- Email: [email protected]
Patellofemoral joint stress during stair ascent and descent in persons with and without patellofemoral pain
Gait and Posture 16 (2002) 115 /123 www.elsevier.com/locate/gaitpost
Patellofemoral joint stress during stair ascent and descent in persons with and without patellofemoral pain Jacklyn Heino Brechter a,*, Christopher M. Powers b a
b
Department of Physical Therapy, Chapman University, One University Drive, Orange, CA 92866, USA Musculoskeletal Biomechanics Research Laboratory and Department of Biokinesiology and Physical Therapy, University of Southern California, Los Angeles, CA, USA Accepted 30 November 2001
Abstract Objective: To determine if persons with patellofemoral pain (PFP) demonstrate elevated patellofemoral joint (PFJ) stress during stair ascent and descent when compared to persons without PFP. Design: A cross sectional study utilizing an experimental and a control group. Background: Ascending and descending stairs is one of the most painful activities of daily living for persons with PFP. Whether or not the pain associated with stair ambulation is the result of elevated joint stress (force per unit area) has not been explored. Methods: 10 subjects with a diagnosis of PFP and 10 subjects without pain completed two phases of data collection, (1) MRI assessment to determine PFJ contact area and (2) comprehensive motion analysis during stair ambulation at self selected climbing velocities. Data obtained from both data collection sessions were utilized as input variables into a biomechanical model to quantify PFJ stress. Results: Although the knee extensor moment and PFJ reaction force (PFJRF) were significantly reduced in the PFP subjects during stair ascent, there was no difference in PFJ stress between groups. Similarly, there were no differences in PFJ stress during stair descent. Conclusion: Our results do not support the hypothesis that subjects with PFP demonstrate greater joint stress during stair ascent and descent compared to subjects without pain. However, subjects with PFP appeared to maintain normal levels of PFJ stress by minimizing the PFJRF. This was accomplished through a slower cadence and a reduced knee extensor moment. Relevance: PFP is a common syndrome causing pain and functional limitations during stair climbing and other activities requiring high levels of quadriceps activity. Information obtained from this study will be useful in understanding the biomechanical mechanisms contributing to functional deficits in the PFP population. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Patella; Patellofemoral; Biomechanics; Stress; Stairs; Gait
1. Introduction Patellofemoral pain (PFP) affects approximately one of four people in the general population [1]. The clinical presentation of PFP varies between individuals, but some features are considered representative. For example, pain is commonly described as being retropatellar or along the medial and lateral borders of the patella. Symptoms are typically exacerbated with sustained sitting (movie-goers knee) and activities requiring high levels of quadriceps activity (i.e. running, squatting, and negotiating stairs) [2,3]. From a functional standpoint,
* Corresponding author. Tel.: /1-714-744-7649; fax: /1-714-7447621 E-mail address: [email protected] (J.H. Brechter).
ascending and descending stairs is one of the most painful activities of daily living for persons with PFP. It has been reported that stair climbing places a higher demand on the knee when compared to level walking as demonstrated by increased knee extensor moments (an indication of quadriceps demand) and greater ranges of knee motion [4,5]. As the patellofemoral joint reaction force (PFJRF) is dependent on the magnitude of the quadriceps force (QF) and the knee flexion angle [6], the compressive force acting between the patella and femoral trochlea during stair ascent and descent would be expected to be significant. In fact, Matthews et al. have reported that the PFJRF during stair ambulation is more than three times that of level walking [7]. The increased PFJRF associated with stair ambulation suggests that patellofemoral joint (PFJ) stress (force per unit contact area) would be elevated as well.
0966-6362/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 9 6 6 - 6 3 6 2 ( 0 2 ) 0 0 0 9 0 - 5
J.H. Brechter, C.M. Powers / Gait and Posture 16 (2002) 115 /123
116
Elevated PFJ stress in the PFP population is of significant concern as stress is thought to be the factor responsible for articular cartilage degeneration and wear [6,8]. Whether or not individuals with PFP demonstrate higher magnitudes of PFJ stress compared to nonpainful individuals has not been explored. Evidence does exist however, suggesting that this may be the case. Hebert et al. reported that individuals with PFP demonstrated increased knee extensor moments during a squatting activity when compared to control subjects [9]. This finding implies that the PFJRF may be elevated as well. In addition, PFP has been reported to be associated with patellar malalignment [6,10,11] which can result in reduced PFJ contact area [12 /14] and an increase in PFJ stress. Using an imaging based biomechanical model of the PFJ (which takes into consideration the inherent variability in PFJ contact areas between individuals), the purpose of this study was to quantify PFJ stress in persons with and without PFP during stair ascent and descent. It was hypothesized that persons with PFP would demonstrate higher PFJ stress compared to a individuals without PFP. Information obtained from this study may be useful in understanding the biomechanical mechanisms contributing to functional deficits in the PFP population.
activities commonly associated with PFP: (a) stair ascent or descent, (b) squatting, (c) kneeling, (d) prolonged sitting, (e) isometric quadriceps contraction [15,16]. Subjects with PFP were excluded from participation if they reported having any of the following: (1) previous history of knee surgery; (2) history of traumatic patellar dislocation; (3) any neurological involvement that would influence gait; (4) any implanted biological devices, such as pacemakers, cochlear implants, clips which could interact with the magnetic field during imaging. Subjects in the comparison group were recruited from the University of Southern California, and matched for gender to those in the PFP group. Inclusion criterion for participation in the comparison group were as follows: (1) no history or diagnosis of knee pathology or trauma; (2) no knee pain with any of the activities described as inclusion criterion for the PFP group; (3) no limitations present that would influence gait; and (4) no implanted biological devices, such as pacemakers, cochlear implants, clips which could interact with the magnetic field during imaging. Prior to participation, all subjects were fully informed as to the nature of the study, and signed a human subjects consent form approved by the Institutional Review Board of the University of Southern California Health Sciences campus. 2.2. Procedure
2. Methods 2.1. Subjects Twenty subjects were recruited for this study, 10 individuals with a diagnosis of PFP (five males, five females) and ten individuals without PFP (five males, five females) (Table 1). The PFP subjects were recruited from orthopaedic clinics in the Los Angeles area. For purposes of this study, PFP subjects were screened to rule out ligamentous instability, internal derangement, or patellar tendonitis. Subjects were accepted into the study if they met the following inclusion criterion (1) pain originating specifically from the patellofemoral articulation (vague or localized) and (2) reproducible pain with at least two of the following functional
All subjects completed two phases of data collection. Phase one consisted of MRI assessment to determine PFJ contact area while phase two consisted of comprehensive motion analysis during stair ascent and stair descent. Data obtained from both data collection sessions were required as input variables into a biomechanical model to quantify PFJ stress. 2.2.1. Magnetic resonance imaging All imaging was performed at the Los Angeles County/University of Southern California Imaging Science Center. Images of the PFJ were obtained using a 1.5T magnet (GE Medical Systems, Milwaukee, WI) and a three-dimensional spoiled gradient recalled echo (3D SPGR) imaging sequence. The following parameters were employed: TR /60 ms, TE /20 ms, Flip
Table 1 Subject characteristics, means (standard deviations) PFP
Age (years) Height (cm) Weight (kg)
Conrtol
Male (n 5)
Female (n 5)
Combined (n 10)
Male (n 5)
Female (n 5)
Combined (n 10)
38.2 (7.7) 178.9 (10.7) 78.1 (16.0)
36.0 (13.4) 166.1 (6.4) 63.4 (8.4)
37.1 (10.4) 167.9 (17.8) 70.8 (14.3)
32.2 (7.0) 177.7 (10.4) 78.1 (7.9)
31.8 (8.1) 163.7 (7.4) 57.6 (12.3)
32 (7.1) 167.2 (4.4) 67.9 (14.5)
PFP, patellofemoral pain group.
J.H. Brechter, C.M. Powers / Gait and Posture 16 (2002) 115 /123
Angle /308, NEX /1.5, matrix size: 512 /224 /28, field of view: 20 /20 cm2 and chemically selective fat suppression. Each slice was 2-mm thick and contiguous with adjacent slices. Prior to scanning, subjects removed any metal such as jewelry and hair clips. Subjects were positioned supine in the MRI Bore with the knee in 08 of knee flexion. One receive-only extremity coil was secured on each side of the PFJ. Subjects rested quietly while the scan was performed. Following completion of the first scan, subjects were re-positioned and the scan repeated with the knee supported in three additional knee flexion angles 20, 40, and 608. Total imaging time was 44 min. 2.2.2. Motion analysis Motion analysis was performed in the Musculoskeletal Biomechanics Research Laboratory at the University of Southern California. Three-dimensional kinematics were obtained using a six-camera motion analysis system (Vicon, Oxford Metrics Ltd, Oxford, England). Movement was sampled at 60 Hz and recorded digitally on an IBM 166 MHz personal computer. Reflective markers (20 mm spheres) placed at specific anatomical landmarks were used to determine motion of the pelvis and lower extremity. Ground reaction forces were collected at a rate of 2500 Hz using an AMTI force plate (Model #OR6-6-1, Newton, Mass). This force plate was situated as the first of a three-step staircase (step height /20.5 cm, tread /27.5 cm) (Fig. 1). Subjects were appropriately attired to permit marker placement directly on the skin of the subject’s pelvis and
117
lower extremity. The involved lower extremity was instrumented for subjects with PFP, while the right lower extremity was instrumented for subjects without PFP. Anthropometric measures were obtained from each subject for use in calculating lower extremity kinetics using inverse dynamics equations. Reflective markers were then taped to the following landmarks: sacrum, anterior superior iliac spine bilaterally, lateral thigh, lateral femoral epicondyle, lateral tibia, lateral malleolus, 2nd metatarsal head, and posterior calcaneous. Subjects were allowed several practice trials to accommodate to the stair apparatus. All participants were instructed to walk in a step over step fashion at a self selected pace. Two trials for each ascending stairs and descending stairs were obtained for each subject. A trial was considered successful if the subject’s instrumented foot landed within the force plate. All kinematic and kinetic data were collected simultaneously. 2.3. Data analysis 2.3.1. Patellofemoral joint contact area Sequential sagittal plane images of the PFJ were displayed for analysis using Signa Advantage medical imaging software (GE Medical Systems). The section of the image containing the patella and surrounding portion of the femur was enlarged to 2.5 times normal view to enhance visualization of the contact area. Contact was defined as areas of patella and femoral approximation in which no distinct separation could be
Fig. 1. Arrangement of the force plate and portable staircase. This arrangement permitted the force plate to become one of the steps during stair ascent and descent.
J.H. Brechter, C.M. Powers / Gait and Posture 16 (2002) 115 /123
118
found between the cartilage borders of the two structures. Since cartilage is relatively bright on the 3D SPGR images obtained, the definition of contact area was operatively defined as ‘white on white’ [17]. Contact area measurements were made twice and averaged. The line of contact between the patella and femur was measured and recorded using the same software used to display the images. When the line of contact was curved, separate straight-line segments were measured. To obtain the contact area for each slice, the length was multiplied by the 2-mm slice thickness. Contact areas calculated from each image were summed to obtain the total PFJ contact area, with values reported in mm [2]. This method has been shown to be reliable and comparable to contact area obtained using Fuji pressure sensitive film in cadaver specimens [17]. Contact area measurements were made twice and averaged. All MRI measurements were made by the same investigator. 2.3.2. Knee joint kinematics and kinetics Reflective markers were identified manually using Vicon Clinical Manager (VCM) software and then automatically digitized. The same software was used to calculate knee joint kinematics and kinetics in the sagittal plane. Only data corresponding to the force plate step was used. To calculate the net joint moment at the knee, anthropometric data, ground reaction forces, and kinematics were used to solve inverse dynamics equations. To facilitate comparison between subjects and groups, the net knee joint moments were normalized by body mass and reported in units of Nm/kg. Data obtained from the two trials were averaged. 2.3.3. Biomechanical model A PFJ model was developed to utilize input from mechanical analysis of the lower extremity and PFJ contact area from the MRI. An overview of the model is illustrated (Fig. 2). Input variables for the model algorithm included knee joint flexion angle, knee extensor moment, and PFJ contact area. The first step of the algorithm was to calculate the QF. The effective lever arm (LA) for the quadriceps muscle group was determined for each knee joint angle position using an equation fit to the data from van Eijden et al. Eq. (1) [18]. At each joint angle, the knee extensor moment (MEXT) was divided by the lever arm to obtain QF Eq. (2). LA(x) (8:0e 05 x3 0:013x2 0:28x0:046)
(1)
2
x , tibiofemoral joint angle; r /0.99 QF(ui ) MEXT (ui )=QLA (ui )
(2)
u , knee flexion angle; i /1 to x ; x, number of frames of data
Fig. 2. Flow chart of PFJ model. *Data obtained from van Eijden et al. [18]. **Data obtained from van Eijden et al. [19].
The second step of the algorithm was to calculate the PFJRF. A constant (k ) was then calculated for each joint angular position by fitting an equation to the data reported by van Eijden et al. Eq. (3) [19]. This constant was multiplied by the QF to obtain the PFJRF Eq. (4). k(x) (4:62e 01 1:47e 03 x 3:84e 05 x2 )=(1 1:62e 02 x 1:55e 04 x2 6:98e 07 x3 )
(3)
x , tibiofemoral joint angle; r2 /0.99 PFJRF(ui ) k QF(ui )
(4)
k , constant; u , knee flexion angle; i /1 to x ; x , number of frames of data. Using the four contact area values obtained from MRI, the final step in the algorithm was to calculate PFJ stress. A straight line was fit between each two consecutive data points to provide approximate PFJ contact area values for each knee flexion angle from 08 to 608. For knee flexion angles greater than 608, the straight line fit between contact area at 408 and 608 was extrapolated out to the maximum knee flexion angle measured. PFJ stress was then calculated by dividing the PFJRF by the contact area for the knee flexion angle corresponding to the PFJRF value Eq. (5) PFJStress (ui ) PFJRF(ui )=Area(ui )
(5)
Area, contact area; u , knee flexion angle; i /1 to x ; x , number of frames of data. The model output was PFJRF, PFJstress, and utilized contact area (contact area corresponding to a given knee flexion angle). Each variable was normalized to the stance phase of stair ascent or descent.
J.H. Brechter, C.M. Powers / Gait and Posture 16 (2002) 115 /123
119
2.4. Statistical analysis
3.3. Net knee joint moments
Variables used for statistical analysis included cadence, peak PFJstress, PFJ stress-time integral, peak PFJRF, PFJRF-time integral, peak knee extensor moment, peak knee joint flexion and extension angle, and mean utilized contact area. The PFJ stress-time integral and the PFJRF-time integral were calculated by taking the area under the PFJstress curve and the PFJRF curve respectively. The use of the normalized PFJRF and PFJ stress data for calculation of the integrals allowed for easier comparison of the integral data between groups. The mean utilized contact area was calculated by taking the average of the utilized contact area over the stance period. Comparisons between groups were made using trimmed means and Yuen’s test for comparison of two independent groups with unequal variances [20]. This analysis was repeated for each variable and significance levels were set at P B/0.05. All statistics were computed using a custom macro written in Microsoft Excel ’97.
During stair ascent, the PFP group demonstrated a significantly lower peak knee extensor moment when compared to the control group (0.81 vs. 1.16 Nm/kg; P /0.04) (Fig. 4). The same trend was observed during stair descent (0.98 vs. 1.12 Nm/kg), however this difference was not statistically significant (P /0.09) (Fig. 4). 3.4. Patellofemoral joint reaction force During stair ascent, peak PFJRF was significantly less in the PFP group when compared to the control group (25.0 vs. 37.3 N/kg; P /0.02) (Fig. 5). In addition, the PFJRF-time integral, during stair ascent, was significantly less in the PFP group when compared to the control group (288.2 vs. 501.9 N/kg s% Stance Phase; P /0.01) (Fig. 5). In contrast, there was no significant group difference with respect to the peak PFJRF during stair descent. However, there was a trend toward reduced PFJRF-time integral in the PFP group during stair descent (464.4 vs. 605.9 s % Stance Phase; P /0.07) (Fig. 5).
3. Results 3.5. Utilized patellofemoral joint contact area 3.1. Cadence During stair ascent, the PFP group adopted a significantly slower cadence compared with the control group (115.9 vs. 142.1 steps/min; P /0.016). Similarly, the PFP group demonstrated a significantly reduced cadence compared with the control group during stair descent (115.4 vs. 153.6 steps/min; P /0.004).
3.2. Knee kinematics There was no significant difference in knee kinematics between the PFP group and the control group during stair ascent or stair descent (Fig. 3).
There was no significant difference in the mean utilized contact area between the PFP and control groups during stair ascent or stair descent (Fig. 6). 3.6. Patellofemoral joint stress During stair ascent, there was no significant difference in peak PFJ stress between the PFP group and the control group (Fig. 7). Furthermore, there was no significant group difference with respect to the PFJ stress-time integral (Fig. 7). Similarly, during stair descent there was no significant group difference for either peak PFJ or the PFJ stress-time integral (Fig. 7). Although peak PFJ stress was not significantly different between groups, the time at which peak PFJ stress occurred was significantly earlier for the PFP group
Fig. 3. Knee joint angle plotted as a function of the stance phase for both the PFP group and the control group (CTRL) during (A) ascending stairs and (B) descending stairs. There was no significant difference between groups for peak knee flexion or peak knee extension.
120
J.H. Brechter, C.M. Powers / Gait and Posture 16 (2002) 115 /123
Fig. 4. Net knee joint moment plotted as a function of the stance phase for both the PFP group and the CTRL group during (A) ascending stairs and (B) descending stairs. *Indicates that peak extension moment is significantly smaller in the PFP group when compared with the CTRL group.
when compared to the control group (50 vs. 76% Stance Phase; P /0.05) (Fig. 7).
4. Discussion As persons with PFP typically complain of pain during stair ambulation, it was hypothesized that these individuals would demonstrate elevated PFJ stress when compared to individuals without PFP. In other words, the elevated PFJ stress would act to cause the PFP. This hypothesis was not supported by the current study as no significant group differences were found for either peak PFJ stress or PFJ stress-time integral during stair ascent or descent. When evaluating the individual components of PFJ stress, however, group differences were observed. During stair ascent, peak PFJRF and the PFJRF-time integral were significantly reduced in the PFP group, with values being 33 and 43% lower respectively than the control group. Thus, the patients could be trying to avoid PFP by reducing the magnitude of the force across the PFJ. The reduced PFJRF in the PFP group was attributed to a 50% reduction in the peak knee extensor moment as there were no differences in knee kinematics. The lower knee extensor moment and PFJRF are suggestive of quadriceps avoidance, which has been proposed by Powers and colleagues to be a compensatory strategy by which subjects with PFP reduce the
forces acting across the PFJ [16]. The reduced knee extensor moment was likely achieved through a significant reduction in stair climbing cadence as moments have been shown to be influenced by walking speed [21]. Although not measured, the PFP group was observed to employ a forward trunk lean which could have been an additional strategy to reduce the knee extensor moment. Such a strategy, as reported by Ernst et al., would bring the body center of mass closer to the knee joint center subsequently reducing the moment about the knee joint [22]. Despite the fact that the PFJRF was reduced in the PFP group during stair ascent, PFJ stress was similar between groups. This discrepancy can be explained by the trend toward decreased utilized contact area in the PFP group. Although not statistically significant, the observed 14% reduction in contact area helped offset the decrease in stress that would be expected with a reduction in PFJRF (the result being a stress curve comparable to the control group). In contrast to stair ascent, there were no significant differences in the peak PFJRF or the PFJRF-time integral during stair descent. Despite a trend towards decreased knee extensor moment, no significant group differences were found in the knee extensor moment, in spite of the lower cadence in the PFP group (75% of normal). The inability of PFP subjects to reduce the PFJRF during stair descent, as found in stair ascent, may be related to differences in compensation options
Fig. 5. PFJRF plotted as a function of the stance phase for both the PFP group and the CTRL group during (A) ascending stairs and (B) descending stairs. *Indicates that peak PFJRF is significantly smaller in the PFP group when compared with the CTRL group. **Indicates that the PFJRF-time integral is significantly smaller in the PFP group when compared with the CTRL group.
J.H. Brechter, C.M. Powers / Gait and Posture 16 (2002) 115 /123
121
Fig. 6. Utilized PFJ contact area plotted as a function of the stance phase for both the PFP group and the CTRL group during (A) ascending stairs and (B) descending stairs. There was no significant difference between groups.
between these two tasks. The lack of significant differences in the PFJRF during stair descent, as found in stair ascent, may be related to differences in compensation options between these two tasks. For example, subjects with PFP selected a similar cadence during the two tasks (115.9 and 115.4 steps/min respectively for stair ascent and stair descent) yet the peak knee moment during stair ascent was less than 83% of the peak during stair descent. A similar yet larger difference was noted in peak PFJRF, with the peak during stair ascent 76% as large as that during stair descent. Thus, the increase in the knee moment and PFJRF during stair descent was not a function of cadence and must be attributed to other factors. Although the PFJ stress-time integral during stair descent was similar between groups, the pattern of PFJ stress was different, with peak PFJ stress occurring significantly earlier in the stance phase for the PFP group. This early peak in PFJ stress may be explained by the reduced utilized contact area at lesser knee flexion angles in the PFP group. Given similar shaped PFJRF curves, the smaller utilized contact area would result in greater peak PFJ stress during the first 60% of the stair descent stance phase. As the knee flexion angle increased (during the last 40% stair descent stance phase) the PFJ contact area in the PFP group also increased (relative to the control group), resulting in decreased PFJ stress late in the stance phase. The decreased contact area in the
PFP group at lesser knee flexion angles (ie. 0 /308) is consistent with the premise that patellar malalignment and subsequent reduction in contact area would likely be evident before the patella becomes firmly seated in the trochlear groove [13]. Comparison of the ascending and descending stair conditions revealed that peak PFJ stress values were similar. However, the PFJ stress-time integral was more than 1.5 times greater during descending stairs when compared to ascending stairs. As the PFJ stress-time integral represents the stress experienced by the PFJ over the stance phase, a larger integral is indicative of greater overall joint pressure. The greater PFJ stresstime integral during stair descent is consistent with clinical complaints, as patients frequently report greater pain during stair descent compared to stair ascent. Caution must be made in generalizing these findings to the entire PFP population as only ten subjects were evaluated. It is likely that the degree of compensation during stair ambulation may vary depending on the severity of pain complaints and the nature of the injury. In addition, the model used in this study assumed a planar representation of the PFJ for calculation of PFJ stress. While the error associated with such an assumption is not known, any error would be consistent across both groups, making comparisons between populations valid. The models’ use of net joint moments from inverse dynamics equations also removes the ability to isolate
Fig. 7. PFJ stress plotted as a function of the stance phase for both the PFP group and the CTRL group during (A) ascending stairs and (B) descending stairs. *Indicates that the time to peak PFJ stress was significantly earlier in the PFP group when compared with the CTRL group.
122
J.H. Brechter, C.M. Powers / Gait and Posture 16 (2002) 115 /123
any single contributing factor to the net joint moment. One effect of this net value for the joint moment is to underestimate the QF in the presence of hamstring activity or other rotational forces opposing the extensor moment, thus underestimating the magnitude of the PFJRF and the PFJ stress. The MRI assessment protocol also may contribute to some error in the calculation of PFJ stress. Firstly, the MRI technique utilized to assess PFJ contact area was performed with the quadriceps muscle group relaxed and may not be representative of the magnitude of contact area present with quadriceps contraction. To avoid motion artifact in the MRI and obtain a loaded view of the patella, subjects would need to isometrically contract and hold without any motion for the 10:49 min required to obtain the contact area at each joint angle. This was not physically possible. Secondly, the space inside the MRI bore did not allow larger joint angles than 608 of knee flexion, thus extrapolation was required between 608 up to the peak knee flexion range of motion utilized by each subject. The error associated with this extrapolation is not known, however any error would be consistent across both groups, making comparisons between populations valid. The quantity of contact area also has been reported to peak about 608 of knee flexion [13,23,24] thus any extrapolation beyond 608 has likely resulted in an overestimation of the contact area and thus underestimation of the PFJ stress. Further research is required to address both of these limitations.
5. Conclusion Subjects with PFP did not demonstrate increased PFJ stress during stair ascent and descent when compared to a pain free control group. During stair ascent, stress was modulated by a reduction in the PFJRF, which was accomplished through a reduction in the knee extensor moment and a slower cadence. During stair descent, although adopting a similar cadence to that in stair ascent, subjects with PFP demonstrated higher peak knee moment and peak PFJRF, which accounted for the lack of significant differences during stair descent. These findings suggest that individuals with PFP employ compensatory strategies to maintain normal levels of joint stress during stair ambulation.
Acknowledgements This study was funded in part by the Foundation for Physical Therapy.
References [1] Devereaux M, Lachmann S. Patellofemoral arthralgia in athletes attending a sports injury clinic. Britain Journal of Sports Medicine 1984;18:18 /21. [2] Shelton GL, Thigpen LK. Rehabilitation of patellofemoral dysfunction: a review of literature. Journal of Orthopaedic and Sports Physical Therapy 1991;14:243 /9. [3] Wilk KE, Davies GJ, Mangine RE, Malone TR. Patellofemoral disorders: a classification system and clinical guidelines for nonoperative rehabilitation. Journal of Orthopaedic and Sports Physical Therapy 1998;28:307 /22. [4] Andriacchi TPP, Andersson BJM, Fermier RWBS, Stern DD. Galante JOM: a study of lower-limb mechanics during stairclimbing. The Journal of Bone and Joint Surgery 1980;62-A:749 / 57. [5] Asplund DJ, Hall SJ. Kinetics and myoelectric activity during stair-climbing ergometry. Journal of Orthopaedic and Sports Physical Therapy 1995;22:247 /53. [6] Fulkerson JP, Shea KP. Mechanical basis for patellofemoral pain and cartilage breakdown. In: Ewing JW, editor. Articular cartilage and knee joint function: basic science and arthroscopy. New York: Raven Press, 1990:93 /101. [7] Matthews LS, Sonstegard DA, Henke JA. Load bearing characteristics of the patello-femoral joint. Acta Orthopaedica Scandinavia 1977;18:511 /6. [8] Maquet P. Mechanics and osteoarthritis of the patellofemoral joint. Clinical Orthopaedics and Related Research 1979;144:71 /3. [9] Hebert LJ, Gravel D, Arsenault AB, Tremblay G. Patellofemoral pain syndrome: the possible role of an inadequate neuromuscular mechanism. Clinical Biomechanics 1994;9:93 /7. [10] Biloff C. Patellofemoral malalignment. The Journal of Manual and Manipulative Therapy 1995;3:39 /43. [11] Fulkerson JP, Hungerford DS. Disorders of the patellofemoral joint. Baltimore: Williams & Wilkins, 1990. [12] Fujikawa K, Seedhom BB, Wright V. Biomechanics of the patello-femoral joint. Part II: a study of the effect of simulated femoro-tibial varus deformity on the congruity of the patellofemoral compartment and movement of the patella. Engineering in Medicine 1983;12:13 /21. [13] Hefzy MS, Jackson WT, Saddemi SR, Hsieh Y-F. Effects of tibial rotations on patellar tracking and patello-femoral contact areas. Journal of Biomedical Engineering 1992;14:329 /43. [14] Lee TQ, Anzel SH, Bennett KA, Pang D, Kim WC. The influence of fixed rotational deformities of the femur on the patellofemoral contact pressures in human cadaver knees. Clinical Orthopedics 1994;302:69 /74. [15] Powers CM, Heino JG, Rao S, Perry J. The influence of patellofemoral pain on lower limb loading during gait. Clinical Biomechanics 1999;14:722 /8. [16] Powers CM, Landel R, Perry J. Timing and intensity of vastus muscle activity during functional activities in subjects with and without patellofemoral pain. Physical Therapy 1996;76:946 /55. [17] Heino JG, Powers CM, Terk MR, Lee TQ. Measurement of patellofemoral joint contact area using magnetic resonance imaging. Transactions of the Orthopaedic Research Society (Abstract) 1999;45:99. [18] van Eijden TMGJ, Kouwenhoven E, Verburg J, Weijs WA. A mathematical model of the patellofemoral joint. Journal of Biomechanics 1986;19:219 /29. [19] van Eijden TMGJ, de Boer W, Weijs WA. The orientation of the distal part of the quadriceps femoris muscle as a function of the knee flexion-extension angle. Journal of Biomechanics 1985;18:803 /9. [20] Yuen K. The two sample trimmed t for unequal population variances. Biometrika 1974;61:165 /75.
J.H. Brechter, C.M. Powers / Gait and Posture 16 (2002) 115 /123 [21] Winter DA. Kinematic and kinetic patterns in human gait: Variability and compensating effects. Human Movement Science 1984;3:51 /76. [22] Ernst G, Kawaguchi J, Saliba E. Effect of patellar taping on knee kinetics of patients with patellofemoral pain syndrome. Journal of Orthopedic and Sports Physical Therapy(Abstract) 1999;29:661 / 7.
123
[23] Hehne H-JM. Biomechanics of the Patellofemoral Joint and Its Clinical Relevance. Clinical Orthopaedics and Related Research 1990;258:73 /85. [24] Huberti HH, Hayes WC. Patellofemoral contact pressures: the influence of Q-angle and tendofemoral contact. The Journal of Bone and Joint Surgery 1984;16A:715 /23.
Patellofemoral joint stress during stair ascent and descent in persons with and without patellofemoral pain Jacklyn Heino Brechter a,*, Christopher M. Powers b a
b
Department of Physical Therapy, Chapman University, One University Drive, Orange, CA 92866, USA Musculoskeletal Biomechanics Research Laboratory and Department of Biokinesiology and Physical Therapy, University of Southern California, Los Angeles, CA, USA Accepted 30 November 2001
Abstract Objective: To determine if persons with patellofemoral pain (PFP) demonstrate elevated patellofemoral joint (PFJ) stress during stair ascent and descent when compared to persons without PFP. Design: A cross sectional study utilizing an experimental and a control group. Background: Ascending and descending stairs is one of the most painful activities of daily living for persons with PFP. Whether or not the pain associated with stair ambulation is the result of elevated joint stress (force per unit area) has not been explored. Methods: 10 subjects with a diagnosis of PFP and 10 subjects without pain completed two phases of data collection, (1) MRI assessment to determine PFJ contact area and (2) comprehensive motion analysis during stair ambulation at self selected climbing velocities. Data obtained from both data collection sessions were utilized as input variables into a biomechanical model to quantify PFJ stress. Results: Although the knee extensor moment and PFJ reaction force (PFJRF) were significantly reduced in the PFP subjects during stair ascent, there was no difference in PFJ stress between groups. Similarly, there were no differences in PFJ stress during stair descent. Conclusion: Our results do not support the hypothesis that subjects with PFP demonstrate greater joint stress during stair ascent and descent compared to subjects without pain. However, subjects with PFP appeared to maintain normal levels of PFJ stress by minimizing the PFJRF. This was accomplished through a slower cadence and a reduced knee extensor moment. Relevance: PFP is a common syndrome causing pain and functional limitations during stair climbing and other activities requiring high levels of quadriceps activity. Information obtained from this study will be useful in understanding the biomechanical mechanisms contributing to functional deficits in the PFP population. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Patella; Patellofemoral; Biomechanics; Stress; Stairs; Gait
1. Introduction Patellofemoral pain (PFP) affects approximately one of four people in the general population [1]. The clinical presentation of PFP varies between individuals, but some features are considered representative. For example, pain is commonly described as being retropatellar or along the medial and lateral borders of the patella. Symptoms are typically exacerbated with sustained sitting (movie-goers knee) and activities requiring high levels of quadriceps activity (i.e. running, squatting, and negotiating stairs) [2,3]. From a functional standpoint,
* Corresponding author. Tel.: /1-714-744-7649; fax: /1-714-7447621 E-mail address: [email protected] (J.H. Brechter).
ascending and descending stairs is one of the most painful activities of daily living for persons with PFP. It has been reported that stair climbing places a higher demand on the knee when compared to level walking as demonstrated by increased knee extensor moments (an indication of quadriceps demand) and greater ranges of knee motion [4,5]. As the patellofemoral joint reaction force (PFJRF) is dependent on the magnitude of the quadriceps force (QF) and the knee flexion angle [6], the compressive force acting between the patella and femoral trochlea during stair ascent and descent would be expected to be significant. In fact, Matthews et al. have reported that the PFJRF during stair ambulation is more than three times that of level walking [7]. The increased PFJRF associated with stair ambulation suggests that patellofemoral joint (PFJ) stress (force per unit contact area) would be elevated as well.
0966-6362/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 9 6 6 - 6 3 6 2 ( 0 2 ) 0 0 0 9 0 - 5
J.H. Brechter, C.M. Powers / Gait and Posture 16 (2002) 115 /123
116
Elevated PFJ stress in the PFP population is of significant concern as stress is thought to be the factor responsible for articular cartilage degeneration and wear [6,8]. Whether or not individuals with PFP demonstrate higher magnitudes of PFJ stress compared to nonpainful individuals has not been explored. Evidence does exist however, suggesting that this may be the case. Hebert et al. reported that individuals with PFP demonstrated increased knee extensor moments during a squatting activity when compared to control subjects [9]. This finding implies that the PFJRF may be elevated as well. In addition, PFP has been reported to be associated with patellar malalignment [6,10,11] which can result in reduced PFJ contact area [12 /14] and an increase in PFJ stress. Using an imaging based biomechanical model of the PFJ (which takes into consideration the inherent variability in PFJ contact areas between individuals), the purpose of this study was to quantify PFJ stress in persons with and without PFP during stair ascent and descent. It was hypothesized that persons with PFP would demonstrate higher PFJ stress compared to a individuals without PFP. Information obtained from this study may be useful in understanding the biomechanical mechanisms contributing to functional deficits in the PFP population.
activities commonly associated with PFP: (a) stair ascent or descent, (b) squatting, (c) kneeling, (d) prolonged sitting, (e) isometric quadriceps contraction [15,16]. Subjects with PFP were excluded from participation if they reported having any of the following: (1) previous history of knee surgery; (2) history of traumatic patellar dislocation; (3) any neurological involvement that would influence gait; (4) any implanted biological devices, such as pacemakers, cochlear implants, clips which could interact with the magnetic field during imaging. Subjects in the comparison group were recruited from the University of Southern California, and matched for gender to those in the PFP group. Inclusion criterion for participation in the comparison group were as follows: (1) no history or diagnosis of knee pathology or trauma; (2) no knee pain with any of the activities described as inclusion criterion for the PFP group; (3) no limitations present that would influence gait; and (4) no implanted biological devices, such as pacemakers, cochlear implants, clips which could interact with the magnetic field during imaging. Prior to participation, all subjects were fully informed as to the nature of the study, and signed a human subjects consent form approved by the Institutional Review Board of the University of Southern California Health Sciences campus. 2.2. Procedure
2. Methods 2.1. Subjects Twenty subjects were recruited for this study, 10 individuals with a diagnosis of PFP (five males, five females) and ten individuals without PFP (five males, five females) (Table 1). The PFP subjects were recruited from orthopaedic clinics in the Los Angeles area. For purposes of this study, PFP subjects were screened to rule out ligamentous instability, internal derangement, or patellar tendonitis. Subjects were accepted into the study if they met the following inclusion criterion (1) pain originating specifically from the patellofemoral articulation (vague or localized) and (2) reproducible pain with at least two of the following functional
All subjects completed two phases of data collection. Phase one consisted of MRI assessment to determine PFJ contact area while phase two consisted of comprehensive motion analysis during stair ascent and stair descent. Data obtained from both data collection sessions were required as input variables into a biomechanical model to quantify PFJ stress. 2.2.1. Magnetic resonance imaging All imaging was performed at the Los Angeles County/University of Southern California Imaging Science Center. Images of the PFJ were obtained using a 1.5T magnet (GE Medical Systems, Milwaukee, WI) and a three-dimensional spoiled gradient recalled echo (3D SPGR) imaging sequence. The following parameters were employed: TR /60 ms, TE /20 ms, Flip
Table 1 Subject characteristics, means (standard deviations) PFP
Age (years) Height (cm) Weight (kg)
Conrtol
Male (n 5)
Female (n 5)
Combined (n 10)
Male (n 5)
Female (n 5)
Combined (n 10)
38.2 (7.7) 178.9 (10.7) 78.1 (16.0)
36.0 (13.4) 166.1 (6.4) 63.4 (8.4)
37.1 (10.4) 167.9 (17.8) 70.8 (14.3)
32.2 (7.0) 177.7 (10.4) 78.1 (7.9)
31.8 (8.1) 163.7 (7.4) 57.6 (12.3)
32 (7.1) 167.2 (4.4) 67.9 (14.5)
PFP, patellofemoral pain group.
J.H. Brechter, C.M. Powers / Gait and Posture 16 (2002) 115 /123
Angle /308, NEX /1.5, matrix size: 512 /224 /28, field of view: 20 /20 cm2 and chemically selective fat suppression. Each slice was 2-mm thick and contiguous with adjacent slices. Prior to scanning, subjects removed any metal such as jewelry and hair clips. Subjects were positioned supine in the MRI Bore with the knee in 08 of knee flexion. One receive-only extremity coil was secured on each side of the PFJ. Subjects rested quietly while the scan was performed. Following completion of the first scan, subjects were re-positioned and the scan repeated with the knee supported in three additional knee flexion angles 20, 40, and 608. Total imaging time was 44 min. 2.2.2. Motion analysis Motion analysis was performed in the Musculoskeletal Biomechanics Research Laboratory at the University of Southern California. Three-dimensional kinematics were obtained using a six-camera motion analysis system (Vicon, Oxford Metrics Ltd, Oxford, England). Movement was sampled at 60 Hz and recorded digitally on an IBM 166 MHz personal computer. Reflective markers (20 mm spheres) placed at specific anatomical landmarks were used to determine motion of the pelvis and lower extremity. Ground reaction forces were collected at a rate of 2500 Hz using an AMTI force plate (Model #OR6-6-1, Newton, Mass). This force plate was situated as the first of a three-step staircase (step height /20.5 cm, tread /27.5 cm) (Fig. 1). Subjects were appropriately attired to permit marker placement directly on the skin of the subject’s pelvis and
117
lower extremity. The involved lower extremity was instrumented for subjects with PFP, while the right lower extremity was instrumented for subjects without PFP. Anthropometric measures were obtained from each subject for use in calculating lower extremity kinetics using inverse dynamics equations. Reflective markers were then taped to the following landmarks: sacrum, anterior superior iliac spine bilaterally, lateral thigh, lateral femoral epicondyle, lateral tibia, lateral malleolus, 2nd metatarsal head, and posterior calcaneous. Subjects were allowed several practice trials to accommodate to the stair apparatus. All participants were instructed to walk in a step over step fashion at a self selected pace. Two trials for each ascending stairs and descending stairs were obtained for each subject. A trial was considered successful if the subject’s instrumented foot landed within the force plate. All kinematic and kinetic data were collected simultaneously. 2.3. Data analysis 2.3.1. Patellofemoral joint contact area Sequential sagittal plane images of the PFJ were displayed for analysis using Signa Advantage medical imaging software (GE Medical Systems). The section of the image containing the patella and surrounding portion of the femur was enlarged to 2.5 times normal view to enhance visualization of the contact area. Contact was defined as areas of patella and femoral approximation in which no distinct separation could be
Fig. 1. Arrangement of the force plate and portable staircase. This arrangement permitted the force plate to become one of the steps during stair ascent and descent.
J.H. Brechter, C.M. Powers / Gait and Posture 16 (2002) 115 /123
118
found between the cartilage borders of the two structures. Since cartilage is relatively bright on the 3D SPGR images obtained, the definition of contact area was operatively defined as ‘white on white’ [17]. Contact area measurements were made twice and averaged. The line of contact between the patella and femur was measured and recorded using the same software used to display the images. When the line of contact was curved, separate straight-line segments were measured. To obtain the contact area for each slice, the length was multiplied by the 2-mm slice thickness. Contact areas calculated from each image were summed to obtain the total PFJ contact area, with values reported in mm [2]. This method has been shown to be reliable and comparable to contact area obtained using Fuji pressure sensitive film in cadaver specimens [17]. Contact area measurements were made twice and averaged. All MRI measurements were made by the same investigator. 2.3.2. Knee joint kinematics and kinetics Reflective markers were identified manually using Vicon Clinical Manager (VCM) software and then automatically digitized. The same software was used to calculate knee joint kinematics and kinetics in the sagittal plane. Only data corresponding to the force plate step was used. To calculate the net joint moment at the knee, anthropometric data, ground reaction forces, and kinematics were used to solve inverse dynamics equations. To facilitate comparison between subjects and groups, the net knee joint moments were normalized by body mass and reported in units of Nm/kg. Data obtained from the two trials were averaged. 2.3.3. Biomechanical model A PFJ model was developed to utilize input from mechanical analysis of the lower extremity and PFJ contact area from the MRI. An overview of the model is illustrated (Fig. 2). Input variables for the model algorithm included knee joint flexion angle, knee extensor moment, and PFJ contact area. The first step of the algorithm was to calculate the QF. The effective lever arm (LA) for the quadriceps muscle group was determined for each knee joint angle position using an equation fit to the data from van Eijden et al. Eq. (1) [18]. At each joint angle, the knee extensor moment (MEXT) was divided by the lever arm to obtain QF Eq. (2). LA(x) (8:0e 05 x3 0:013x2 0:28x0:046)
(1)
2
x , tibiofemoral joint angle; r /0.99 QF(ui ) MEXT (ui )=QLA (ui )
(2)
u , knee flexion angle; i /1 to x ; x, number of frames of data
Fig. 2. Flow chart of PFJ model. *Data obtained from van Eijden et al. [18]. **Data obtained from van Eijden et al. [19].
The second step of the algorithm was to calculate the PFJRF. A constant (k ) was then calculated for each joint angular position by fitting an equation to the data reported by van Eijden et al. Eq. (3) [19]. This constant was multiplied by the QF to obtain the PFJRF Eq. (4). k(x) (4:62e 01 1:47e 03 x 3:84e 05 x2 )=(1 1:62e 02 x 1:55e 04 x2 6:98e 07 x3 )
(3)
x , tibiofemoral joint angle; r2 /0.99 PFJRF(ui ) k QF(ui )
(4)
k , constant; u , knee flexion angle; i /1 to x ; x , number of frames of data. Using the four contact area values obtained from MRI, the final step in the algorithm was to calculate PFJ stress. A straight line was fit between each two consecutive data points to provide approximate PFJ contact area values for each knee flexion angle from 08 to 608. For knee flexion angles greater than 608, the straight line fit between contact area at 408 and 608 was extrapolated out to the maximum knee flexion angle measured. PFJ stress was then calculated by dividing the PFJRF by the contact area for the knee flexion angle corresponding to the PFJRF value Eq. (5) PFJStress (ui ) PFJRF(ui )=Area(ui )
(5)
Area, contact area; u , knee flexion angle; i /1 to x ; x , number of frames of data. The model output was PFJRF, PFJstress, and utilized contact area (contact area corresponding to a given knee flexion angle). Each variable was normalized to the stance phase of stair ascent or descent.
J.H. Brechter, C.M. Powers / Gait and Posture 16 (2002) 115 /123
119
2.4. Statistical analysis
3.3. Net knee joint moments
Variables used for statistical analysis included cadence, peak PFJstress, PFJ stress-time integral, peak PFJRF, PFJRF-time integral, peak knee extensor moment, peak knee joint flexion and extension angle, and mean utilized contact area. The PFJ stress-time integral and the PFJRF-time integral were calculated by taking the area under the PFJstress curve and the PFJRF curve respectively. The use of the normalized PFJRF and PFJ stress data for calculation of the integrals allowed for easier comparison of the integral data between groups. The mean utilized contact area was calculated by taking the average of the utilized contact area over the stance period. Comparisons between groups were made using trimmed means and Yuen’s test for comparison of two independent groups with unequal variances [20]. This analysis was repeated for each variable and significance levels were set at P B/0.05. All statistics were computed using a custom macro written in Microsoft Excel ’97.
During stair ascent, the PFP group demonstrated a significantly lower peak knee extensor moment when compared to the control group (0.81 vs. 1.16 Nm/kg; P /0.04) (Fig. 4). The same trend was observed during stair descent (0.98 vs. 1.12 Nm/kg), however this difference was not statistically significant (P /0.09) (Fig. 4). 3.4. Patellofemoral joint reaction force During stair ascent, peak PFJRF was significantly less in the PFP group when compared to the control group (25.0 vs. 37.3 N/kg; P /0.02) (Fig. 5). In addition, the PFJRF-time integral, during stair ascent, was significantly less in the PFP group when compared to the control group (288.2 vs. 501.9 N/kg s% Stance Phase; P /0.01) (Fig. 5). In contrast, there was no significant group difference with respect to the peak PFJRF during stair descent. However, there was a trend toward reduced PFJRF-time integral in the PFP group during stair descent (464.4 vs. 605.9 s % Stance Phase; P /0.07) (Fig. 5).
3. Results 3.5. Utilized patellofemoral joint contact area 3.1. Cadence During stair ascent, the PFP group adopted a significantly slower cadence compared with the control group (115.9 vs. 142.1 steps/min; P /0.016). Similarly, the PFP group demonstrated a significantly reduced cadence compared with the control group during stair descent (115.4 vs. 153.6 steps/min; P /0.004).
3.2. Knee kinematics There was no significant difference in knee kinematics between the PFP group and the control group during stair ascent or stair descent (Fig. 3).
There was no significant difference in the mean utilized contact area between the PFP and control groups during stair ascent or stair descent (Fig. 6). 3.6. Patellofemoral joint stress During stair ascent, there was no significant difference in peak PFJ stress between the PFP group and the control group (Fig. 7). Furthermore, there was no significant group difference with respect to the PFJ stress-time integral (Fig. 7). Similarly, during stair descent there was no significant group difference for either peak PFJ or the PFJ stress-time integral (Fig. 7). Although peak PFJ stress was not significantly different between groups, the time at which peak PFJ stress occurred was significantly earlier for the PFP group
Fig. 3. Knee joint angle plotted as a function of the stance phase for both the PFP group and the control group (CTRL) during (A) ascending stairs and (B) descending stairs. There was no significant difference between groups for peak knee flexion or peak knee extension.
120
J.H. Brechter, C.M. Powers / Gait and Posture 16 (2002) 115 /123
Fig. 4. Net knee joint moment plotted as a function of the stance phase for both the PFP group and the CTRL group during (A) ascending stairs and (B) descending stairs. *Indicates that peak extension moment is significantly smaller in the PFP group when compared with the CTRL group.
when compared to the control group (50 vs. 76% Stance Phase; P /0.05) (Fig. 7).
4. Discussion As persons with PFP typically complain of pain during stair ambulation, it was hypothesized that these individuals would demonstrate elevated PFJ stress when compared to individuals without PFP. In other words, the elevated PFJ stress would act to cause the PFP. This hypothesis was not supported by the current study as no significant group differences were found for either peak PFJ stress or PFJ stress-time integral during stair ascent or descent. When evaluating the individual components of PFJ stress, however, group differences were observed. During stair ascent, peak PFJRF and the PFJRF-time integral were significantly reduced in the PFP group, with values being 33 and 43% lower respectively than the control group. Thus, the patients could be trying to avoid PFP by reducing the magnitude of the force across the PFJ. The reduced PFJRF in the PFP group was attributed to a 50% reduction in the peak knee extensor moment as there were no differences in knee kinematics. The lower knee extensor moment and PFJRF are suggestive of quadriceps avoidance, which has been proposed by Powers and colleagues to be a compensatory strategy by which subjects with PFP reduce the
forces acting across the PFJ [16]. The reduced knee extensor moment was likely achieved through a significant reduction in stair climbing cadence as moments have been shown to be influenced by walking speed [21]. Although not measured, the PFP group was observed to employ a forward trunk lean which could have been an additional strategy to reduce the knee extensor moment. Such a strategy, as reported by Ernst et al., would bring the body center of mass closer to the knee joint center subsequently reducing the moment about the knee joint [22]. Despite the fact that the PFJRF was reduced in the PFP group during stair ascent, PFJ stress was similar between groups. This discrepancy can be explained by the trend toward decreased utilized contact area in the PFP group. Although not statistically significant, the observed 14% reduction in contact area helped offset the decrease in stress that would be expected with a reduction in PFJRF (the result being a stress curve comparable to the control group). In contrast to stair ascent, there were no significant differences in the peak PFJRF or the PFJRF-time integral during stair descent. Despite a trend towards decreased knee extensor moment, no significant group differences were found in the knee extensor moment, in spite of the lower cadence in the PFP group (75% of normal). The inability of PFP subjects to reduce the PFJRF during stair descent, as found in stair ascent, may be related to differences in compensation options
Fig. 5. PFJRF plotted as a function of the stance phase for both the PFP group and the CTRL group during (A) ascending stairs and (B) descending stairs. *Indicates that peak PFJRF is significantly smaller in the PFP group when compared with the CTRL group. **Indicates that the PFJRF-time integral is significantly smaller in the PFP group when compared with the CTRL group.
J.H. Brechter, C.M. Powers / Gait and Posture 16 (2002) 115 /123
121
Fig. 6. Utilized PFJ contact area plotted as a function of the stance phase for both the PFP group and the CTRL group during (A) ascending stairs and (B) descending stairs. There was no significant difference between groups.
between these two tasks. The lack of significant differences in the PFJRF during stair descent, as found in stair ascent, may be related to differences in compensation options between these two tasks. For example, subjects with PFP selected a similar cadence during the two tasks (115.9 and 115.4 steps/min respectively for stair ascent and stair descent) yet the peak knee moment during stair ascent was less than 83% of the peak during stair descent. A similar yet larger difference was noted in peak PFJRF, with the peak during stair ascent 76% as large as that during stair descent. Thus, the increase in the knee moment and PFJRF during stair descent was not a function of cadence and must be attributed to other factors. Although the PFJ stress-time integral during stair descent was similar between groups, the pattern of PFJ stress was different, with peak PFJ stress occurring significantly earlier in the stance phase for the PFP group. This early peak in PFJ stress may be explained by the reduced utilized contact area at lesser knee flexion angles in the PFP group. Given similar shaped PFJRF curves, the smaller utilized contact area would result in greater peak PFJ stress during the first 60% of the stair descent stance phase. As the knee flexion angle increased (during the last 40% stair descent stance phase) the PFJ contact area in the PFP group also increased (relative to the control group), resulting in decreased PFJ stress late in the stance phase. The decreased contact area in the
PFP group at lesser knee flexion angles (ie. 0 /308) is consistent with the premise that patellar malalignment and subsequent reduction in contact area would likely be evident before the patella becomes firmly seated in the trochlear groove [13]. Comparison of the ascending and descending stair conditions revealed that peak PFJ stress values were similar. However, the PFJ stress-time integral was more than 1.5 times greater during descending stairs when compared to ascending stairs. As the PFJ stress-time integral represents the stress experienced by the PFJ over the stance phase, a larger integral is indicative of greater overall joint pressure. The greater PFJ stresstime integral during stair descent is consistent with clinical complaints, as patients frequently report greater pain during stair descent compared to stair ascent. Caution must be made in generalizing these findings to the entire PFP population as only ten subjects were evaluated. It is likely that the degree of compensation during stair ambulation may vary depending on the severity of pain complaints and the nature of the injury. In addition, the model used in this study assumed a planar representation of the PFJ for calculation of PFJ stress. While the error associated with such an assumption is not known, any error would be consistent across both groups, making comparisons between populations valid. The models’ use of net joint moments from inverse dynamics equations also removes the ability to isolate
Fig. 7. PFJ stress plotted as a function of the stance phase for both the PFP group and the CTRL group during (A) ascending stairs and (B) descending stairs. *Indicates that the time to peak PFJ stress was significantly earlier in the PFP group when compared with the CTRL group.
122
J.H. Brechter, C.M. Powers / Gait and Posture 16 (2002) 115 /123
any single contributing factor to the net joint moment. One effect of this net value for the joint moment is to underestimate the QF in the presence of hamstring activity or other rotational forces opposing the extensor moment, thus underestimating the magnitude of the PFJRF and the PFJ stress. The MRI assessment protocol also may contribute to some error in the calculation of PFJ stress. Firstly, the MRI technique utilized to assess PFJ contact area was performed with the quadriceps muscle group relaxed and may not be representative of the magnitude of contact area present with quadriceps contraction. To avoid motion artifact in the MRI and obtain a loaded view of the patella, subjects would need to isometrically contract and hold without any motion for the 10:49 min required to obtain the contact area at each joint angle. This was not physically possible. Secondly, the space inside the MRI bore did not allow larger joint angles than 608 of knee flexion, thus extrapolation was required between 608 up to the peak knee flexion range of motion utilized by each subject. The error associated with this extrapolation is not known, however any error would be consistent across both groups, making comparisons between populations valid. The quantity of contact area also has been reported to peak about 608 of knee flexion [13,23,24] thus any extrapolation beyond 608 has likely resulted in an overestimation of the contact area and thus underestimation of the PFJ stress. Further research is required to address both of these limitations.
5. Conclusion Subjects with PFP did not demonstrate increased PFJ stress during stair ascent and descent when compared to a pain free control group. During stair ascent, stress was modulated by a reduction in the PFJRF, which was accomplished through a reduction in the knee extensor moment and a slower cadence. During stair descent, although adopting a similar cadence to that in stair ascent, subjects with PFP demonstrated higher peak knee moment and peak PFJRF, which accounted for the lack of significant differences during stair descent. These findings suggest that individuals with PFP employ compensatory strategies to maintain normal levels of joint stress during stair ambulation.
Acknowledgements This study was funded in part by the Foundation for Physical Therapy.
References [1] Devereaux M, Lachmann S. Patellofemoral arthralgia in athletes attending a sports injury clinic. Britain Journal of Sports Medicine 1984;18:18 /21. [2] Shelton GL, Thigpen LK. Rehabilitation of patellofemoral dysfunction: a review of literature. Journal of Orthopaedic and Sports Physical Therapy 1991;14:243 /9. [3] Wilk KE, Davies GJ, Mangine RE, Malone TR. Patellofemoral disorders: a classification system and clinical guidelines for nonoperative rehabilitation. Journal of Orthopaedic and Sports Physical Therapy 1998;28:307 /22. [4] Andriacchi TPP, Andersson BJM, Fermier RWBS, Stern DD. Galante JOM: a study of lower-limb mechanics during stairclimbing. The Journal of Bone and Joint Surgery 1980;62-A:749 / 57. [5] Asplund DJ, Hall SJ. Kinetics and myoelectric activity during stair-climbing ergometry. Journal of Orthopaedic and Sports Physical Therapy 1995;22:247 /53. [6] Fulkerson JP, Shea KP. Mechanical basis for patellofemoral pain and cartilage breakdown. In: Ewing JW, editor. Articular cartilage and knee joint function: basic science and arthroscopy. New York: Raven Press, 1990:93 /101. [7] Matthews LS, Sonstegard DA, Henke JA. Load bearing characteristics of the patello-femoral joint. Acta Orthopaedica Scandinavia 1977;18:511 /6. [8] Maquet P. Mechanics and osteoarthritis of the patellofemoral joint. Clinical Orthopaedics and Related Research 1979;144:71 /3. [9] Hebert LJ, Gravel D, Arsenault AB, Tremblay G. Patellofemoral pain syndrome: the possible role of an inadequate neuromuscular mechanism. Clinical Biomechanics 1994;9:93 /7. [10] Biloff C. Patellofemoral malalignment. The Journal of Manual and Manipulative Therapy 1995;3:39 /43. [11] Fulkerson JP, Hungerford DS. Disorders of the patellofemoral joint. Baltimore: Williams & Wilkins, 1990. [12] Fujikawa K, Seedhom BB, Wright V. Biomechanics of the patello-femoral joint. Part II: a study of the effect of simulated femoro-tibial varus deformity on the congruity of the patellofemoral compartment and movement of the patella. Engineering in Medicine 1983;12:13 /21. [13] Hefzy MS, Jackson WT, Saddemi SR, Hsieh Y-F. Effects of tibial rotations on patellar tracking and patello-femoral contact areas. Journal of Biomedical Engineering 1992;14:329 /43. [14] Lee TQ, Anzel SH, Bennett KA, Pang D, Kim WC. The influence of fixed rotational deformities of the femur on the patellofemoral contact pressures in human cadaver knees. Clinical Orthopedics 1994;302:69 /74. [15] Powers CM, Heino JG, Rao S, Perry J. The influence of patellofemoral pain on lower limb loading during gait. Clinical Biomechanics 1999;14:722 /8. [16] Powers CM, Landel R, Perry J. Timing and intensity of vastus muscle activity during functional activities in subjects with and without patellofemoral pain. Physical Therapy 1996;76:946 /55. [17] Heino JG, Powers CM, Terk MR, Lee TQ. Measurement of patellofemoral joint contact area using magnetic resonance imaging. Transactions of the Orthopaedic Research Society (Abstract) 1999;45:99. [18] van Eijden TMGJ, Kouwenhoven E, Verburg J, Weijs WA. A mathematical model of the patellofemoral joint. Journal of Biomechanics 1986;19:219 /29. [19] van Eijden TMGJ, de Boer W, Weijs WA. The orientation of the distal part of the quadriceps femoris muscle as a function of the knee flexion-extension angle. Journal of Biomechanics 1985;18:803 /9. [20] Yuen K. The two sample trimmed t for unequal population variances. Biometrika 1974;61:165 /75.
J.H. Brechter, C.M. Powers / Gait and Posture 16 (2002) 115 /123 [21] Winter DA. Kinematic and kinetic patterns in human gait: Variability and compensating effects. Human Movement Science 1984;3:51 /76. [22] Ernst G, Kawaguchi J, Saliba E. Effect of patellar taping on knee kinetics of patients with patellofemoral pain syndrome. Journal of Orthopedic and Sports Physical Therapy(Abstract) 1999;29:661 / 7.
123
[23] Hehne H-JM. Biomechanics of the Patellofemoral Joint and Its Clinical Relevance. Clinical Orthopaedics and Related Research 1990;258:73 /85. [24] Huberti HH, Hayes WC. Patellofemoral contact pressures: the influence of Q-angle and tendofemoral contact. The Journal of Bone and Joint Surgery 1984;16A:715 /23.