradicts findings reported in studies that used JPR to test proprioceptive deficits before and after TKA and between surgical and non-surgical knees; as participants post-TKA showed no deficiencies in the operated knee compared to the non-operated knee (Barrett et al., 1991; Fuchs et al., 1999; Isaac et al., 2007; Skinner et al., 1984). The contradictions reported in these studies could be explained by the different methods (JPR & TDPM) and varying group characteristics (pre-TKA vs post-TKA) used in each study. The JPR is a static test that was used in previous research studies to either match the surgical knee angle with the ipsilateral knee (Barrett et al., 1991; Fuchs et al., 1999; Isaac et al., 2007; Skinner et al., 1984) or the contralateral healthy knee (Levinger et al., 2011) to measure the segment angle relative to the previous segment, and thus mainly relies on memory, muscle movement, and inter-hemispheric communication (Wikstrom et al., 2006). These factors represent difficulties in measuring the mechanoreceptors around the knee capsule and joint that are affected by TKA surgery. Instead, the JPR test measures a variety of mechanisms that can be affected by other factors independently associated with TKA, such as reduction in memory consolidation or a sedentary lifestyle. JPR may not be an appropriate measure of the direct cause of reduced proprioception due to sacrificed ligaments and replaced biological tissues during TKA. In addition, multiple studies tested post-operation conditions compared to pre-surgery (Barrett et al., 1991; Isaac et al., 2007). OA leads to elevated pain, cartilage and bone loss and increased knee ligament laxity (Barrett et al., 1991; Fortier and Basset, 2012; Fuchs et al., 1999; Isaac et al., 2007; Ju et al., 2010; Kaya, 2015); since these factors can reduce proprioception, comparing comparison of the surgical to the non-surgical limb may be a better method to explain knee proprioception improvements reported post-TKA.

TDPM was used in this study because it is a direct measure of knee kinesthesia and rate of movement. It relies on afferent neurons with no cognitive function to initiate any lower limb movement except the reaction time needed to push the hand held button (Wikstrom et al., 2006). Surgical angle deficiencies present in this study (Flexion = 4.67° SD 3.94°; Extension = 3.58 SD 3.21) were higher than those presented by Géza Pap et al., 2000 (2.2° SD 0.9°) when using the TDPM and lower than those presented by Skinner (Skinner et al., 1984) (8.8° SD 3.2°). This could be explained by the different methods used between each study. In the current study, we separated flexion and extension trials, and participants were specifically told which direction the knee would be moving to eliminate any confusion and results in the optimal proprioception (Refshauge et al., 1995). Previous reports combined both flexion and extension trials together and did not inform participants on the direction of movement (Pap et al., 2000; Skinner et al., 1984), which may lead to patient confusion. We used a higher angular velocity of 1.0°/s compared to previous studies that used 0.5°/s–0.6°/s (Pap et al., 2000; Skinner et al., 1984), since it was reported that higher angular velocities would lead to faster detection rates (Refshauge et al., 1995). Specifically in the knee joint, 1.0°/s was reported to have no significant angle detection differences with as high as 50°/s, as compared to velocities lower than 1.0°/s that showed significant angle detection differences compared to angular velocities higher than 1.0°/s (Refshauge et al., 1995). In addition, we only recruited unilateral post-TKA participants with both cruciate ligaments sacrificed and medial and lateral collateral ligaments retained in the surgical knee and healthy contralateral joints. The higher detection error reported in our results compared to Pap (Pap et al., 2000) could be due to the sacrificed cruciate ligaments as they did not mention the specific sacrificed and retained ligaments during surgery. We used a starting angle of 50° in flexion and 90° in extension as was explained by Pincivero et al. (2001) that more extended knee angles > 30°, results in less knee movement before detection. The methods used in this study was carefully chosen to produce the highest sensitivity and exclude all factors that lead to higher detection angle irrelevant to joint proprioception. Using higher angular velocity, separating flexion/extension trials, instructing patients on the direction of movement to eliminate any confusion, using only primary unilateral TKA, and using the Biodex (Biodex Medical Inc., Shirley, NY, USA) would also explain the lower detection errors reported in this study compared to Skinner (Skinner et al., 1984). Based on our results, and comparisons with previous literature, TDPM may be used as a direct measure of kinesthesia in the knee in clinical settings with the methods employed in this study.

Secondly, this study aimed to determine if balance is an appropriate measure of single joint proprioception. Unlike previous results reported by Gauchard et al. (2010), participants in this study at 1 year after TKA failed to show less stable balance compared to healthy controls during all SOT conditions and sensory ratios, but instead showed more stable balance during somatosensory demanding conditions (Condition 5 & 6 with statistical significance recorded in condition 6). Gauchard et al. (2010) used an earlier testing period after surgery (3–5 weeks post-TKA) which may account for different findings between studies. As hypothesized, TKA participants who showed knee proprioceptive deficits did not show any balance deficits compared to age-matched healthy adults. Our second hypothesis was also supported since no correlations were recorded between knee proprioception and all SOT conditions and ratios. Two theories could explain the reason behind the reported higher balance control in participants post-TKA compared to healthy controls. During low perturbations, the ankle joint controls sway, while hip strategies are used during high perturbations (Creath et al., 2005). Further, participants with lower limb proprioceptive loss rely more on hip movement compared to ankle joint control (Horlings et al., 2008). Hence, the first theory may indicate that TKA participants compensated for the deficient knee proprioception using ankle and hip joints strategies to control sway. On the other hand, a second theory may indicate the methods used to estimate COM sway from participant’s height and linear COP sway as calculated in the Balance Master may not accurately measure postural control. Horlings et al. reported that unlike participants with vestibular loss, participants with lower limb proprioceptive loss can only be identified from vestibular loss and healthy participants using AP and ML measurement of pelvis sway, as oppose of just measuring the AP sway (Horlings et al., 2008). Those differences can only be identified with a combination of conditions similar to condition 1, 4, and 5 of the SOT (Horlings et al., 2008). If so, one can hypothesize that the use of traditional measures of linear COP sway may not sufficiently test postural sway in unhealthy population with a higher risk of falls. Clark et al. was the only group that used linear and non-linear calculations of COP to measure differences in static balance between 4 and 12 weeks after TKA (Clark et al., 2017). Interestingly, only high velocity ML COP path length, AP COP mean instantaneous frequency and AP COP Detrended Fluctuation Analysis (DFA) showed significant differences between the two timelines post-TKA when compared to linear measures presented as peak COP displacement, COP standard deviation (SD), and COP root mean square (RMS) (Clark et al., 2017). The SOT uses COP displacement as a measure of balance, and thus, the second theory may be more suited than the first theory according to findings presented by Clark et al. (2017). To our best knowledge, Gauchard et al. were the only group with a previously recorded study to measure the differences between participants 3–5 weeks post-TKA and age-matched controls; their findings showed more stable balance in the control group, which is one more piece of evidence to strengthen the second theory (Gauchard et al., 2010). Moreover, subjective report of balance through the ABC scale did not detect any significant differences between groups. Both groups showed high ABC scores with a minimum of 81% in the TKA group and 83% in the control group. Nevertheless, balance is a multi-joint task that requires somatosensory, visual, vestibular, reaction time, and muscle strength to produce the appropriate motor control (Bascuas et al., 2013; Hill et al., 2013). Based on the results presented in this paper, balance is not an appropriate measure of single joint proprioception, specifically knee joint proprioception deficits post-TKA.

Deficient surgical knee proprioception may lead to a more risk of falls post-TKA (Levinger et al., 2011), and in order to test this hypothesis, future research should focus on measuring balance using dynamic methods. The SOT measures standing balance without changing feet’s contact with the ground while introducing perturbations, but individuals are more prone to falls during locomotion. Hence, the effect of knee proprioception deficits on dynamic balance (maintaining balance during locomotion) should be investigated post-TKA in order to investigate its contribution to risk of falls. One limitation of the current study is the small sample size. However, the difference in knee flexion TDPM demonstrated a large effect size of 0.22 (partial eta squared) with a power of 0.85. A moderate effect size of 0.11 was present during extension TDPM with a power of 0.493. Another limitation was the lack of muscle strength measurements. However, it has been reported that the strength of knee extensor and flexor muscles are not contributors to knee proprioception post-TKA (Wada et al., 2002), but it may be necessary for future research to measure their contribution on balance alongside knee proprioception post-TKA. The equilibrium score of the SOT only accounts for the COM sway angle in the AP direction, which could be a limitation, although the SOT is a standard clinical assessment. Future research may also take physical activity levels into account and study its effect on knee proprioception and lower limb motor control post-TKA, as it was previously reported that intensive training leads to enhanced knee detection angle recognition using the TDPM (Lephart et al., 1996; Skinner et al., 1984). Although we report significant deficient surgical knee proprioception detection angle in participants after unilateral TKA with both cruciate ligaments retained and collateral ligaments sacrificed, we are not certain if retaining cruciate ligaments may lead to enhanced knee proprioception due to the lack of including a TKA comparison group with cruciate ligaments retained. Hence, future research should investigate the differences between retained and sacrificed cruciate ligaments (ACL & PCL) on knee proprioception post-TKA.

5. Conclusion

Our main finding indicates participants one year after unilateral TKA with cruciate ligaments sacrificed and medial and lateral collateral ligaments retained show deficient surgical knee proprioception compared to non-surgical and knees of age matched individuals. Since individuals with cruciate ligaments sacrificed show deficient knee proprioception post-surgery, one should investigate how retaining cruciate ligaments may lead to reduced proprioception deficiencies in surgical knee. Moreover, linear measures of COP as an indicator of balance was a generally poor predictor of individual joint proprioception, however additional balance measures should be investigated as limited evidence of the association between proprioception and balance was found. We should also focus on investigating the effects of deficient surgical knee proprioception on dynamic balance during locomotion to minimize risk of falls and improve functional outcomes post-TKA.

Acknowledgements

This work was supported by the National Institutes of Health [NIH/NIGMS P20 GM109090 and NIH/NICHD R15 HD094194] and the Office of Research and Creative Activity at the University of Nebraska Omaha. The authors would like to thank Tyler Hamer, Katlyn Nimtz and Monica Barajas for their assistance in data collection and processing.

Role of external funding sources

No involvement in the study design, collection, analysis and interpretation of data, or in the writing of the manuscript or decision to submit for publication.

Footnotes

Declaration of Competing Interest

None.

References

  1. Akram SB, Frank JS, Patla AE, Allum JHJ, 2008. Balance control during continuous rotational perturbations of the support surface. Gait Posture 27 (3), 393–398. 10.1016/J.GAITPOST.2007.05.006. [DOI] [PubMed] [Google Scholar]
  2. American Joint Replacement Registry, 2017. Fourth AJRR Annual Report on Hip and Knee Arthroplasty Data. pp. 1–71. http://www.ajrr.net/images/annual_reports/AJRR-2017-Annual-Report—Final.pdf?hsCtaTracking=61a2fedc-0f48-4241-a282-2bca6e4b68ad%7Ce4ed88d8-6b8e-4d0c-9a47-d1c169759e83.
  3. Amis AA, Gupte CM, Bull AMJ, Edwards A, 2006. Anatomy of the posterior cruciate ligament and the meniscofemoral ligaments. Knee Surgery, Sport Traumatol Arthrosc 14 (3), 257–263. 10.1007/s00167-005-0686-x. [DOI] [PubMed] [Google Scholar]
  4. Barrett DS, Cobb AG, Bentley G, 1991. Joint proprioception in normal, osteoarthritic and replaced knees. J Bone Joint Surg Br 73 (1), 53–56. 10.4172/2157-7595.1000119. [DOI] [PubMed] [Google Scholar]
  5. Bascuas I, Tejero M, Monleón S, Boza R, Muniesa JM, Belmonte R, 2013. Balance 1 year after TKA: correlation with clinical variables. Orthopedics 36 (1), e6–12. 10.3928/01477447-20121217-11. [DOI] [PubMed] [Google Scholar]
  6. Baumann F, Bahadin Ö, Krutsch W, et al. , 2017. Proprioception after bicruciate-retaining total knee arthroplasty is comparable to unicompartmental knee arthroplasty. Knee Surgery, Sport Traumatol Arthrosc 25 (6), 1697–1704. 10.1007/s00167-016-4121-2. [DOI] [PubMed] [Google Scholar]
  7. Clark RA, Seah FJT, Chong HC, et al. , 2017. Standing balance post total knee arthroplasty: sensitivity to change analysis from four to twelve weeks in 466 patients. Osteoarthr. Cartil 25 (1), 42–45. 10.1016/j.joca.2016.08.009. [DOI] [PubMed] [Google Scholar]
  8. Creath R, Kiemel T, Horak F, Peterka R, Jeka J, 2005. A unified view of quiet and perturbed stance: simultaneous co-existing excitable modes. Neurosci. Lett 377 (2), 75–80. 10.1016/J.NEULET.2004.11.071. [DOI] [PubMed] [Google Scholar]
  9. Fortier S, Basset FA, 2012. The effects of exercise on limb proprioceptive signals. J. Electromyogr. Kinesiol 22 (6), 795–802. 10.1016/j.jelekin.2012.04.001. [DOI] [PubMed] [Google Scholar]
  10. Fransson P-A, Johansson R, Hafström A, Magnusson M, 2000. Methods for evaluation of postural control adaptation. Gait Posture 12 (1), 14–24. 10.1016/S0966-6362(00)00052-7. [DOI] [PubMed] [Google Scholar]
  11. Fuchs S, Thorwesten L, Niewerth S, 1999. Proprioceptive function in knees with and without total knee arthroplasty. Am J Phys Med Rehabil 78 (1), 39–45. http://www.ncbi.nlm.nih.gov/pubmed/9923428, Accessed date: 19 June 2017. [DOI] [PubMed] [Google Scholar]
  12. Gauchard GC, Vançon G, Meyer P, Mainard D, Perrin PP, 2010. On the role of knee joint in balance control and postural strategies: effects of total knee replacement in elderly subjects with knee osteoarthritis. Gait Posture 32 (2), 155–160. 10.1016/j.gaitpost.2010.04.002. [DOI] [PubMed] [Google Scholar]
  13. Goble DJ, 2010. Proprioceptive acuity assessment via joint position matching: from basic science to general practice. Phys. Ther 90 (8), 1176–1184. 10.2522/ptj.20090399. [DOI] [PubMed] [Google Scholar]
  14. Hill KD, Dip Physio G, Williams SB, Chen J, Moran H, Hunt S, 2013. Balance and falls risk in women with lower limb osteoarthritis or rheumatoid arthritis. J Clin Gerontol Geriatics 4, 22–28. 10.1016/j.jcgg.2012.10.003. [DOI] [Google Scholar]
  15. Horlings CGC, Küng UM, Bloem BR, et al. , 2008. Identifying deficits in balance control following vestibular or proprioceptive loss using posturographic analysis of stance tasks. Clin. Neurophysiol 119 (10), 2338–2346. 10.1016/j.clinph.2008.07.221. [DOI] [PubMed] [Google Scholar]
  16. Isaac SM, Barker KL, Danial IN, Beard DJ, Dodd CA, Murray DW, 2007. Does arthroplasty type influence knee joint proprioception? A longitudinal prospective study comparing total and unicompartmental arthroplasty. Knee 14 (3), 212–217. 10.1016/j.knee.2007.01.001. [DOI] [PubMed] [Google Scholar]
  17. Johansson H, Sjölander P, Sojka P, 1991. Receptors in the knee joint ligaments and their role in the biomechanics of the joint. Crit. Rev. Biomed. Eng 18 (5), 341–368. http://www.ncbi.nlm.nih.gov/pubmed/2036801, Accessed date: 8 June 2018. [PubMed] [Google Scholar]
  18. Ju Y-Y, Wang C-W, Cheng H-YK, 2010. Effects of active fatiguing movement versus passive repetitive movement on knee proprioception. Clin. Biomech 25 (7), 708–712. 10.1016/j.clinbiomech.2010.04.017. [DOI] [PubMed] [Google Scholar]
  19. Kaya Defne, 2015. Proprioception: The Forgotten Sixth Sense. OMICS Group eBooks https://pdfs.semanticscholar.org/e226/45948d2a9d2552821b614a23b636b9dfc170.pdf, Accessed date: 9 June 2018. [Google Scholar]
  20. Keshner EA, Allum JHJ, Pfaltz CR, 1987. Postural coactivation and adaptation in the sway stabilizing responses of normals and patients with bilateral vestibular deficit. Exp. Brain Res. 69 (1), 77–92. 10.1007/BF00247031. [DOI] [PubMed] [Google Scholar]
  21. Lephart SM, Giraldo JL, Borsa PA, Fu FH, 1996. Knee joint proprioception: a comparison between female intercollegiate gymnasts and controls. Knee Surg. Sports Traumatol. Arthrosc 4 (2), 121–124. http://www.ncbi.nlm.nih.gov/pubmed/8884734, Accessed date: 20 June 2018. [DOI] [PubMed] [Google Scholar]
  22. Levinger P, Menz HB, Wee E, Feller JA, Bartlett JR, Bergman NR, 2011. Physiological risk factors for falls in people with knee osteoarthritis before and early after knee replacement surgery. Knee Surgery, Sport Traumatol Arthrosc 19 (7), 1082–1089. 10.1007/s00167-010-1325-8. [DOI] [PubMed] [Google Scholar]
  23. Natus Medical Incorporated, 2016. Sensory Organization Test (SOT) SOT Comprehensive Report Sensory Organization Test (SOT) Equilibrium Score. Pleasanton, CA: http://balanceandmobility.com/wp-content/uploads/018528A_NCM_SOT_brochure_EN-US_lo-res.pdf, Accessed date: 7 June 2018. [Google Scholar]
  24. Pap G, Meyer M, Weiler H, Machner A, Awiszus F, 2000. Proprioception after total knee arthroplasty: a comparison with clinical outcome. Acta Orthop. Scand 71 (2), 153–159. 10.1080/000164700317413120. [DOI] [PubMed] [Google Scholar]
  25. Peretz C, Herman T, Hausdorff JM, Giladi N, 2006. Assessing fear of falling: can a short version of the activities-specific balance confidence scale be useful? Mov.Disord 21 (12), 2101–2105. 10.1002/mds.21113. [DOI] [PubMed] [Google Scholar]
  26. Pincivero DM, Bachmeier B, Coelho AJ, 2001. The effects of joint angle and reliability on knee proprioception. Med. Sci. Sports Exerc 33 (10), 1708–1712. [DOI] [PubMed] [Google Scholar]
  27. Powell LE, Myers AM, 1995. The Activities-Specific Balance Confidence (ABC) scale. Journals Gerontol - Ser A Biol Sci Med Sci 50A (1), M28–M34. 10.1093/gerona/50A.1.M28. [DOI] [PubMed] [Google Scholar]
  28. Refshauge KM, Chan R, Taylor JL, Mccloskey DI, 1995. Detection of movements imposed on human hip, knee, ankle and toe joints. J. Physiol. 488 (1), 231–241. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1156716/pdf/jphysiol00308-0229.pdf, Accessed date: 10 June 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Skinner HB, Barrack RL, Cook SD, Haddad RJ, Haddad RJ Jr., 1984. Joint position sense in total knee arthroplasty. J. Orthop. Res 1 (5), 276–283. 10.1002/jor.1100010307. [DOI] [PubMed] [Google Scholar]
  30. Swanik CB, Lephart SM, Rubash HE, 2004. Proprioception, kinesthesia, and balance after total knee arthroplasty with cruciate-retaining and posterior stabilized prostheses. J. Bone Joint Surg. Am 86-A (2), 328–334. [DOI] [PubMed] [Google Scholar]
  31. Swinkels A, Newman JH, Allain TJ, 2008. A prospective observational study of falling before and after knee replacement surgery. Age Ageing 38 (2), 175–181. 10.1093/ageing/afn229. [DOI] [PubMed] [Google Scholar]
  32. Vandekerckhove PJTK, Parys R, Tampere T, Linden P, Van den Daelen L, Verdonk PC, 2015. Does cruciate retention primary total knee arthroplasty affect proprioception, strength and clinical outcome? Knee Surgery, Sport Traumatol Arthrosc 23 (6), 1644–1652. 10.1007/s00167-014-3384-8. [DOI] [PubMed] [Google Scholar]
  33. Wada M, Kawahara H, Shimada S, Miyazaki T, Baba H, 2002. Joint proprioception before and after total knee arthroplasty. Clin. Orthop. Relat. Res 403, 161–167. http://www.ncbi.nlm.nih.gov/pubmed/12360022, Accessed date: 10 June 2018. [DOI] [PubMed] [Google Scholar]
  34. Waddington G, Adams R, 1999. Discrimination of active plantarflexion and inversion movements after ankle injury. Aust J Physiother 45 (1), 7–13. http://www.ncbi.nlm.nih.gov/pubmed/11676748, Accessed date: 24 October 2017. [DOI] [PubMed] [Google Scholar]
  35. Wikstrom EA, Tillman MD, Chmielewski TL, Borsa PA, 2006. Measurement and evaluation of dynamic joint stability of the knee and ankle after injury. Sports Med. 36 (5), 393–410. http://www.ncbi.nlm.nih.gov/pubmed/16646628, Accessed date: 10 June 2018. [DOI] [PubMed] [Google Scholar]

RESOURCES