- Email: [email protected]
Effects of step length on patellofemoral joint stress in female runners with and without patellofemoral pain
Clinical Biomechanics 29 (2014) 243–247
Contents lists available at ScienceDirect
Clinical Biomechanics journal homepage: www.elsevier.com/locate/clinbiomech
Effects of step length on patellofemoral joint stress in female runners with and without patellofemoral pain John D. Willson a,⁎, Ryan Sharpee b, Stacey A. Meardon a, Thomas W. Kernozek b a b
Department of Physical Therapy, East Carolina University, Greenville, NC 28734, USA La Crosse Institute for Movement Science, Department of Health Professions — Physical Therapy Program, University of Wisconsin — La Crosse, La Crosse, WI 54601, USA
a r t i c l e
i n f o
Article history: Received 22 July 2013 Accepted 23 December 2013 Keywords: Knee Kinetics Gait Rehabilitation
a b s t r a c t Background: Patellofemoral pain is common among runners and is frequently attributed to increased patellofemoral joint stress. The purpose of our study was to examine the effects of changing step length during running on patellofemoral joint stress per step and stress per mile in females with and without patellofemoral pain. Methods: Ten female runners with patellofemoral pain and 13 healthy female runners performed running trials at 3.7 m/s in three conditions: preferred step length, at least +10% step length, and at least −10% step length. Knee flexion angles and internal knee extension moments served as inputs for a biomechanical model to estimate patellofemoral joint stress per step. We also estimated total patellofemoral joint stress per mile based on the number of steps necessary to run a mile during each condition. Findings: Patellofemoral joint stress per step increased 31% in the long step length condition (P b .001) and decreased 22.2% in the short step length condition (P b .001). Despite the inverse relationship between step length and number of steps required to run a mile, patellofemoral joint stress per mile increased 14% in the long step length condition (P b .001) and decreased 7.5% in the short step length condition (P b .001). Interpretation: These results suggest a direct relationship between step length and patellofemoral joint loads. Total stress per mile experienced at the patellofemoral joint decreased with a short step length despite the greater number of steps necessary to cover this distance. These findings may have relevance with respect to both prevention and treatment of patellofemoral joint pain. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction Patellofemoral pain (PFP) is a common injury, particularly among females and runners. It has been determined that roughly 13% of females 18–35 years of age in the general population meet subjective criteria for PFP (Roush and Curtis, 2012). Females in the military were found to be twice as likely to develop new cases of PFP compared to males (Boling et al., 2009). Finally, PFP was found to be the most common orthopedic condition among runners seeking care for their injuries (Taunton et al., 2002). Patellofemoral joint stress (PFJS) is commonly associated with the etiology and exacerbation of patellofemoral joint symptoms (Dye, 2005; Powers, 2003). Greater PFJS during squatting, walking, and running was observed among people with PFP compared with those without PFP (Brechter and Powers, 2002; Farrokhi et al., 2011; Wirtz et al., 2012). Repetitive exposure to this elevated PFJS during running may result in pain due to increased subchondral bone metabolic activity (Draper ⁎ Corresponding author at: Department of Physical Therapy, East Carolina University, 600 Moye Blvd, Greenville, NC 27834, USA. E-mail address: [email protected] (J.D. Willson). 0268-0033/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.clinbiomech.2013.12.016
et al., 2012) or elevated patellar water content (Ho et al., 2013). If increased PFJS is associated with PFP, it follows from these studies that interventions to reduce PFJS during running may benefit both prevention and treatment efforts. Modification of step length during running may result in favorable changes in PFJS. A 10% decrease in step length during running resulted in a 34% decrease in negative tibiofemoral joint work per step during treadmill running (Heiderscheit et al., 2011). A reduction in patellofemoral joint reaction force was also reported among runners without PFP while running with a decreased step length compared to their preferred step length (Lenhart et al., 2013). However, the effect of step length changes during running on PFJS has not been reported. Additionally, it is not clear if the increased number of steps required to run a given distance would result in a similar cumulative PFJS that may effectively mitigate these potential benefits at the patellofemoral joint. The purpose of this study was to test if step length changes during running affect total PFJS per step and cumulative PFJS per mile. We hypothesized that PFJS per step and PFJS per mile would decrease while running with a shorter step length and increase while running with a greater step length. We also tested if the effect of step length
244
J.D. Willson et al. / Clinical Biomechanics 29 (2014) 243–247
changes during running were similar among females with and without PFP. We hypothesized that the effect of step length changes would be similar among females with and without PFP. 2. Methods Our protocol was approved by the university institutional review board and all subjects provided their informed consent prior to participation. Using an alpha level of 0.05, a beta level of 0.2, and the expected variability PFJS integral per step during running determined from pilot testing, at least 17 participants were deemed to be necessary to identify differences between step length conditions with effect sizes greater than 0.6. Participants who were 18–35 years old, ran at least 10 miles per week and reported their activity level as greater than or equal to 5 out of 10 on the Tegner activity scale (Tegner and Lysholm, 1985) were recruited for this study. All subjects who were pregnant, reported a known cardiovascular pathology, had surgery to either lower extremity over the last 12 months or sustained a traumatic injury to either knee joint within 6 months of the study were excluded. All participants reported use of a rearfoot strike pattern during their routine training runs. Potential participants with a primary complaint of knee pain during running were screened by a licensed physical therapist for specific criteria to be included in the PFP group (Brechter and Powers, 2002; Farrokhi et al., 2011; Wirtz et al., 2012). These criteria included a verbal pain score of at least a 3 (moderate) on a 10 point verbal pain scale during running and squatting, prolonged sitting, ascending or descending stairs, or jumping. Potential participants must have also described pain behind or adjacent to the patella and not solely at the iliotibial band, patellar tendon, or knee-joint line. Knee symptoms were required to be of insidious onset and present for at least 2 months in duration. Participants with PFP had to report that their symptoms were exacerbated with manual compression of the patella into the trochlear groove with the knee in 15° of flexion or with palpation of the medial or lateral patellar retinaculum against the posterior patellar surface. Finally, participants with PFP were required to score less than 85/100 on the Anterior Knee Pain Scale, a patient-reported questionnaire developed to evaluate symptoms and functional limitations associated with patellofemoral joint disorders (Kujala et al., 1993). Any potential participants were excluded if they presented for these screening tests with signs and symptoms of a meniscus or ligament injury, were currently receiving supervised treatment for PFP, or reported symptoms in either foot, ankle, hip or low back that were exacerbated by running. Screening of 18 potential participants with complaints of knee pain resulted in the inclusion of 10 females with PFP for further testing. For these participants, the most symptomatic lower extremity was chosen for analysis. Thirteen females of similar age, height, weight, and running experience but no recent injuries or symptoms during running also participated, resulting in 23 total participants (Table 1). The right lower extremity of the healthy control group subjects was used for analysis. All participants wore the same type of shoe (model 629, New Balance, Boston, MA) during testing in order to reduce variability that may be caused by different shoe absorption properties. Subjects were prepared for 3D motion analysis testing during running by attaching reflective markers (diameter = 14 mm) to the right limb. The three-dimensional coordinates of these markers were used to track the motion of the pelvis, femur, shank, and foot,
each modeled as a rigid body. Anatomical markers used to establish the segmental-coordinate systems were placed over each iliac crest, the greater trochanters, medial and lateral femoral condyles, medial and lateral proximal tibia, medial and lateral malleoli, the first and fifth metatarsal heads, and the tip of the shoe. The proximal end of the thigh segment was described by the ipsilateral greater trochanter marker and the calculated location of the hip joint center. The hip joint center was identified using a Newton iterative spherical fitting algorithm from data recorded during a standing trial where the instrumented leg was moved in a prescribed fashion prior to the running trials (Hicks and Richards, 2005). During this trial, participants stood on their contralateral leg while moving their free leg through two arcs of approximately 80° hip flexion and two arcs of 50° hip abduction. Following both the standing calibration trial and hip center movement trial, each of the anatomical markers were removed. Tracking markers, which remained in place for all of the running trials, were positioned as a cluster of three markers on the rearfoot of the shoe, a cluster of four markers on the posterior shank, a cluster of four markers on the lateral thigh, and three markers for the pelvis on each anterior superior iliac spine and at the L5–S1 interspace. For all running trials, participants were asked to run at 3.7 m/s (± 5%) along a 23 m runway as indicated by the forward velocity of the sacral marker in the lab coordinate system at the moment of contact with the force platform. Subjects initially ran with their preferred step length, which was calculated immediately after each running trial by taking the difference in location of the distal heel marker between subsequent steps of the right and left legs at initial contact. Subjects were then asked to run with a step length at least 10% greater and 10% less than their preferred step length (in random order) at the same running velocity. Feedback on step length was provided to the participants following each trial in the modified step length conditions. Trials were discarded and repeated during the modified step length conditions if step length increased or decreased less than 10% from the preferred condition. Thus, only trials in which participants modified step length by more than 10% were used for analysis. Participants were instructed to maintain their preferred foot strike pattern in each condition. After at least five practice trials, five trials were collected for further analysis in each condition. During each trial, marker data were collected at 120 Hz using an eight camera motion capture system (Motion Analysis Corporation, Santa Rosa, CA, USA) positioned around the runway. The marker trajectories were digitally filtered at 15 Hz using a low pass, fourth order Butterworth recursive filter. Ground reaction forces were recorded at 1200 Hz from a force platform (Model 4080, Bertec Corporation, Columbus, OH) flush with the surface of the runway and digitally filtered at 15 Hz using a low pass, fourth order Butterworth recursive filter for the calculation of net joint moments using the inverse dynamics approach (Bisseling and Hof, 2006; Kristianslund et al., 2012). Sagittal plane knee joint angle and net internal joint moment during the stance phase of each running trial were calculated with Visual 3D software (C-Motion Inc., Rockville, MD). Kinematic data were not normalized to the static neutral trial. In other words, 0° corresponded to an erect posture at the hip and knee. PFJS during running was calculated by dividing estimated patellofemoral joint reaction force by patellofemoral contact area. Patellofemoral joint reaction force was estimated using a biomechanical model previously described by Salem and Powers (Salem and Powers,
Table 1 Mean (SD) participant demographics.
All participants Healthy Patellofemoral pain
Participants (n)
Age (yr)
Height (m)
Weight (kg)
Miles/week
23 13 10
20.9 (2.9) 21.0 (2.3) 20.8 (3.7)
1.69 (.05) 1.70 (.05) 1.69 (.05)
61.8 (7.0) 61.2 (6.0) 62.5 (8.5)
19.9 (11.1) 18.3 (9.7) 21.9 (13.0)
J.D. Willson et al. / Clinical Biomechanics 29 (2014) 243–247
2001). This model has been used to estimate PFJS during walking, resisted squatting, stair climbing, and running in previous studies (Brechter and Powers, 2002; Kulmala et al., 2013; Wallace et al., 2002; Wirtz et al., 2012). This model uses the sagittal plane tibiofemoral angle and net knee joint moment data obtained via inverse dynamics. The internal knee extensor moment data during the stance phase of each running trial (in Nm) was divided by the effective lever arm for the quadriceps as a function of knee flexion angle to obtain quadriceps force expressed relative to body mass. The effective lever arm of the quadriceps is based on cadaver data presented by van Eijden et al. (1986) as represented by the equation from Salem and Powers (2001). Patellofemoral joint reaction force was then calculated by multiplying calculated quadriceps force by the ratio of patellofemoral compression force to quadriceps force as a function of knee flexion angle presented by van Eijden et al. (1986) and represented mathematically by Salem and Powers (2001). Patellofemoral contact area during weight bearing may differ between females with and without PFP during weight bearing activities (Connolly et al., 2009). Thus, the estimated patellofemoral contact area as a function of knee flexion angle was unique for each group based on data presented by Connolly et al (2009). In their study, patellofemoral contact area was determined for males and females at 15°, 30°, and 45° knee flexion using magnetic resonance imaging as subjects stood with 45% body weight through each leg. These data were then linearly interpolated to provide patellofemoral contact area as a function of knee flexion angle for each group. Patellofemoral joint reaction force was divided by this group-specific patellofemoral joint contact area to estimate PFJS during running. Dependent variables of interest relative to the PFJS model included peak knee flexion angle, peak knee extension moment, peak PFJS and the PFJS-time integral during the stance phase of running. The average of these variables of interest from all five running trials in each condition was calculated and used for analysis. Changing a runner's step length will affect the number of steps required to run a given distance. Therefore, we also calculated total PFJS stress per mile during running by multiplying the PFJS-time integral during a single stance phase by the number of steps required to run a mile, based on the step length each participant demonstrated during each condition. The effect of step length on the dependent variables of interest were analyzed using separate two-factor mixed ANOVAs (2 [group] × 3 [step length condition]) and polynomial contrasts. We set an alpha level of .01 for these tests to maintain a family-wise type I error rate of no more than 5%. Significant step length condition effects were analyzed in post hoc tests using the LSD approach. Polynomial contrasts were used to identify significant linear trends across step length conditions for dependent variables with significant main effects of step length. Significant group × step length interaction terms were analyzed using independent t-tests to compare changes between groups between step length conditions. Each dependent variable was first tested against a normal distribution using separate Kolmognorov– Smirnov tests.
3. Results All participants successfully manipulated step length for each condition. Average step length increased 14.6% and stance time increased 8.7% during the increased step length condition compared to the preferred condition (Table 2). Conversely, average step length decreased 16.1% and stance time decreased 7% during the reduced step length condition compared to the preferred condition. Participants demonstrated both greater peak knee flexion and peak knee extensor moment during the increased step length condition (Table 3, P b .001). Although the knee extensor moment decreased between each step length condition for both females with and without PFP, the decrease between the preferred and reduced step length conditions was less among females with PFP(P = .002).
245
Table 2 Mean (SD) step length and stance time measured for all participants during each condition while running at 3.7 m/s. Variable Step length (m) All participants Healthy Patellofemoral pain Stance time (s) All participants Healthy Patellofemoral pain
10% increased
Preferred
10% reduced
1.675 (.077) 1.673 (.077) 1.677 (.081)
1.461 (.058) 1.459 (.063) 1.463 (.054)
1.226 (.071) 1.205 (.077) 1.254 (.054)
0.250 (.023) 0.252 (.026) 0.247 (.019)
0.230 (.020) 0.229 (.021) 0.232 (.020)
0.214 (.021) 0.213 (.023) 0.214 (.019)
A longer step length resulted in a 9.2% increase in peak PFJS for both groups (Table 3, P b .001) and a reduced step length resulted in a 16.3% average decrease in peak PFJS (P b .001). However, a significant group × step length interaction was identified for peak PFJS. Followup tests revealed that the 8.7% change in peak PFJS among participants with PFP between the preferred and reduced step length conditions was smaller than the 21.6% decrease noted among healthy participants (P = .005). Main effects of step length condition were identified for PFJS-time integral per step (Table 3, P b .001) and PFJS-time integral per mile (P b .001). Analysis of polynomial contrasts revealed a linear trend for both PFJS-time integral per step and PFJS-time integral per mile across step length conditions (P b 0.001). Follow up tests indicated that, compared with their preferred step length, PFJS-time integral per step increased 31% when participants ran with an increased step length (P b .001) and decreased 22.2% when participants ran with a reduced step length (Fig. 1, P b .001). Additionally, despite the inverse relationship between step length and number of steps required to run a given distance, PFJS-time integral per mile increased 14.2% in the increased step length condition (P b .001) and decreased 7.5% in the reduced step length condition (P b .001). No significant group main effects or group × step length condition interaction terms were identified for PFJS-time integral per step or PFJS-time integral per mile.
4. Discussion The purpose of this study was to test the hypotheses that step length changes result in a proportional change PFJS-time integral per step and PFJS-time integral per mile. We also tested if changes in step length result in similar changes in peak PFJS, PFJS-time integral, and PFJS-time integral per mile among females with and without PFP. In support of our hypotheses, we observed that step length has a significant effect on patellofemoral joint stress in both females with and without PFP. We also observed that these step length effects per mile were not entirely mitigated by taking fewer steps per mile with at least a 10% increase in step length and a greater number of steps per mile with at least a 10% decrease in step length. The pathoetiology of PFP is commonly attributed to increased PFJS during weight bearing activities (Farrokhi et al., 2011; Heino Brechter and Powers, 2002). While it is not possible to measure PFJS directly in vivo, physiological events that occur in response to increased PFJS such as increased patellar water content and patellar bone metabolic activity support this premise. Ho et al. (2013) reported that both PFP and patellar water content increased among females with PFP over the course of a 40 minute running session. The increase in patellar water content measured among females with PFP in this study was twice as large as the increase observed among pain-free females in an earlier study (Ho et al., 2013). Draper et al. (2012) reported a significant association between greater patellar and trochlear bone metabolic activity and pain intensity among males and females with PFP. To the extent that increased PFJS accounts for these physiological changes at the
246
J.D. Willson et al. / Clinical Biomechanics 29 (2014) 243–247
Table 3 Mean (SD) gait measurements among participants while running in each step length condition. Variable Peak knee flexion (degrees) All participants Healthy Patellofemoral pain Peak knee extensor moment (Nm/kg ∗ m) All participants Healthy Patellofemoral pain Peak PFJ stress (MPa) All participants Healthy PFPS PFJ stress–time integral/step (MPa ∗ s) All participants Healthy Patellofemoral pain PFJ Stress-time integral/mile (MPa ∗ s) All participants Healthy Patellofemoral pain a b
10% increased
Preferred
10% reduced
P-value
43.6 (6.2) 44.4 (6.3) 42.5 (6.4)
40.0 (5.0) 40.5 (5.2) 39.4 (5.0)
37.4 (6.8) 37.6 (6.8) 37.1 (7.2)
Group × condition: Group: Condition:
.58 .65 b.001b
1.49 (.21) 1.54 (.19) 1.43 (.21)
1.40 (.20) 1.48 (.17) 1.30 (.20)
1.19 (.24) 1.17 (.25) 1.20 (.24)
Group × condition: Group: Condition:
.003a .29 b.001b
10.7 (2.4) 10.9 (1.9) 10.4 (2.9)
9.8 (1.9) 10.2 (1.4) 9.3 (2.3)
8.2 (2.4) 8.0 (2.2) 8.5 (2.6)
Group × condition: Group: Condition:
.01a .72 b.001b
1.18 (.37) 1.23 (.34) 1.12 (.42)
0.90 (.26) 0.94 (.22) 0.85 (.31)
0.70 (.29) 0.70 (.31) 0.71 (.29)
Group × condition: Group: Condition:
.11 .62 b.001b
563 (171) 590 (164) 529 (181)
493 (139) 516 (121) 463 (160)
456 (182) 459 (183) 453 (190)
Group × condition: Group: Condition:
.15 .56 b.001b
Greater decrease among healthy group between preferred and 10% reduced step length conditions. 10% increased N preferred N 10% reduced.
patellofemoral joint, our results may have implications for the prevention or treatment of PFP in runners. The results of our study are consistent with earlier studies of step length effects on knee running mechanics. A 34% decrease in negative tibiofemoral joint work per step and a 25% decrease in positive tibiofemoral joint work has been reported among healthy runners in response to a 10% reduction in step length (Heiderscheit et al., 2011). This same study identified a 47% increase in negative tibiofemoral joint work and a 22% increase in positive tibiofemoral joint work per step following a 10% increase in step length (Heiderscheit et al., 2011). A 15% increase in stride length during running also resulted in a 12% increase in shock attenuation, while a 15% decrease in stride length resulted in a 22% decrease in shock attenuation (Mercer et al., 2003). Since the quadriceps muscles contribute to the greater knee extensor moment used to increase knee joint work or attenuate shock at impact while running with an increased step length, the model used in this study would yield an increase in PFJS. The changes in PFJS-time integral per step observed in our study with different step lengths can be attributed to several factors. Participants in this study and in previous studies of step length and running mechanics demonstrated increased peak knee flexion and peak
Fig. 1. Average patellofemoral joint stress for each step length condition (n = 23). Shaded areas represent 1 standard error of the mean for each condition.
knee extensor moment as step length increased (Derrick et al., 1998; Heiderscheit et al., 2011). As knee flexion angle increases, patellofemoral contact area also increases (Besier et al., 2005). This would typically disperse patellofemoral joint reaction forces resulting in a decrease PFJS. However, the increase in knee extensor moment was greater than the increase in estimated patellofemoral contact area for the 10% increase in step length. This relatively large increase in knee extensor moment with the 10% increase in step length may be attributed to a longer flight time, increased downward velocity of the COM at initial contact, and greater peak ground reaction forces (Heiderscheit et al., 2011; Mercer et al., 2003). In our study as in previous studies, a longer step length was associated with a longer stance phase (Derrick et al., 1998; Heiderscheit et al., 2011; Mercer et al., 2003). The combined effects of greater PFJS and a longer stance phase both contributed to the greater PFJS-time integral for the 10% increase in step length. The results of our study support our hypothesis that the sparing effect per step at the PFJ with a 10% decreased step length persisted despite the greater number of steps required to run a given distance. Conversely, potentially detrimental effects per step with increased step length were not eliminated due to a smaller number of steps required to run a given distance. Running with a shorter step length has also been found to reduce the probability of tibial stress fractures, despite the increased number of steps required to run a given distance.(Edwards et al., 2009) However, these conclusions are based on the assumption that modifications to step length measured for five individual steps are maintained throughout the duration of the running session. Further study appears necessary to document that step length changes are feasible among runners with and without PFP and that these changes can result in decreased PFJS over multiple consecutive steps during a prolonged run. Although no data were collected as part of our study to quantify the effectiveness of step length training as an intervention for PFP, step length training may be a viable treatment option for runners with PFP. Gait training interventions used to modify running mechanics among individuals with PFP suggest that runners can make persistent changes in technique with sufficient practice, even after the removal of the training stimulus (Cheung and Davis, 2011; Willy et al., 2012). If PFJS exacerbates symptoms of PFP, training programs that promote a shorter step length may lessen symptoms of PFP and result in improved tolerance for their preferred mode of exercise (Wille et al., 2013). Further, as the etiology of PFP is often described in the context of increased PFJS, (Dye, 2005; Powers, 2003) the increase in PFJS-time integral per mile with
J.D. Willson et al. / Clinical Biomechanics 29 (2014) 243–247
increasing step length observed in our study also raises the question of the feasibility and effectiveness of training programs that reduce step length as a preventative measure for the development of PFP. A training program to decrease step length for healthy runners demonstrating an increased step length was effective for reducing preferred step length, but it is not known to what effect, if any, this had on injury risk (Morgan et al., 1994). Future studies of the effectiveness of step length training for prevention or treatment of PFP appear justified. The metabolic consequence of step length modifications was not measured in this study. However, habitual runners tend to utilize a step length that minimizes oxygen consumption (Cavanagh and Williams, 1982; Hamill et al., 1995) and large modifications to step length during running may adversely affect running economy. For example, a 20% decrease in step length was found to typically increase oxygen consumption up to 13% (Cavanagh and Williams, 1982). The 16% average step length decrease in this study may therefore adversely affect running economy. However, we believe the metabolic cost associated with this step length decrease may be a worthwhile sacrifice for recreational runners with PFP, provided a reduction in symptoms accompanies this change. Notable limitations are associated with this study. The PFJS model used in this study includes estimates of patellofemoral contact area and quadriceps moment arm that are not specific to our participants. The model is also not sensitive to cocontraction of the knee flexors or transverse and frontal plane hip and knee joint rotations. Each of these factors may significantly influence PFJS. As such, these data are best estimates of PFJS during running. This study was also not powered to identify group × step length interaction terms with a small effect size. A post hoc power analysis suggests that our sample size was sufficient to identify group × step length interaction terms with a standard mean difference greater than 0.7. Interaction terms that were not identified due to inadequate study power may have limited clinical relevance. Finally, we tested the effects of a single increase and decrease in step length during running. As such, questions remain regarding the effect of larger (or smaller) modifications to step length on PFJS estimates during running. Previous step length interventions during running range from 6 to 30% (Hobara et al., 2012; Morgan et al., 1994). Future studies with a broader range of step length changes are necessary to develop a greater understanding of the relationship between step length and PFJS during running. 5. Conclusion In our study we observed a significant increase in peak PFJS and PFJStime integral when running with at least a 10% increased step length and a significant decrease in these same variables when running with at least a 10% reduced step length. The PFJS loading effects calculated per step were not completely mitigated by the inverse relationship between step length and number of steps per mile. To the extent that PFJS is associated with the etiology or exacerbation of PFP, step length training may be a viable treatment option for individuals with PFP or as a preventative measure for healthy runners. References Besier, T.F., Draper, C.E., Gold, G.E., Beaupre, G.S., Delp, S.L., 2005. Patellofemoral joint contact area increases with knee flexion and weight-bearing. J. Orthop. Res. 23, 345–350.
247
Bisseling, R.W., Hof, A.L., 2006. Handling of impact forces in inverse dynamics. J. Biomech. 39, 2438–2444. Boling, M., Padua, D., Marshall, S., Guskiewicz, K., Pyne, S., Beutler, A., 2009. Gender differences in the incidence and prevalence of patellofemoral pain syndrome. Scand. J. Med. Sci. Sports 37, 2108–2116. Brechter, J.H., Powers, C.M., 2002. Patellofemoral joint stress during stair ascent and descent in persons with and without patellofemoral pain. Gait Posture 16, 115–123. Cavanagh, P.R., Williams, K.R., 1982. The effect of stride length variation on oxygen uptake during distance running. Med. Sci. Sports Exerc. 14, 30–35. Cheung, R.T., Davis, I.S., 2011. Landing pattern modification to improve patellofemoral pain in runners: a case series. J. Orthop. Sports Phys. Ther. 41, 914–919. Connolly, K.D., Ronsky, J.L., Westover, L.M., Kupper, J.C., Frayne, R., 2009. Differences in patellofemoral contact mechanics associated with patellofemoral pain syndrome. J. Biomech. 42, 2802–2807. Derrick, T.R., Hamill, J., Caldwell, G.E., 1998. Energy absorption of impacts during running at various stride lengths. Med. Sci. Sports Exerc. 30, 128–135. Draper, C.E., Fredericson, M., Gold, G.E., Besier, T.F., Delp, S.L., Beaupre, G.S., et al., 2012. Patients with patellofemoral pain exhibit elevated bone metabolic activity at the patellofemoral joint. J. Orthop. Res. 30, 209–213. Dye, S.F., 2005. The pathophysiology of patellofemoral pain: a tissue homeostasis perspective. Clin. Orthop. Relat. Res. 100–110. Edwards, W.B., Taylor, D., Rudolphi, T.J., Gillette, J.C., Derrick, T.R., 2009. Effects of stride length and running mileage on a probabilistic stress fracture model. Med. Sci. Sports Exerc. 41, 2177–2184. Farrokhi, S., Keyak, J.H., Powers, C.M., 2011. Individuals with patellofemoral pain exhibit greater patellofemoral joint stress: a finite element analysis study. Osteoarthritis Cartilage 19, 287–294. Hamill, J., Derrick, T.R., Holt, K.G., 1995. Shock attenuation and stride frequency during running. Hum. Mov. Sci. 14, 45–60. Heiderscheit, B.C., Chumanov, E.S., Michalski, M.P., Wille, C.M., Ryan, M.B., 2011. Effects of step rate manipulation on joint mechanics during running. Med. Sci. Sports Exerc. 43, 296–302. Hicks, J.L., Richards, J.G., 2005. Clinical applicability of using spherical fitting to find hip joint centers. Gait Posture 22, 138–145. Ho, K.Y., Hu, H.H., Colletti, P.M., Powers, C.M., 2013. Running-induced patellofemoral pain fluctuates with changes in patella water content. Eur. J. Sport Sci. http://dx.doi.org/ 10.1080/17461391.2013.862872. Hobara, H., Sato, T., Sakaguchi, M., Nakazawa, K., 2012. Step frequency and lower extremity loading during running. Int. J. Sports Med. 33, 310–313. Kristianslund, E., Krosshaug, T., van den Bogert, A.J., 2012. Effect of low pass filtering on joint moments from inverse dynamics: implications for injury prevention. J. Biomech. 45, 666–671. Kujala, U.M., Jaakkola, L.H., Koskinen, S.K., Taimela, S., Hurme, M., Nelimarkka, O., 1993. Scoring of patellofemoral disorders. Arthroscopy 9, 159–163. Kulmala, J.P., Avela, J., Pasanen, K., Parkkari, J., 2013. Forefoot strikers exhibit lower running-induced knee loading than rearfoot strikers. Med. Sci. Sports Exerc. 45, 2306–2313. Lenhart, R.L., Thelen, D.G., Wille, C.M., Chumanov, E.S., Heiderscheit, B.C., 2013. Increasing running step rate reduces patellofemoral joint forces. Med. Sci. Sports Exerc. http:// dx.doi.org/10.1249/MSS.0b013e3182a78c3a. Mercer, J.A., Devita, P., Derrick, T.R., Bates, B.T., 2003. Individual effects of stride length and frequency on shock attenuation during running. Med. Sci. Sports Exerc. 35, 307–313. Morgan, D., Martin, P., Craib, M., Caruso, C., Clifton, R., Hopewell, R., 1994. Effect of step length optimization on the aerobic demand of running. J. Appl. Physiol. 77, 245–251. Powers, C.M., 2003. The influence of altered lower-extremity kinematics on patellofemoral joint dysfunction: a theoretical perspective. J. Orthop. Sports Phys. Ther. 33, 639–646. Roush, J.R., Curtis, Bay R., 2012. Prevalence of anterior knee pain in 18–35 year-old females. Int. J. Sports Phys. Ther. 7, 396–401. Salem, G.J., Powers, C.M., 2001. Patellofemoral joint kinetics during squatting in collegiate women athletes. Clin. Biomech. 16, 424–430. Taunton, J., Ryan, M., Clement, D., McKenzie, D., Lloyd-Smith, D., Zumbo, B., 2002. A retrospective case-control analysis is 2002 running injuries. Br. J. Sports Med. 36, 95–101. Tegner, Y., Lysholm, J., 1985. Rating systems in the evaluation of knee ligament injuries. Clin. Orthop. 43–49. van Eijden, T., Kouwenhoven, E., Verburg, J., Weijs, W., 1986. A mathematical model of the patellofemoral joint. J. Biomech. 19, 219–229. Wallace, D.A., Salem, G.J., Salinas, R., Powers, C.M., 2002. Patellofemoral joint kinetics while squatting with and without an external load. J. Orthop. Sports Phys. Ther. 32, 141–148. Wille, C., Chumanov, E., Schubert, A., Kempf, J., Heiderscheit, B., 2013. Running step rate modification to reduce anterior knee pain in runners. J. Orthop. Sports Phys. Ther. 43, A119–A120. Willy, R.W., Scholz, J.P., Davis, I.S., 2012. Mirror gait retraining for the treatment of patellofemoral pain in female runners. Clin. Biomech. 27, 1045–1051. Wirtz, A.D., Willson, J.D., Kernozek, T.W., Hong, D.A., 2012. Patellofemoral joint stress during running in females with and without patellofemoral pain. Knee 19, 703–708.
Contents lists available at ScienceDirect
Clinical Biomechanics journal homepage: www.elsevier.com/locate/clinbiomech
Effects of step length on patellofemoral joint stress in female runners with and without patellofemoral pain John D. Willson a,⁎, Ryan Sharpee b, Stacey A. Meardon a, Thomas W. Kernozek b a b
Department of Physical Therapy, East Carolina University, Greenville, NC 28734, USA La Crosse Institute for Movement Science, Department of Health Professions — Physical Therapy Program, University of Wisconsin — La Crosse, La Crosse, WI 54601, USA
a r t i c l e
i n f o
Article history: Received 22 July 2013 Accepted 23 December 2013 Keywords: Knee Kinetics Gait Rehabilitation
a b s t r a c t Background: Patellofemoral pain is common among runners and is frequently attributed to increased patellofemoral joint stress. The purpose of our study was to examine the effects of changing step length during running on patellofemoral joint stress per step and stress per mile in females with and without patellofemoral pain. Methods: Ten female runners with patellofemoral pain and 13 healthy female runners performed running trials at 3.7 m/s in three conditions: preferred step length, at least +10% step length, and at least −10% step length. Knee flexion angles and internal knee extension moments served as inputs for a biomechanical model to estimate patellofemoral joint stress per step. We also estimated total patellofemoral joint stress per mile based on the number of steps necessary to run a mile during each condition. Findings: Patellofemoral joint stress per step increased 31% in the long step length condition (P b .001) and decreased 22.2% in the short step length condition (P b .001). Despite the inverse relationship between step length and number of steps required to run a mile, patellofemoral joint stress per mile increased 14% in the long step length condition (P b .001) and decreased 7.5% in the short step length condition (P b .001). Interpretation: These results suggest a direct relationship between step length and patellofemoral joint loads. Total stress per mile experienced at the patellofemoral joint decreased with a short step length despite the greater number of steps necessary to cover this distance. These findings may have relevance with respect to both prevention and treatment of patellofemoral joint pain. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction Patellofemoral pain (PFP) is a common injury, particularly among females and runners. It has been determined that roughly 13% of females 18–35 years of age in the general population meet subjective criteria for PFP (Roush and Curtis, 2012). Females in the military were found to be twice as likely to develop new cases of PFP compared to males (Boling et al., 2009). Finally, PFP was found to be the most common orthopedic condition among runners seeking care for their injuries (Taunton et al., 2002). Patellofemoral joint stress (PFJS) is commonly associated with the etiology and exacerbation of patellofemoral joint symptoms (Dye, 2005; Powers, 2003). Greater PFJS during squatting, walking, and running was observed among people with PFP compared with those without PFP (Brechter and Powers, 2002; Farrokhi et al., 2011; Wirtz et al., 2012). Repetitive exposure to this elevated PFJS during running may result in pain due to increased subchondral bone metabolic activity (Draper ⁎ Corresponding author at: Department of Physical Therapy, East Carolina University, 600 Moye Blvd, Greenville, NC 27834, USA. E-mail address: [email protected] (J.D. Willson). 0268-0033/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.clinbiomech.2013.12.016
et al., 2012) or elevated patellar water content (Ho et al., 2013). If increased PFJS is associated with PFP, it follows from these studies that interventions to reduce PFJS during running may benefit both prevention and treatment efforts. Modification of step length during running may result in favorable changes in PFJS. A 10% decrease in step length during running resulted in a 34% decrease in negative tibiofemoral joint work per step during treadmill running (Heiderscheit et al., 2011). A reduction in patellofemoral joint reaction force was also reported among runners without PFP while running with a decreased step length compared to their preferred step length (Lenhart et al., 2013). However, the effect of step length changes during running on PFJS has not been reported. Additionally, it is not clear if the increased number of steps required to run a given distance would result in a similar cumulative PFJS that may effectively mitigate these potential benefits at the patellofemoral joint. The purpose of this study was to test if step length changes during running affect total PFJS per step and cumulative PFJS per mile. We hypothesized that PFJS per step and PFJS per mile would decrease while running with a shorter step length and increase while running with a greater step length. We also tested if the effect of step length
244
J.D. Willson et al. / Clinical Biomechanics 29 (2014) 243–247
changes during running were similar among females with and without PFP. We hypothesized that the effect of step length changes would be similar among females with and without PFP. 2. Methods Our protocol was approved by the university institutional review board and all subjects provided their informed consent prior to participation. Using an alpha level of 0.05, a beta level of 0.2, and the expected variability PFJS integral per step during running determined from pilot testing, at least 17 participants were deemed to be necessary to identify differences between step length conditions with effect sizes greater than 0.6. Participants who were 18–35 years old, ran at least 10 miles per week and reported their activity level as greater than or equal to 5 out of 10 on the Tegner activity scale (Tegner and Lysholm, 1985) were recruited for this study. All subjects who were pregnant, reported a known cardiovascular pathology, had surgery to either lower extremity over the last 12 months or sustained a traumatic injury to either knee joint within 6 months of the study were excluded. All participants reported use of a rearfoot strike pattern during their routine training runs. Potential participants with a primary complaint of knee pain during running were screened by a licensed physical therapist for specific criteria to be included in the PFP group (Brechter and Powers, 2002; Farrokhi et al., 2011; Wirtz et al., 2012). These criteria included a verbal pain score of at least a 3 (moderate) on a 10 point verbal pain scale during running and squatting, prolonged sitting, ascending or descending stairs, or jumping. Potential participants must have also described pain behind or adjacent to the patella and not solely at the iliotibial band, patellar tendon, or knee-joint line. Knee symptoms were required to be of insidious onset and present for at least 2 months in duration. Participants with PFP had to report that their symptoms were exacerbated with manual compression of the patella into the trochlear groove with the knee in 15° of flexion or with palpation of the medial or lateral patellar retinaculum against the posterior patellar surface. Finally, participants with PFP were required to score less than 85/100 on the Anterior Knee Pain Scale, a patient-reported questionnaire developed to evaluate symptoms and functional limitations associated with patellofemoral joint disorders (Kujala et al., 1993). Any potential participants were excluded if they presented for these screening tests with signs and symptoms of a meniscus or ligament injury, were currently receiving supervised treatment for PFP, or reported symptoms in either foot, ankle, hip or low back that were exacerbated by running. Screening of 18 potential participants with complaints of knee pain resulted in the inclusion of 10 females with PFP for further testing. For these participants, the most symptomatic lower extremity was chosen for analysis. Thirteen females of similar age, height, weight, and running experience but no recent injuries or symptoms during running also participated, resulting in 23 total participants (Table 1). The right lower extremity of the healthy control group subjects was used for analysis. All participants wore the same type of shoe (model 629, New Balance, Boston, MA) during testing in order to reduce variability that may be caused by different shoe absorption properties. Subjects were prepared for 3D motion analysis testing during running by attaching reflective markers (diameter = 14 mm) to the right limb. The three-dimensional coordinates of these markers were used to track the motion of the pelvis, femur, shank, and foot,
each modeled as a rigid body. Anatomical markers used to establish the segmental-coordinate systems were placed over each iliac crest, the greater trochanters, medial and lateral femoral condyles, medial and lateral proximal tibia, medial and lateral malleoli, the first and fifth metatarsal heads, and the tip of the shoe. The proximal end of the thigh segment was described by the ipsilateral greater trochanter marker and the calculated location of the hip joint center. The hip joint center was identified using a Newton iterative spherical fitting algorithm from data recorded during a standing trial where the instrumented leg was moved in a prescribed fashion prior to the running trials (Hicks and Richards, 2005). During this trial, participants stood on their contralateral leg while moving their free leg through two arcs of approximately 80° hip flexion and two arcs of 50° hip abduction. Following both the standing calibration trial and hip center movement trial, each of the anatomical markers were removed. Tracking markers, which remained in place for all of the running trials, were positioned as a cluster of three markers on the rearfoot of the shoe, a cluster of four markers on the posterior shank, a cluster of four markers on the lateral thigh, and three markers for the pelvis on each anterior superior iliac spine and at the L5–S1 interspace. For all running trials, participants were asked to run at 3.7 m/s (± 5%) along a 23 m runway as indicated by the forward velocity of the sacral marker in the lab coordinate system at the moment of contact with the force platform. Subjects initially ran with their preferred step length, which was calculated immediately after each running trial by taking the difference in location of the distal heel marker between subsequent steps of the right and left legs at initial contact. Subjects were then asked to run with a step length at least 10% greater and 10% less than their preferred step length (in random order) at the same running velocity. Feedback on step length was provided to the participants following each trial in the modified step length conditions. Trials were discarded and repeated during the modified step length conditions if step length increased or decreased less than 10% from the preferred condition. Thus, only trials in which participants modified step length by more than 10% were used for analysis. Participants were instructed to maintain their preferred foot strike pattern in each condition. After at least five practice trials, five trials were collected for further analysis in each condition. During each trial, marker data were collected at 120 Hz using an eight camera motion capture system (Motion Analysis Corporation, Santa Rosa, CA, USA) positioned around the runway. The marker trajectories were digitally filtered at 15 Hz using a low pass, fourth order Butterworth recursive filter. Ground reaction forces were recorded at 1200 Hz from a force platform (Model 4080, Bertec Corporation, Columbus, OH) flush with the surface of the runway and digitally filtered at 15 Hz using a low pass, fourth order Butterworth recursive filter for the calculation of net joint moments using the inverse dynamics approach (Bisseling and Hof, 2006; Kristianslund et al., 2012). Sagittal plane knee joint angle and net internal joint moment during the stance phase of each running trial were calculated with Visual 3D software (C-Motion Inc., Rockville, MD). Kinematic data were not normalized to the static neutral trial. In other words, 0° corresponded to an erect posture at the hip and knee. PFJS during running was calculated by dividing estimated patellofemoral joint reaction force by patellofemoral contact area. Patellofemoral joint reaction force was estimated using a biomechanical model previously described by Salem and Powers (Salem and Powers,
Table 1 Mean (SD) participant demographics.
All participants Healthy Patellofemoral pain
Participants (n)
Age (yr)
Height (m)
Weight (kg)
Miles/week
23 13 10
20.9 (2.9) 21.0 (2.3) 20.8 (3.7)
1.69 (.05) 1.70 (.05) 1.69 (.05)
61.8 (7.0) 61.2 (6.0) 62.5 (8.5)
19.9 (11.1) 18.3 (9.7) 21.9 (13.0)
J.D. Willson et al. / Clinical Biomechanics 29 (2014) 243–247
2001). This model has been used to estimate PFJS during walking, resisted squatting, stair climbing, and running in previous studies (Brechter and Powers, 2002; Kulmala et al., 2013; Wallace et al., 2002; Wirtz et al., 2012). This model uses the sagittal plane tibiofemoral angle and net knee joint moment data obtained via inverse dynamics. The internal knee extensor moment data during the stance phase of each running trial (in Nm) was divided by the effective lever arm for the quadriceps as a function of knee flexion angle to obtain quadriceps force expressed relative to body mass. The effective lever arm of the quadriceps is based on cadaver data presented by van Eijden et al. (1986) as represented by the equation from Salem and Powers (2001). Patellofemoral joint reaction force was then calculated by multiplying calculated quadriceps force by the ratio of patellofemoral compression force to quadriceps force as a function of knee flexion angle presented by van Eijden et al. (1986) and represented mathematically by Salem and Powers (2001). Patellofemoral contact area during weight bearing may differ between females with and without PFP during weight bearing activities (Connolly et al., 2009). Thus, the estimated patellofemoral contact area as a function of knee flexion angle was unique for each group based on data presented by Connolly et al (2009). In their study, patellofemoral contact area was determined for males and females at 15°, 30°, and 45° knee flexion using magnetic resonance imaging as subjects stood with 45% body weight through each leg. These data were then linearly interpolated to provide patellofemoral contact area as a function of knee flexion angle for each group. Patellofemoral joint reaction force was divided by this group-specific patellofemoral joint contact area to estimate PFJS during running. Dependent variables of interest relative to the PFJS model included peak knee flexion angle, peak knee extension moment, peak PFJS and the PFJS-time integral during the stance phase of running. The average of these variables of interest from all five running trials in each condition was calculated and used for analysis. Changing a runner's step length will affect the number of steps required to run a given distance. Therefore, we also calculated total PFJS stress per mile during running by multiplying the PFJS-time integral during a single stance phase by the number of steps required to run a mile, based on the step length each participant demonstrated during each condition. The effect of step length on the dependent variables of interest were analyzed using separate two-factor mixed ANOVAs (2 [group] × 3 [step length condition]) and polynomial contrasts. We set an alpha level of .01 for these tests to maintain a family-wise type I error rate of no more than 5%. Significant step length condition effects were analyzed in post hoc tests using the LSD approach. Polynomial contrasts were used to identify significant linear trends across step length conditions for dependent variables with significant main effects of step length. Significant group × step length interaction terms were analyzed using independent t-tests to compare changes between groups between step length conditions. Each dependent variable was first tested against a normal distribution using separate Kolmognorov– Smirnov tests.
3. Results All participants successfully manipulated step length for each condition. Average step length increased 14.6% and stance time increased 8.7% during the increased step length condition compared to the preferred condition (Table 2). Conversely, average step length decreased 16.1% and stance time decreased 7% during the reduced step length condition compared to the preferred condition. Participants demonstrated both greater peak knee flexion and peak knee extensor moment during the increased step length condition (Table 3, P b .001). Although the knee extensor moment decreased between each step length condition for both females with and without PFP, the decrease between the preferred and reduced step length conditions was less among females with PFP(P = .002).
245
Table 2 Mean (SD) step length and stance time measured for all participants during each condition while running at 3.7 m/s. Variable Step length (m) All participants Healthy Patellofemoral pain Stance time (s) All participants Healthy Patellofemoral pain
10% increased
Preferred
10% reduced
1.675 (.077) 1.673 (.077) 1.677 (.081)
1.461 (.058) 1.459 (.063) 1.463 (.054)
1.226 (.071) 1.205 (.077) 1.254 (.054)
0.250 (.023) 0.252 (.026) 0.247 (.019)
0.230 (.020) 0.229 (.021) 0.232 (.020)
0.214 (.021) 0.213 (.023) 0.214 (.019)
A longer step length resulted in a 9.2% increase in peak PFJS for both groups (Table 3, P b .001) and a reduced step length resulted in a 16.3% average decrease in peak PFJS (P b .001). However, a significant group × step length interaction was identified for peak PFJS. Followup tests revealed that the 8.7% change in peak PFJS among participants with PFP between the preferred and reduced step length conditions was smaller than the 21.6% decrease noted among healthy participants (P = .005). Main effects of step length condition were identified for PFJS-time integral per step (Table 3, P b .001) and PFJS-time integral per mile (P b .001). Analysis of polynomial contrasts revealed a linear trend for both PFJS-time integral per step and PFJS-time integral per mile across step length conditions (P b 0.001). Follow up tests indicated that, compared with their preferred step length, PFJS-time integral per step increased 31% when participants ran with an increased step length (P b .001) and decreased 22.2% when participants ran with a reduced step length (Fig. 1, P b .001). Additionally, despite the inverse relationship between step length and number of steps required to run a given distance, PFJS-time integral per mile increased 14.2% in the increased step length condition (P b .001) and decreased 7.5% in the reduced step length condition (P b .001). No significant group main effects or group × step length condition interaction terms were identified for PFJS-time integral per step or PFJS-time integral per mile.
4. Discussion The purpose of this study was to test the hypotheses that step length changes result in a proportional change PFJS-time integral per step and PFJS-time integral per mile. We also tested if changes in step length result in similar changes in peak PFJS, PFJS-time integral, and PFJS-time integral per mile among females with and without PFP. In support of our hypotheses, we observed that step length has a significant effect on patellofemoral joint stress in both females with and without PFP. We also observed that these step length effects per mile were not entirely mitigated by taking fewer steps per mile with at least a 10% increase in step length and a greater number of steps per mile with at least a 10% decrease in step length. The pathoetiology of PFP is commonly attributed to increased PFJS during weight bearing activities (Farrokhi et al., 2011; Heino Brechter and Powers, 2002). While it is not possible to measure PFJS directly in vivo, physiological events that occur in response to increased PFJS such as increased patellar water content and patellar bone metabolic activity support this premise. Ho et al. (2013) reported that both PFP and patellar water content increased among females with PFP over the course of a 40 minute running session. The increase in patellar water content measured among females with PFP in this study was twice as large as the increase observed among pain-free females in an earlier study (Ho et al., 2013). Draper et al. (2012) reported a significant association between greater patellar and trochlear bone metabolic activity and pain intensity among males and females with PFP. To the extent that increased PFJS accounts for these physiological changes at the
246
J.D. Willson et al. / Clinical Biomechanics 29 (2014) 243–247
Table 3 Mean (SD) gait measurements among participants while running in each step length condition. Variable Peak knee flexion (degrees) All participants Healthy Patellofemoral pain Peak knee extensor moment (Nm/kg ∗ m) All participants Healthy Patellofemoral pain Peak PFJ stress (MPa) All participants Healthy PFPS PFJ stress–time integral/step (MPa ∗ s) All participants Healthy Patellofemoral pain PFJ Stress-time integral/mile (MPa ∗ s) All participants Healthy Patellofemoral pain a b
10% increased
Preferred
10% reduced
P-value
43.6 (6.2) 44.4 (6.3) 42.5 (6.4)
40.0 (5.0) 40.5 (5.2) 39.4 (5.0)
37.4 (6.8) 37.6 (6.8) 37.1 (7.2)
Group × condition: Group: Condition:
.58 .65 b.001b
1.49 (.21) 1.54 (.19) 1.43 (.21)
1.40 (.20) 1.48 (.17) 1.30 (.20)
1.19 (.24) 1.17 (.25) 1.20 (.24)
Group × condition: Group: Condition:
.003a .29 b.001b
10.7 (2.4) 10.9 (1.9) 10.4 (2.9)
9.8 (1.9) 10.2 (1.4) 9.3 (2.3)
8.2 (2.4) 8.0 (2.2) 8.5 (2.6)
Group × condition: Group: Condition:
.01a .72 b.001b
1.18 (.37) 1.23 (.34) 1.12 (.42)
0.90 (.26) 0.94 (.22) 0.85 (.31)
0.70 (.29) 0.70 (.31) 0.71 (.29)
Group × condition: Group: Condition:
.11 .62 b.001b
563 (171) 590 (164) 529 (181)
493 (139) 516 (121) 463 (160)
456 (182) 459 (183) 453 (190)
Group × condition: Group: Condition:
.15 .56 b.001b
Greater decrease among healthy group between preferred and 10% reduced step length conditions. 10% increased N preferred N 10% reduced.
patellofemoral joint, our results may have implications for the prevention or treatment of PFP in runners. The results of our study are consistent with earlier studies of step length effects on knee running mechanics. A 34% decrease in negative tibiofemoral joint work per step and a 25% decrease in positive tibiofemoral joint work has been reported among healthy runners in response to a 10% reduction in step length (Heiderscheit et al., 2011). This same study identified a 47% increase in negative tibiofemoral joint work and a 22% increase in positive tibiofemoral joint work per step following a 10% increase in step length (Heiderscheit et al., 2011). A 15% increase in stride length during running also resulted in a 12% increase in shock attenuation, while a 15% decrease in stride length resulted in a 22% decrease in shock attenuation (Mercer et al., 2003). Since the quadriceps muscles contribute to the greater knee extensor moment used to increase knee joint work or attenuate shock at impact while running with an increased step length, the model used in this study would yield an increase in PFJS. The changes in PFJS-time integral per step observed in our study with different step lengths can be attributed to several factors. Participants in this study and in previous studies of step length and running mechanics demonstrated increased peak knee flexion and peak
Fig. 1. Average patellofemoral joint stress for each step length condition (n = 23). Shaded areas represent 1 standard error of the mean for each condition.
knee extensor moment as step length increased (Derrick et al., 1998; Heiderscheit et al., 2011). As knee flexion angle increases, patellofemoral contact area also increases (Besier et al., 2005). This would typically disperse patellofemoral joint reaction forces resulting in a decrease PFJS. However, the increase in knee extensor moment was greater than the increase in estimated patellofemoral contact area for the 10% increase in step length. This relatively large increase in knee extensor moment with the 10% increase in step length may be attributed to a longer flight time, increased downward velocity of the COM at initial contact, and greater peak ground reaction forces (Heiderscheit et al., 2011; Mercer et al., 2003). In our study as in previous studies, a longer step length was associated with a longer stance phase (Derrick et al., 1998; Heiderscheit et al., 2011; Mercer et al., 2003). The combined effects of greater PFJS and a longer stance phase both contributed to the greater PFJS-time integral for the 10% increase in step length. The results of our study support our hypothesis that the sparing effect per step at the PFJ with a 10% decreased step length persisted despite the greater number of steps required to run a given distance. Conversely, potentially detrimental effects per step with increased step length were not eliminated due to a smaller number of steps required to run a given distance. Running with a shorter step length has also been found to reduce the probability of tibial stress fractures, despite the increased number of steps required to run a given distance.(Edwards et al., 2009) However, these conclusions are based on the assumption that modifications to step length measured for five individual steps are maintained throughout the duration of the running session. Further study appears necessary to document that step length changes are feasible among runners with and without PFP and that these changes can result in decreased PFJS over multiple consecutive steps during a prolonged run. Although no data were collected as part of our study to quantify the effectiveness of step length training as an intervention for PFP, step length training may be a viable treatment option for runners with PFP. Gait training interventions used to modify running mechanics among individuals with PFP suggest that runners can make persistent changes in technique with sufficient practice, even after the removal of the training stimulus (Cheung and Davis, 2011; Willy et al., 2012). If PFJS exacerbates symptoms of PFP, training programs that promote a shorter step length may lessen symptoms of PFP and result in improved tolerance for their preferred mode of exercise (Wille et al., 2013). Further, as the etiology of PFP is often described in the context of increased PFJS, (Dye, 2005; Powers, 2003) the increase in PFJS-time integral per mile with
J.D. Willson et al. / Clinical Biomechanics 29 (2014) 243–247
increasing step length observed in our study also raises the question of the feasibility and effectiveness of training programs that reduce step length as a preventative measure for the development of PFP. A training program to decrease step length for healthy runners demonstrating an increased step length was effective for reducing preferred step length, but it is not known to what effect, if any, this had on injury risk (Morgan et al., 1994). Future studies of the effectiveness of step length training for prevention or treatment of PFP appear justified. The metabolic consequence of step length modifications was not measured in this study. However, habitual runners tend to utilize a step length that minimizes oxygen consumption (Cavanagh and Williams, 1982; Hamill et al., 1995) and large modifications to step length during running may adversely affect running economy. For example, a 20% decrease in step length was found to typically increase oxygen consumption up to 13% (Cavanagh and Williams, 1982). The 16% average step length decrease in this study may therefore adversely affect running economy. However, we believe the metabolic cost associated with this step length decrease may be a worthwhile sacrifice for recreational runners with PFP, provided a reduction in symptoms accompanies this change. Notable limitations are associated with this study. The PFJS model used in this study includes estimates of patellofemoral contact area and quadriceps moment arm that are not specific to our participants. The model is also not sensitive to cocontraction of the knee flexors or transverse and frontal plane hip and knee joint rotations. Each of these factors may significantly influence PFJS. As such, these data are best estimates of PFJS during running. This study was also not powered to identify group × step length interaction terms with a small effect size. A post hoc power analysis suggests that our sample size was sufficient to identify group × step length interaction terms with a standard mean difference greater than 0.7. Interaction terms that were not identified due to inadequate study power may have limited clinical relevance. Finally, we tested the effects of a single increase and decrease in step length during running. As such, questions remain regarding the effect of larger (or smaller) modifications to step length on PFJS estimates during running. Previous step length interventions during running range from 6 to 30% (Hobara et al., 2012; Morgan et al., 1994). Future studies with a broader range of step length changes are necessary to develop a greater understanding of the relationship between step length and PFJS during running. 5. Conclusion In our study we observed a significant increase in peak PFJS and PFJStime integral when running with at least a 10% increased step length and a significant decrease in these same variables when running with at least a 10% reduced step length. The PFJS loading effects calculated per step were not completely mitigated by the inverse relationship between step length and number of steps per mile. To the extent that PFJS is associated with the etiology or exacerbation of PFP, step length training may be a viable treatment option for individuals with PFP or as a preventative measure for healthy runners. References Besier, T.F., Draper, C.E., Gold, G.E., Beaupre, G.S., Delp, S.L., 2005. Patellofemoral joint contact area increases with knee flexion and weight-bearing. J. Orthop. Res. 23, 345–350.
247
Bisseling, R.W., Hof, A.L., 2006. Handling of impact forces in inverse dynamics. J. Biomech. 39, 2438–2444. Boling, M., Padua, D., Marshall, S., Guskiewicz, K., Pyne, S., Beutler, A., 2009. Gender differences in the incidence and prevalence of patellofemoral pain syndrome. Scand. J. Med. Sci. Sports 37, 2108–2116. Brechter, J.H., Powers, C.M., 2002. Patellofemoral joint stress during stair ascent and descent in persons with and without patellofemoral pain. Gait Posture 16, 115–123. Cavanagh, P.R., Williams, K.R., 1982. The effect of stride length variation on oxygen uptake during distance running. Med. Sci. Sports Exerc. 14, 30–35. Cheung, R.T., Davis, I.S., 2011. Landing pattern modification to improve patellofemoral pain in runners: a case series. J. Orthop. Sports Phys. Ther. 41, 914–919. Connolly, K.D., Ronsky, J.L., Westover, L.M., Kupper, J.C., Frayne, R., 2009. Differences in patellofemoral contact mechanics associated with patellofemoral pain syndrome. J. Biomech. 42, 2802–2807. Derrick, T.R., Hamill, J., Caldwell, G.E., 1998. Energy absorption of impacts during running at various stride lengths. Med. Sci. Sports Exerc. 30, 128–135. Draper, C.E., Fredericson, M., Gold, G.E., Besier, T.F., Delp, S.L., Beaupre, G.S., et al., 2012. Patients with patellofemoral pain exhibit elevated bone metabolic activity at the patellofemoral joint. J. Orthop. Res. 30, 209–213. Dye, S.F., 2005. The pathophysiology of patellofemoral pain: a tissue homeostasis perspective. Clin. Orthop. Relat. Res. 100–110. Edwards, W.B., Taylor, D., Rudolphi, T.J., Gillette, J.C., Derrick, T.R., 2009. Effects of stride length and running mileage on a probabilistic stress fracture model. Med. Sci. Sports Exerc. 41, 2177–2184. Farrokhi, S., Keyak, J.H., Powers, C.M., 2011. Individuals with patellofemoral pain exhibit greater patellofemoral joint stress: a finite element analysis study. Osteoarthritis Cartilage 19, 287–294. Hamill, J., Derrick, T.R., Holt, K.G., 1995. Shock attenuation and stride frequency during running. Hum. Mov. Sci. 14, 45–60. Heiderscheit, B.C., Chumanov, E.S., Michalski, M.P., Wille, C.M., Ryan, M.B., 2011. Effects of step rate manipulation on joint mechanics during running. Med. Sci. Sports Exerc. 43, 296–302. Hicks, J.L., Richards, J.G., 2005. Clinical applicability of using spherical fitting to find hip joint centers. Gait Posture 22, 138–145. Ho, K.Y., Hu, H.H., Colletti, P.M., Powers, C.M., 2013. Running-induced patellofemoral pain fluctuates with changes in patella water content. Eur. J. Sport Sci. http://dx.doi.org/ 10.1080/17461391.2013.862872. Hobara, H., Sato, T., Sakaguchi, M., Nakazawa, K., 2012. Step frequency and lower extremity loading during running. Int. J. Sports Med. 33, 310–313. Kristianslund, E., Krosshaug, T., van den Bogert, A.J., 2012. Effect of low pass filtering on joint moments from inverse dynamics: implications for injury prevention. J. Biomech. 45, 666–671. Kujala, U.M., Jaakkola, L.H., Koskinen, S.K., Taimela, S., Hurme, M., Nelimarkka, O., 1993. Scoring of patellofemoral disorders. Arthroscopy 9, 159–163. Kulmala, J.P., Avela, J., Pasanen, K., Parkkari, J., 2013. Forefoot strikers exhibit lower running-induced knee loading than rearfoot strikers. Med. Sci. Sports Exerc. 45, 2306–2313. Lenhart, R.L., Thelen, D.G., Wille, C.M., Chumanov, E.S., Heiderscheit, B.C., 2013. Increasing running step rate reduces patellofemoral joint forces. Med. Sci. Sports Exerc. http:// dx.doi.org/10.1249/MSS.0b013e3182a78c3a. Mercer, J.A., Devita, P., Derrick, T.R., Bates, B.T., 2003. Individual effects of stride length and frequency on shock attenuation during running. Med. Sci. Sports Exerc. 35, 307–313. Morgan, D., Martin, P., Craib, M., Caruso, C., Clifton, R., Hopewell, R., 1994. Effect of step length optimization on the aerobic demand of running. J. Appl. Physiol. 77, 245–251. Powers, C.M., 2003. The influence of altered lower-extremity kinematics on patellofemoral joint dysfunction: a theoretical perspective. J. Orthop. Sports Phys. Ther. 33, 639–646. Roush, J.R., Curtis, Bay R., 2012. Prevalence of anterior knee pain in 18–35 year-old females. Int. J. Sports Phys. Ther. 7, 396–401. Salem, G.J., Powers, C.M., 2001. Patellofemoral joint kinetics during squatting in collegiate women athletes. Clin. Biomech. 16, 424–430. Taunton, J., Ryan, M., Clement, D., McKenzie, D., Lloyd-Smith, D., Zumbo, B., 2002. A retrospective case-control analysis is 2002 running injuries. Br. J. Sports Med. 36, 95–101. Tegner, Y., Lysholm, J., 1985. Rating systems in the evaluation of knee ligament injuries. Clin. Orthop. 43–49. van Eijden, T., Kouwenhoven, E., Verburg, J., Weijs, W., 1986. A mathematical model of the patellofemoral joint. J. Biomech. 19, 219–229. Wallace, D.A., Salem, G.J., Salinas, R., Powers, C.M., 2002. Patellofemoral joint kinetics while squatting with and without an external load. J. Orthop. Sports Phys. Ther. 32, 141–148. Wille, C., Chumanov, E., Schubert, A., Kempf, J., Heiderscheit, B., 2013. Running step rate modification to reduce anterior knee pain in runners. J. Orthop. Sports Phys. Ther. 43, A119–A120. Willy, R.W., Scholz, J.P., Davis, I.S., 2012. Mirror gait retraining for the treatment of patellofemoral pain in female runners. Clin. Biomech. 27, 1045–1051. Wirtz, A.D., Willson, J.D., Kernozek, T.W., Hong, D.A., 2012. Patellofemoral joint stress during running in females with and without patellofemoral pain. Knee 19, 703–708.