Volume 89, Issue 10 , Pages 1965-1969, October 2008
Bilateral Frontal Plane Mechanics After Unilateral Total Knee Arthroplasty
Article Outline
Abstract
Milner CE, O'Bryan ME. Bilateral frontal plane mechanics after unilateral total knee arthroplasty.
Objective
To compare frontal plane knee mechanics among the operated and nonoperated limbs after total knee arthroplasty (TKA) and a healthy control limb.
Design
A cross-sectional analysis with age-matched control group.
Setting
A biomechanics and sports medicine laboratory.
Participants
Subjects (n=16; 8 men, 8 women; mean age, 61±7y; height, 1.71±0.10m; weight, 87.5±15.1kg) and age-matched healthy controls (n=16; 8 men, 8 women; mean age, 63±7y; height, 1.7±.09m; weight 72.5±13.9kg).
Interventions
Not applicable.
Main Outcome Measures
Peak knee adduction angle, first peak knee adduction moment, and the frontal plane knee angle and moment at loading peak during the stance phase of walking.
Results
Peak knee adduction angle (P=.176), and the frontal plane knee angle (P=.116) and moment (P=.260) at loading peak were similar across the operated, nonoperated, and healthy control limbs. The first peak knee adduction moment was higher in the nonoperated limb when compared with the operated limb and with the healthy control (P=.003). First peak knee adduction moment was similar in the operated knee and the healthy control.
Conclusions
The greater first peak knee adduction moment in the nonoperated knee indicates a possible mechanism for the predictable deterioration of this knee after unilateral TKA of the contralateral knee.
Key Words: Adult, Aged, Arthroplasty, Replacement, knee, Gait, Kinematics, Kinetics, Rehabilitation
List of Abbreviations: ES, effect size, OA, osteoarthritis, TKA, total knee arthroplasty
OSTEOARTHRITIS IS ONE of the most common sources of disability in aging Americans today, affecting nearly 20% of all adults.1 More specifically, the knee is the most common site for OA to develop and cause disability.1 The number of TKAs increased over 80% from 1990 to 2000.2 The surgery has proven to be very successful in reducing the pain associated with OA and improving function of the operated limb.3 However, recent epidemiologic evidence suggests that a primary TKA leads to a predictable pattern of deterioration in the other major joints of the lower extremities, which may ultimately result in a second joint replacement.4 A statistically predictable pattern of deterioration after an initial knee replacement was determined in this study. More specifically, among those with an initial TKA for OA, about one third went on to have a second joint replacement.4 Of these, 92% had the contralateral knee replaced.4 This predictable pattern of deterioration in the contralateral knee suggests that gait mechanics in the contralateral limb should be considered in this population.
Ideally, gait patterns would return to normal following a TKA. However, evidence suggests that abnormal gait patterns observed presurgery are retained postsurgery.5, 6 Smith et al5 found that around 70% of patients who had abnormal flexor or extensor gait patterns postsurgery showed those same patterns before surgery.5 This retention of presurgery gait patterns suggests that gait mechanics do not spontaneously revert to a normative pattern after TKA.
The predictable pattern of progression of OA in the contralateral knee joint after a primary unilateral TKA4 suggests that there may be asymmetry between the operated and nonoperated limb after TKA. However, there has been little consideration of the contralateral limb in the literature. Typically, studies of people with TKA have examined knee mechanics in the operated limb only. Recently, researchers have begun to examine the effects of TKA on the contralateral limb.7, 8 However, these 2 studies made comparisons between limbs, but did not compare the results with age-matched controls. Comparison with a healthy control group would indicate whether differences in either the operated or nonoperated limb were due to the TKA or simply reflective of gait changes occurring with age.
Many studies have found differences in the gait of subjects with knee OA, the precursor of TKA compared with healthy age-matched controls.8, 10, 11, 12 Furthermore, it has been suggested that OA initiation and progression are directly related to an abnormal magnitude and distribution of load on the joint.12, 13 Both frontal and sagittal plane knee mechanics are affected by the progression of OA.9, 10, 11, 12, 14, 15 Changes in the frontal plane in particular have been associated with progression to severe OA.9, 10, 14 More specifically, subjects with advanced OA of the knee have been found to walk with a greater peak external knee adduction moment than healthy controls9, 10, 12 or the uninvolved limb.15 In addition, subjects with advanced OA have also been found to walk with a greater peak knee adduction angle.10 This evidence is from studies of subjects with knee OA. However, it suggests that frontal plane mechanics in both the operated and nonoperated knees after unilateral TKA may be important in the progression of OA in the nonoperated knee.
The purpose of this study was to compare frontal plane knee mechanics among the operated and nonoperated limbs after TKA and a healthy control limb. Specifically, we compared the peak knee adduction angle and the first peak external knee adduction moment among the limbs. In addition, because OA progression appears to be related to loading, we also compared the frontal plane angle and frontal plane moment of the knee at the vertical loading peak during the weight-acceptance phase of stance. It was expected that these variables would be greater in both the operated and the nonoperated knee of the TKA group compared with the healthy controls. Interlimb comparisons within the TKA group will indicate whether asymmetry was present in frontal plane knee mechanics after TKA.
Methods
Participants
A group of 16 subjects (8 men, 8 women; age, 61±7y; height, 1.71±0.1m; weight, 87.5±15.1kg), who had undergone a unilateral TKA previously, were compared with an age-matched healthy control group (8 men, 8 women; age, 63±7y; height, 1.7±.09m; weight, 72.5±13.9kg). Activity levels for each group, scored out of 18 (Lower Extremity Activity Scale16), were 12±3 in the arthroplasty group and 15±2 in the control group. We recruited all participants from the local community through newspaper advertisements and word of mouth, as well as with flyers in local fitness centers, church meeting halls, and senior centers. Before participation in the study, each subject gave their written informed consent to participate. The institution's human subjects review board approved all procedures prior to the start of the study.
We excluded participants aged over 75 years or with a body mass index of 40kg/m2 or higher (extremely obese) from the study. Additionally, participants who reported any pain in the lower back, or lower extremities, including the feet were excluded. Consequently, all subjects were asymptomatic for OA in their lower-extremity joints at the time of participation in the study. We excluded participants from the TKA group if they were still receiving treatment or physical therapy for the surgery, if they had another lower-extremity joint replaced, or if they were currently injured. Subjects in the arthroplasty group were an average of 28 months postsurgery (range, 4–96mo). Participants in the control group were free of lower-extremity injuries or surgeries that would affect their gait, and had never undergone a partial or total joint arthroplasty at any lower-extremity joint. We also excluded participants from either group if they had a history of rheumatoid arthritis; they were unable to walk without an aid (eg, cane); or they were unable to provide informed consent or follow directions.
Data Collection
We used a 7-camera motion capture systema to collect gait data for the lower extremity at 120Hz. For the collection of ground reaction force data, we used 2 force platformsb sampling at 1200Hz and synchronized with the motion capture system. We positioned the force platforms so that 2 consecutive steps could be recorded. To measure walking velocity, we used 2 photocells connected to a timer. Retro-reflective markers were placed on the pelvis and both lower extremities to determine the 3-dimensional kinematics of walking.
We placed 2 molded thermoplastic shells with 4 noncolinear markers on each lower extremity. One was placed on the lateral aspect of the proximal thigh, and the other on the posterior aspect of the distal shank. To establish the position of the rear foot, we placed 3 noncolinear markers directly onto each heel. Prior to data collection, several markers were placed on each lower extremity of the subject to establish the anatomic coordinate system for each segment. Anatomical markers were placed at the greater trochanter, lateral, and medial knee at the level of the lateral femoral epicondyle, lateral, and medial malleolus at the level of the lateral malleolus, and the first and fifth metatarsal head of each lower extremity. These markers were removed after the standing calibration trial was completed.
During data collection, subjects wore the same model of hiking sandalc to reduce the influence of footwear on their gait. The subjects were instructed to walk at their own pace across the laboratory (arthroplasty group, 1.25±0.17m/s; control group, 1.42±0.16m/s). We collected consecutive left and right stance phase data as the subject contacted each forceplate. Several trials were conducted before data collection to establish a consistent walking speed and a starting point that resulted in each foot landing cleanly on the force platform. If the subject did not make clean contact with the forceplate, or appeared to alter their gait in order to make contact, that trial was disregarded. Five acceptable trials were collected.
Data Analysis
The data reported here are part of an ongoing study of gait after TKA. We processed all data in Visual 3D.d Marker trajectories were low-pass filtered at 6Hz and kinetic data were low-pass filtered at 50Hz with recursive fourth order Butterworth filters. Peak knee adduction angle was the maximum adduction angle of the knee during stance. First peak external knee adduction moment was the maximum adduction moment during the first half of stance phase. The vertical loading peak was the maximum vertical ground reaction force during the first half of stance phase. The frontal plane angle and moment at loading peak were calculated at that point. Both the first peak external knee adduction moment and the moment at loading peak were normalized to each subject's fat-free weight17 and height. Normalizing for lean body mass in this way removes the influence of differing amounts of body fat between subjects. All variables were calculated for each of the 5 trials, then averaged within the subject and across each limb condition (operated, nonoperated, control). Because there was no reason to prefer a particular side, we used the right knee of the control group for comparison with the TKA limbs.
The study design was a cross-sectional analysis comparing the TKA group with an age-matched control group asymptomatic for OA. Because the knee arthroplasty group was no longer receiving postsurgical rehabilitation treatment, they were considered to be recovered from surgery. Power calculations for this study were done using data from Mundermann et al9 for first peak external knee adduction moment. Sample size was based on predicted power to detect a difference between the groups of 0.4% of body weight by height, with an α of .05 and 80% power. This difference in first peak knee moment was reported by Mundermann between patients with severe OA and controls. Based on calculations made in SamplePower,e a minimum sample size of 14 subjects per group was indicated. Therefore, the inclusion of 16 subjects per group should provide adequate power to detect a difference between groups. Sample-size calculations could not be made for the other variables tested, because data were not available in the literature. However, ES was used to assist in the interpretation of differences between the groups. Each variable was compared among all 3 groups using a 1-way analysis of variance test. If a significant difference was found, a post-hoc Tukey test was used to make pairwise comparisons between the groups to identify where the differences lay.
Results
Peak knee adduction angle was similar across the operated, nonoperated, and control knees (P=.176) (table 1, fig 1). This finding was supported by moderate (operated vs nonoperated knee ES=.60), small (control vs nonoperated knee ES=.47), and no effects (control vs operated knee ES=.16) for the pairwise comparisons. However, the first peak external knee adduction moment was significantly greater in the nonoperated knee compared with both the operated and control knees (P=.003) (see table 1, fig 2). Furthermore, there was a large effect for knee adduction moment of the nonoperated knee compared with both the operated (ES=1.1) and the control (ES=1.06) knees. There was no significant difference (no effect; ES=.19) between the operated knee and the healthy control knee.
Table 1. Frontal Plane Knee Variables in the Operated, Nonoperated, and Control Limbs
| Limb | Peak Knee Adduction Angle (deg) | First Peak Knee Adduction Moment⁎ | Loading Peak | |
|---|---|---|---|---|
| Frontal Plane Knee Angle (deg) | Frontal Plane Knee Moment⁎ | |||
| Operated | 1.8±3.6 | −0.047±0.008 | 0.0±3.0 | −0.038±0.014 |
| Nonoperated | 4.3±4.3 | −0.059±0.015 | 2.8±3.9 | −0.047±0.021 |
| Control | 2.4±3.7 | −0.045±0.013 | 0.8±3.6 | −0.040±0.021 |
| P | .176 | .003 | .116 | .260 |
⁎Normalized to fat-free weight by height. |
Fig 1.
Average curves for the control, operated, and nonoperated knees after a primary, unilateral TKA for knee adduction and abduction angle during the stance phase of walking (adduction and varus angles are positive).
Fig 2.
Average curves for the control, operated, and nonoperated knees after a primary, unilateral TKA for knee adduction and abduction moment during the stance phase of walking (external knee adduction moment is positive). Abbreviation: Nm/ffw
·h, normalized to fat free weight by height.
There was no difference in frontal plane knee angle at loading peak between the knees (P=.116) (see table 1). This finding was supported by moderate (operated vs nonoperated knee ES=.70; control vs nonoperated knee ES=.55) to no effects (control vs operated knee ES=.12) of the pairwise comparisons. Frontal plane knee moment at loading peak was also similar among the groups (P=.26) (see table 1). This result was supported by moderate (operated vs nonoperated knee ES=.53), small (control vs nonoperated knee ES=.42), and no effects (control vs operated knee ES=.15) for the pairwise comparisons.
Discussion
This study compared frontal plane knee mechanics after unilateral TKA bilaterally and with a healthy control group to determine potential effects on the contralateral knee of the primary TKA. Peak knee adduction angle was similar in the operated and nonoperated knees when compared with the healthy control knee. It was expected that both the operated and nonoperated knee would have greater than normative knee adduction angles after TKA. Astephen and Deluzio10 found that subjects with severe OA walk with a larger adduction angle when compared with healthy controls. This finding was expected to be observed in the operated knees after TKA, because Smith et al5, 6 observed retention of presurgery gait patterns in postsurgery gait analysis. It might also be assumed that subjects at risk for OA would walk with a greater than normative adduction angle, although it is unclear whether the angle reported by Astephen and Deluzio10 was a cause or effect of OA. Consequently, we expected to find a greater adduction angle in the nonoperated knee, which is at risk for developing severe OA, according to Shakoor et al.4 However, because these were cross-sectional studies, it is unclear whether subjects who are at risk for developing severe OA walk with a greater adduction angle prior to development of symptomatic OA. The present study does not support the hypothesis that knee adduction angle is an important factor in the predictable deterioration of the nonoperated knee.
The first peak knee adduction moment was greater in the nonoperated knee when compared with both the operated and the healthy control knees but was similar in the operated and control knees. This partially supports our hypothesis that first peak knee adduction moment would be greater in the operated and nonoperated limbs compared with the healthy control. Because greater knee adduction moments have been linked to OA progression in adults,9, 10, 11, 12, 15 these results suggest that the nonoperated knee is at higher risk for developing OA than a healthy control of similar age. Shakoor et al4 found a predictable pattern of OA progression, indicated by subsequent joint replacement, after a primary joint replacement. The greater knee moment found in the present study may indicate a mechanism for this predictable pattern of deterioration of the contralateral knee. The greater moment may lead to further development and progression of OA in the nonoperated limb, leading ultimately to joint replacement. Of course, in this cross-sectional study, cause and effect cannot be determined. However, this finding indicates a direction for future longitudinal, prospective studies of OA progression after primary TKA.
In particular, recent evidence has suggested that factors such as foot mechanics and hip strength may play an important role in the development and progression of knee OA. Decreased internal hip abduction moment has been linked with the progression of knee OA over an 18-month period.18 A mechanism for this relationship was proposed by Chang et al,18 such that hip abductor muscle weakness on the ipsilateral side results in a change in frontal plane alignment when the limb is in stance. Weak hip abductors lead to greater pelvic drop on the swing limb side, resulting in a shift in the body's center of mass toward the swing limb. In turn, this increases the load on the medial compartment of the stance knee. Similarly, foot mechanics may also influence the knee. Several studies have reported that foot progression angle is related to internal knee abduction moment. Rutherford et al19 reported that the shape of the frontal plane knee moment curve during stance can be explained partly by foot progression angle in mild to moderate OA. However, this study is limited by being a cross-sectional analysis; therefore cause and effect cannot be determined. Similarly, Guo et al20 reported that increasing toe-out angle in subjects with moderate OA resulted in a reduction in the second peak internal knee abduction moment. Furthermore, a longitudinal study of knee OA progression over an 18-month period18 found that subjects with greater toe-out angles had not deteriorated as much compared with those with less toe-out. Given the greater knee moment found in the present study, these relationships with hip strength and toe-out angle indicate areas for further investigation in relation to decreasing the risk of development and progression of OA in the nonoperated knee.
Loading has been identified as a mechanism of OA development and progression.13 Therefore, we proposed that frontal plane mechanics at the point of peak ground reaction force during weight acceptance by the stance limb would be a distinguishing feature between the operated, nonoperated, and control limbs. However, there were no differences among any of the knees for the frontal plane knee angle and moment at loading peak. It appears that the absolute peak moment during the first half of stance is a better indicator of interlimb difference. The general trend for the knee moment across the groups is the same for both the value at loading peak and the maximum value (higher in the uninvolved knee), but the differences are less pronounced at loading peak. This observation was supported by the small to moderate effect sizes of the comparisons, which indicate that the magnitude of the differences between groups are much less than the intersubject differences within each group at loading peak. These data suggest that knee moment at loading peak is not a good discriminatory variable when comparing knees with and without arthroplasty. Therefore, we recommend that first peak internal knee abduction moment be used to compare knee status after arthroplasty, particularly in relation to potential development and progression of OA.
It should be noted that both groups walked at their freely chosen normative walking speed and the arthroplasty group walked approximately 12% (.17m/s) slower than the control group. It is unclear what effect this small difference in walking speed had on the biomechanic variables measured. There has been some investigation into the effect of speed on gait mechanics, but these have typically compared grossly different speeds. However, Hanlon and Anderson21 used regression methods to predict the change in average peak knee flexion angle during stance with a 0.5-m/s increase in walking speed in healthy adults. This difference is approximately 3 times greater than observed between our subject groups, and was predicted to change peak knee flexion during stance by 1.5°.21 Therefore, a change in velocity similar to the difference between the arthroplasty and control groups might result in a change in peak knee flexion angle of around 0.5°. They did not report the effects of walking speed on frontal plane mechanics; therefore direct conclusions cannot be drawn. However, these data suggest that this small difference in walking velocity likely had minimal effects on knee mechanics.
Study Limitations
Some limitations of this study should be noted. We recruited all subjects from the surrounding community; the TKA subjects were not referred by a specific surgeon. Consequently, the arthroplasty group was somewhat heterogeneous with a wide range of postoperative times being reported. In addition, the subjects may have had different rehabilitation protocols and may have been fitted with different implants. However, the knee range of motion required for gait is well within the goals for rehabilitation and implant design. Furthermore, the heterogeneity of the groups may be more representative of the general TKA population. It may be useful in future studies to include specific details of implant type and rehabilitation protocol. It should also be noted that the control group was functioning at a high level, with no lower extremity joint pain or previous surgeries. This group should be considered a criterion standard for gait in older adults. The cross-sectional design of this study is also a limitation because it provides only a snapshot of the subjects in the TKA and control groups. No cause and effect relationships can be determined. Cause and effect can only be determined from longitudinal studies that follow participants and record knee biomechanics before surgery and at several intervals postsurgery. However, this study does provide an indication of gait mechanics after recovery from TKA compared with healthy older adults with no lower-extremity pain or previous surgery.
Our results have important clinical implications in terms of rehabilitation strategies after TKA. Although the study was a cross-sectional comparison of persons after knee replacement and healthy older adults, it highlights the importance of considering the contralateral limb after a unilateral TKA. Very few studies have considered the nonoperated limb, but given the predictable pattern of deterioration,4 it is clear that this is a worthwhile focus for future study regarding optimized rehabilitation protocols. In particular, it may be worthwhile to emphasize the importance of breaking the common presurgical habit of relying on the nonoperated limb during walking. This might involve more emphasis on how functional tasks such as walking are completed, in addition to whether they can be completed and how quickly. Typically, objective functional tests used to monitor progress during rehabilitation focus on the ability to complete a walking task and the time taken to do so. This study points toward the importance of taking the biomechanics of both knees into consideration in assessing the success of rehabilitation in returning the patient to normative function. Once the mechanics operated in this deterioration are better understood, target rehabilitation interventions can be developed to lessen the effect that a primary unilateral TKA will have on the contralateral limb. These interventions might include gait retraining, muscle strengthening, or prophylactic bracing. The aim of these interventions would be to correct the abnormal mechanics in the contralateral knee after TKA to reduce the risk of further joint degeneration.
Conclusions
Peak knee adduction angle, as well as frontal plane knee angle and moment at loading peak, were similar among the operated, nonoperated, and healthy control limbs. However, first peak knee adduction moment was greater in the nonoperated knee when compared with both the operated knee and the healthy control, with the first peak knee adduction moment being similar in the operated knee and healthy control. These results indicate the importance of focusing on the nonoperated limb, in addition to the operated, during rehabilitation after a unilateral TKA.
Suppliers
References
- The effects of specific medical conditions on the functional limitations of elders in the Framingham Study. Am J Public Health. 1994;84:351–358
- . Trends in total knee replacement surgeries and implications for public health, 1990-2000. Public Health Rep. 2005;120:278–282
- . The risk of contralateral total knee arthroplasty after knee replacement for osteoarthritis. J Rheumatol. 2003;30:1822–1824
- . Nonrandom evolution of end-stage osteoarthritis of the lower limbs. Arthritis Rheum. 2002;46:3185–3189
- . Pre-surgery knee joint loading patterns during walking predict the presence and severity of anterior knee pain after total knee arthroplasty. J Orthop Res. 2004;22:260–266
- . A kinematic and kinetic analysis of walking after total knee arthroplasty with and without patellar resurfacing. Clin Biomech (Bristol, Avon). 2006;21:379–386
- . Altered loading during walking and sit-to-stand is affected by quadriceps weakness after total knee arthroplasty. J Orthop Res. 2005;23:1083–1090
- . Interlimb asymmetry during walking following unilateral total knee arthroplasty. Gait Posture. 2008;28:69–73
- . Secondary gait changes in patients with medial compartment knee osteoarthritis: increased load at the ankle, knee, and hip during walking. Arthritis Rheum. 2005;52:2835–2844
- . Changes in frontal plane dynamics and the loading response phase of the gait cycle are characteristic of severe knee osteoarthritis application of a multidimensional analysis technique. Clin Biomech (Bristol, Avon). 2005;20:209–217
- . Gender differences exist in osteoarthritic gait. Clin Biomech (Bristol, Avon). 2007;22:400–409
- Increased knee joint loads during walking are present in subjects with knee osteoarthritis. Osteoarthritis Cartilage. 2002;10:573–579
- . A hypothesis matrix for studying biomechanical factors associated with the initiation and progression of posttraumatic osteoarthritis. Med Hypotheses. 2005;64:1157–1161
- Knee adduction moment, serum hyaluronan level, and disease severity in medial tibiofemoral osteoarthritis. Arthritis Rheum. 1998;41:1233–1240
- . Associations among knee adduction moment, frontal plane ground reaction force, and lever arm during walking in patients with knee osteoarthritis. J Biomech. 2006;39:2213–2220
- Development and validation of a lower-extremity activity scale (Use for patients treated with revision total knee arthroplasty). J Bone Joint Surg Am. 2005;87:1985–1994
- . Body mass index as a measure of body fatness: age- and sex-specific prediction formulas. Br J Nutr. 1991;65:105–114
- Hip abduction moment and protection against medial tibiofemoral osteoarthritis progression. Arthritis Rheum. 2005;52:3515–3519
- . Foot progression angle and the knee adduction moment: a cross-sectional investigation in knee osteoarthritis. Osteoarthritis Cartilage. 2008;16:883–889
- . The influence of foot progression angle on the knee adduction moment during walking and stair climbing in pain free individuals with knee osteoarthritis. Gait Posture. 2007;26:436–441
- . Prediction methods to account for the effect of gait speed on lower limb angular kinematics. Gait Posture. 2006;24:280–288
No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated.
PII: S0003-9993(08)00438-3
doi:10.1016/j.apmr.2008.02.034
© 2008 American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation. Published by Elsevier Inc. All rights reserved.
Volume 89, Issue 10 , Pages 1965-1969, October 2008
