CONSISTENCY OF DYNAMIC KNEE VALGUS KINEMATICS AND PAIN ACROSS FUNCTIONAL TASKS IN FEMALES WITH PATELLOFEMORAL PAIN: A CROSS-SECTIONAL STUDY

Abstract

Background and Purpose:

Dynamic knee valgus has been associated with patellofemoral pain (PFP) during high-level tasks, however, repeated lower-level stresses may be an alternative pain mechanism. The primary purpose of the current study was to examine the consistency of dynamic knee valgus and task-elicited pain demonstrated by females with PFP across four common functional tasks (stair ascent, stair descent, sit-to-stand, and stand-to-sit). A secondary purpose was to assess the correlation between the clinical test of single-limb squat and functional tasks.

Hypothesis:

Females with patellofemoral pain will demonstrate a positive relationship in magnitude of dynamic knee valgus and task-elicited pain across functional tasks. Individuals who demonstrated greater dynamic knee valgus and task-elicited pain during the clinical test of single-limb squat would demonstrate greater dynamic knee valgus and task elicited pain during stair ascent/descent and sit-to-stand/stand-to-sit tasks.

Study Design:

Cross-sectional study; secondary analysis of a feasibility intervention study.

Methods:

Twenty-three women with patellofemoral pain (age: 21.8 SD 3.7 years; BMI: 22.2 SD 2.0 kg/m²) participated. Three-dimensional kinematic data were captured during task completion. Hip and knee frontal and transverse plane angles at 45 ° of knee flexion, and pain using a visual analog scale, were assessed during single-limb squat, stair ascent/descent, and sit-to-stand. Pearson product-moment correlation coefficients were calculated to examine between-task relationships for each variable at the pre-intervention assessment.

Results:

Correlation coefficients between tasks ranged from 0.23-0.76 for hip frontal plane measures (7/10 significant relationships, p<0.02), 0.31-0.90 for hip transverse plane measures (7/10 significant, p<0.01), 0.87-0.95 for knee frontal plane measures (10/10 significant, p<0.01), and 0.54-0.86 for knee transverse plane measures (10/10 significant, p<0.01). Correlations spanned 0.59-0.85 for pain during tasks (10/10 significant, p<0.01).

Conclusion:

Females with patellofemoral pain demonstrated positive correlations in dynamic knee valgus kinematics and task-elicited pain across five tasks. Movement and pain during the clinical test of single-limb squat test also was correlated with movement and pain during the functional tasks of stair ascent/descent and sit-to-stand.

Level of Evidence:

Level 2b.

INTRODUCTION

Dynamic knee valgus has been described as a combination of hip adduction, hip medial rotation, knee abduction, and knee lateral rotation.^1,2 Dynamic knee valgus or its components have been identified as movement impairment risk factors for the development of a variety of musculoskeletal pain problems.^1,3-5 Often, dynamic knee valgus is studied in an athletic population, examining whether it is a contributing factor to pain problems in different sports and different knee pain conditions.^4,6-9 Assessment frequently occurs during clinical tests or sports-related tasks that may more closely mimic the demands placed on the knee during sports or other higher intensity activities. Examples include single-limb squat,² running,¹⁰ or vertical drop jump.^11-13 The information gathered from these tests is valuable, particularly given dynamic knee valgus demonstrated during such tasks is considered a risk factor for an acute injury, such as an anterior cruciate ligament tear, that occurs during sport activity.^14,15

Dynamic knee valgus, however, also is thought to be a potential contributing factor to insidious or chronic knee pain conditions such as patellofemoral pain (PFP).¹⁶ Furthermore, chronic pain conditions, such as PFP, exist in elite athletes, recreational athletes, and non-athletes.^17-19 Patellofemoral pain also has been reported in lower-level activities of daily living such as prolonged sitting, squatting, and stair negotiation.^17,20

It has been hypothesized that the cumulative effects of movement impairments performed throughout the day during functional tasks may contribute to the development or persistence of chronic musculoskeletal pain problems.^21,22 If dynamic knee valgus is present throughout the day during routine functional tasks, the repetitive dynamic knee valgus may result in increased accumulation of stress at the patellofemoral joint, contributing to the development of PFP. This stress may compound the stresses already present due to participation in sports related activities, and lead to pain during a variety of low-intensity, functional tasks.

While dynamic knee valgus has been associated with PFP during tasks such as the vertical drop jump and single limb squat, the assessment of dynamic knee valgus in previous studies has typically compared individuals with PFP to individuals without PFP and assessed only one or two tasks at a time. What is unknown is whether individuals demonstrate a similar movement pattern of dynamic knee valgus across a variety of functional tasks such as stair ascent/descent, sit-to-stand or stand-to-sit performed throughout the day. It also is unknown whether individuals report a similar pain pattern across tasks. For example, does a patient who reports higher pain during one task (e.g. stair descent) report higher pain in other tasks (e.g. stand-to-sit). Knowledge of the consistency of dynamic knee valgus and task-elicited pain across common daily tasks would provide insight into the movement-related mechanisms of pain development and inform potential movement-based interventions for women with PFP.

The primary purpose of the current study was to examine the consistency of dynamic knee valgus and task-elicited pain demonstrated by females with PFP across four common functional tasks (stair ascent, stair descent, sit-to-stand, and stand-to-sit). The authors hypothesized that the dynamic knee valgus kinematics demonstrated during one task would be correlated with the dynamic knee valgus kinematics demonstrated during all other tasks. Likewise, pain during one task was expected to be positively correlated with pain during all other tasks.

A secondary purpose was to assess the correlation between the clinical test of single-limb squat and functional tasks. The authors hypothesized that individuals who demonstrated greater dynamic knee valgus and task-elicited pain during the clinical test of single-limb squat would demonstrate greater dynamic knee valgus and task elicited pain during stair ascent/descent and sit-to-stand/stand-to-sit tasks. A better understanding of how well clinical tests relate to functional tasks with respect to movement and task-elicited pain may provide clinicians with the confidence to use clinical tests in place of functional tasks when time, space, or equipment restraints limit assessment of a variety of tasks.

METHODS

These data are a subset of data from a prospective, within-group, double-baseline, feasibility intervention trial.²⁰ Data reported in this cross-sectional study are from pre-treatment assessment.

Participants

Twenty-three women with PFP (mean age: 21.8 SD 3.7 years; body mass index: 22.2 SD 2.0 kg/m²) participated. Twenty participants reported bilateral pain; three participants reported unilateral pain. Participants qualified for the study if they reported pain behind or around the patella for at least two months duration²³ and a minimum average pain during the prior week of a 3/10 using a verbal pain rating scale (0 represents no pain, 10 represents severe pain). Pain had to be reproducible with two of the three following tests: resisted isometric quadriceps contraction performed with the knee in approximately 10 ° of flexion, single-limb squat, or stair descent.²⁴ Women with PFP also had to demonstrate an observable dynamic knee valgus during the descent phase of a single-limb squat test.²⁵ Dynamic knee valgus was considered observable if the researcher visually observed the participant demonstrate a change of 10 ° or greater in the angle between the line that bisects the thigh and a line that bisects the lower leg, during descent of single-limb squat. Potential participants were excluded if they reported a (1) body mass index greater than 30 kg/m² to minimize potential additional error due to excessive skin movement over bony landmarks, (2) history or current report of knee ligament, tendon, or cartilage injury; patellar instability or dislocation; or prior knee surgery, (3) known pregnancy, or (4) neurological involvement that would influence movement. All participants read and signed an informed consent approved by the Saint Louis University institutional review board (IRB# 24433). The study protocol, also approved by the institutional review board, followed federal and state regulations and the Declaration of Helsinki guidelines protecting human participants.

Procedures

Three-dimensional kinematic data were captured with an 8-camera motion analysis system (Vicon, Oxford Metrics LTD, Oxford, England) using methods described previously.² Prior to data collection, reflective markers were placed bilaterally over the iliac crests, anterior superior iliac spines, posterior superior iliac spines, medial and lateral femoral condyles, medial and lateral malleoli, posterior calcanei, lateral and anterior midfoot regions, and 1^st and 5^th metatarsal heads. Thermoplastic shells each with 4 reflective markers were affixed to the lateral mid-thigh and mid-shank regions bilaterally.

Following donning of all reflective markers, participants performed a standing calibration trial, followed by single-limb squat, stair ascent/descent, sit-to-stand/stand-to-sit in randomized order. Due to set-up, if stair ascent or descent was chosen, stair descent or ascent was next, respectively. This also was true for sit-to-stand and stand-to-sit. Participants performed single-limb squat on the involved or more painful limb; the involved side was analyzed during stair ascent/descent and the sit-to-stand/stand-to-sit tasks. For single-limb squat, participants were instructed to keep their arms at their side while bending the knee to at least 60 ° (visually confirmed by investigator). For stair ascent/descent, participants were instructed to ascend/descend the three-step staircase (6” riser; 12” tread) step over step, leading with the involved limb (ascent) or uninvolved limb (descent), using their usual pattern and pace. For sit-to-stand, participants were seated on a rigid bench, with the seat height adjusted such that their hips and knees were flexed to 90 °. They were instructed to face straight ahead and keep their arms at their sides or rest their hands on their thighs as they stood up. They were not allowed to use the chair or their thighs for upper extremity support. For stand-to-sit, participants stood in front of the bench using a self-selected foot placement. Participants were given the same instructions regarding orientation and use of upper extremities as with sit-to-stand. During all tasks, participants were instructed to perform each task at their usual pace, except for single limb squat, in which they were instructed to complete a full squat (down and up) in ∼4 seconds.¹⁰ No additional instructions were given to the participant about the position of the knee relative to the foot or hip. Participants completed three trials of each task after being allowed several practice trials to become familiar with the task.

Pain during each task (task-elicited pain) was assessed using a visual analog scale (VAS). The VAS was a 100mm line with a left anchor of “no pain,” and a right anchor of “worst imaginable pain.” After completion of each condition, participants rated their average pain along the VAS during that particular condition by placing a hash mark on the line. Pain levels were calculated as the distance in millimeters measured from the left anchor to the point where the hash mark crossed the line.

Data Processing

All 3D kinematic data were processed using Visual3D™ software (C-Motion, Inc., Rockville, MD, USA). The marker trajectories were low-pass filtered using a 4^th-order Butterworth filter with a 6 Hz cutoff frequency. A six degrees of freedom model incorporating the pelvis (CODA model, Charnwood Dynamics Ltd., UK), thigh, shank, and foot was used for data processing.^2,26

The thigh frontal plane was defined at the proximal end by the hip joint center and at the distal end by the two femoral epicondyle markers. The shank frontal plane was defined at the proximal end by the thigh distal endpoint and at the distal end by the two malleolus markers. The foot frontal plane was defined at the proximal end by the two malleolus markers and at the distal end by the projection on the floor of the two malleolus markers. The local coordinate system of each segment was located at the proximal endpoint of each segment. The frontal plane defined the orientation of the x-axis (sagittal plane rotation). The z-axis (transverse plane rotation) was aligned so that it passed through the proximal endpoint and the distal endpoint of the segments. The y-axis (frontal plane rotation) was oriented orthogonal to both x and z axes.²⁶ The reference frame of the proximal segment was used to calculate all hip and knee angles, measured to 1/100^th of a degree. Flexion, adduction, and medial rotation at both joints were calculated in the positive direction. For all movement tasks, hip and knee angles were calculated at 45 º of knee flexion to allow assessment of frontal and transverse plane angles when the knee was in a similar sagittal plane position. The average of the three trials for each task was used for further data analysis.

Data Analysis

Data were analyzed with IBM SPSS Statistics version 24 (IBM SPSS Statistics for Windows, Armonk, NY, USA). Descriptive statistics were calculated to determine the mean and standard deviation of the hip frontal and transverse plane angles, knee frontal and transverse plane angles, and pain during each condition. Pearson product-moment correlation coefficients were calculated to examine relationships between conditions for each individual variable. A p value less than 0.05 was considered significant.

RESULTS

Descriptive statistics

Descriptive statistics for hip frontal and transverse plane angles, knee frontal and transverse plane angles, and pain during each condition are provided in Table 1.

Table 1.

Descriptive variables presented as Mean (SD) for pain and hip and knee frontal and transverse plane angles at 45 º of knee flexion during single-limb squat, stair ascent, stair descent, sit-to-stand, and stand-to-sit. In the frontal plane, a positive value indicates adduction, a negative value indicates abduction; in the transverse plane a positive value indicates medial rotation, a negative value indicates lateral rotation.

	Single-Limb Squat	Stair Ascent	Stair Descent	Sit-to-Stand	Stand-to-Sit
Hip
Frontal Plane	7.18 º (4.47 º)	8.78 º (2.78 º)	4.77 º (2.52 º)	-0.45 º (3.50 º)	0.57 º (3.48 º)
Transverse Plane	1.82 º (4.94 º)	1.11 º (4.39 º)	1.68 º (5.34 º)	-3.43 º (5.15 º)	-2.38 º (4.07 º)
Knee
Frontal Plane	-0.82 º (4.95 º)	-0.57 º (5.00 º)	-0.34 º (4.3 º)	-3.56 º (5.14 º)	-4.22 º (5.41 º)
Transverse Plane	4.86 º (4.07 º)	3.92 º (3.89 º)	3.65 º (4.74 º)	7.74 º (4.45 º)	7.39 º (3.68 º)
Pain during Task (VAS; mm)	27.17 (18.68)	13.91 (12.13)	18.39 (16.82)	10.78 (10.20)	11.87 (10.47)

Abbreviations: VAS: Visual Analog Scale, 0-100 with 0 anchored by “no pain” and 100 anchored with “worst pain imaginable”

Pearson Product-Moment Correlations

Hip frontal plane: Pearson product-moment correlations between conditions ranged from r = 0.23 to r = 0.76 with 7 of 10 relationships being statistically significant at p < 0.05 or lower (Table 2).

Table 2.

Hip frontal plane Pearson product-moment correlations between single-limb squat, stair ascent, stair descent, sit-to-stand, and stand-to-sit.

	Single- Limb Squat	Stair Ascent	Stair Descent	Sit-to-Stand	Stand-to-Sit
Single-Limb Squat	-
Stair Ascent	r = 0.67 p < 0.01†	-
Stair Descent	r = 0.23 p  =  0.29	r = 0.50 p  =  0.02*	-
Sit-to-Stand	r = 0.56 p < 0.01†	r = 0.51 p  =  0.02*	r = 0.27 p  =  0.22	-
Stand-to-Sit	r = 0.48 p  =  0.02*	r = 0.54 p < 0.01†	r = 0.27 p  =  0.21	r = 0.76 p < 0.01†	-

*p<0.05, ^†p<0.01

Hip transverse plane: Pearson product-moment correlations between conditions ranged from r = 0.31 to r = 0.90 with 7 of 10 relationships being statistically significant at p < 0.01 or lower (Table 3).

Table 3.

Hip transverse plane Pearson product-moment correlations between single-limb squat, stair ascent, stair descent, sit-to-stand, and stand-to-sit.

	Single-Limb Squat	Stair Ascent	Stair Descent	Sit-to-Stand	Stand-to-Sit
Single-Limb Squat	-
Stair Ascent	r = 0.83 p < 0.01*	-
Stair Descent	r = 0.37 p  =  0.09	r = 0.58 p < 0.01*	-
Sit-to-Stand	r = 0.76 p < 0.01*	r = 0.78 p < 0.01*	r = 0.31 p  =  0.16	-
Stand-to-Sit	r = 0.81 p < 0.01*	r = 0.74 p < 0.01*	r = 0.32 p  =  0.13	r = 0.90 p < 0.01*	-

*p<0.01

Knee frontal plane: Pearson product-moment correlations between conditions ranged from r = 0.87 to r = 0.95 with 10 of 10 relationships being statistically significant at p < 0.01 or lower (Table 4).

Table 4.

Knee frontal plane Pearson product-moment correlations between single-limb squat, stair ascent, stair descent, sit-to-stand, and stand-to-sit.

	Single-Limb Squat	Stair Ascent	Stair Descent	Sit-to-Stand	Stand-to-Sit
Single-Limb Squat	-
Stair Ascent	r = 0.95 p < 0.01*	-
Stair Descent	r = 0.91 p < 0.01*	r = 0.91 p < 0.01*	-
Sit-to-Stand	r = 0.92 p < 0.01*	r = 0.93 p < 0.01*	r = 0.90 p < 0.01*	-
Stand-to-Sit	r = 0.87 p < 0.01*	r = 0.88 p < 0.01*	r = 0.87 p < 0.01*	r = 0.94 p < 0.01*	-

*p<0.01

Knee transverse plane: Pearson product-moment correlations between conditions ranged from r = 0.54 to r = 0.86 with 10 of 10 relationships being statistically significant at p < 0.01 or lower (Table 5).

Table 5.

Knee transverse plane Pearson product-moment correlations between single-limb squat, stair ascent, stair descent, sit-to-stand, and stand-to-sit.

	Single-Limb Squat	Stair Ascent	Stair Descent	Sit-to-Stand	Stand-to-Sit
Single-Limb Squat	-
Stair Ascent	r = 0.75 p < 0.01*	-
Stair Descent	r = 0.66 p < 0.01*	r = 0.83p < 0.01*	-
Sit-to-Stand	r = 0.70 p < 0.01*	r = 0.81 p < 0.01*	r = 0.74 p < 0.01*	-
Stand-to-Sit	r = 0.54 p < 0.01*	r = 0.77 p < 0.01*	r = 0.69 p < 0.01*	r = 0.86 p < 0.01*	-

*p<0.01

Task-elicited pain: Pearson product-moment correlations between conditions ranged from r = 0.59 to r = 0.85 with 10 of 10 relationships being statistically significant at p < 0.01 or lower (Table 6).

Table 6.

Pearson product-moment correlations between single-limb squat, stair ascent, stair descent, sit-to-stand, and stand-to-sit for pain as reported during each task.

	Single-Limb Squat	Stair Ascent	Stair Descent	Sit-to-Stand	Stand-to-Sit
Single-Limb Squat	-
Stair Ascent	r = 0.63 p < 0.01*	-
Stair Descent	r = 0.71 p < 0.01*	r = 0.78 p < 0.01*	-
Sit-to-Stand	r = 0.64 p < 0.01*	r = 0.80 p < 0.01*	r = 0.64 p < 0.01*	-
Stand-to-Sit	r = 0.62 p < 0.01*	r = 0.59 p < 0.01*	r = 0.73 p < 0.01*	r = 0.85 p < 0.01*	-

*p<0.01

DISCUSSION

The primary purpose of the current study was to examine the consistency of dynamic knee valgus and task-elicited pain demonstrated by females with PFP across a variety of functional tasks. The authors hypothesized there would be a strong relationship between the different functional tasks with respect to kinematics and pain. A secondary purpose was to assess the correlation between the clinical test of single-limb squat and functional tasks. The authors hypothesized that individuals who demonstrated greater dynamic knee valgus and task-elicited pain during the clinical test of single-limb squat would demonstrate greater dynamic knee valgus and task-elicited pain during stair ascent/descent and sit-to-stand/stand-to-sit tasks. The results of the current study indicate that females with PFP demonstrate a positive relationship in movement patterns and task-elicited pain across functional tasks. Furthermore, females with PFP who demonstrate greater dynamic knee valgus and pain during the clinical test of single-limb squat would demonstrate greater dynamic knee valgus and pain during stair ascent/descent and sit-to-stand/stand-to-sit.

Existing literature^2,4,6-10 pertaining to movement patterns in people with knee pain/injury conditions provides information about behavior during single tasks or high intensity activities leading to recognition of components of dynamic knee valgus as a risk factor for patellofemoral and other knee pathologies. However, the repetition of a movement pattern performed during functional tasks throughout the day, despite being a lower level of intensity could have a cumulative effect on stress at the knee joint.

Powers et al^1,27 modeled the potential contribution of an impaired movement pattern on the stress at the knee joint, arguing an impaired movement contributes to the development of increased stress and, subsequently pain. Dye²⁸ describes how load applied to the patellofemoral joint that is in excess of what is allowable to maintain tissue homeostasis can result in increased stress on tissues and the perception of pain. Given the demands placed on the knee across the day during routine daily activities such as sit-to-stand or ascending/descending stairs, recognizing impaired movement patterns during daily tasks may provide insight into the cumulative microtrauma placed on tissues throughout the day. This microtrauma may potentially contribute to knee pain. The results of the current study indicate individuals with greater dynamic knee valgus during one task may demonstrate greater dynamic knee valgus across a variety of functional tasks. This suggests a potential for cumulative microtrauma across the day, potentially contributing to the development or persistence of PFP. This concept has support from findings observed in the lumbar spine. Marich et al²⁹ reported that people with low back pain adopted consistent movement patterns when performing a variety of functional tasks. These patterns, different from those of pain-free individuals, were associated with greater functional limitations.

Another primary finding of this study is the positive relationship between conditions in the task-elicited pain reported by participants. On average, participants who reported higher pain in one activity reported higher pain in other activities as well. This has potential negative ramifications in that women with PFP may elect to limit participation in basic activities of daily living because of pain. Studies have shown that higher pain levels in people with musculoskeletal pain are associated with greater perceived interference in functional activities³⁰ and greater disability.³¹ While it is not clear that there is a one-to-one relationship between the level of pain and degree of dynamic knee valgus during a task, there is evidence that increasing dynamic knee valgus results in an immediate increase in PFP in women who performed a single-limb squat task.² As such, dynamic knee valgus as a proposed mechanism of PFP in women is plausible.

Interestingly, the positive correlations observed were stronger in the knee kinematic variables than in those of the hip (knee coefficients ranged from 0.54 to 0.94 and ten out of ten relationships were statistically significant). One explanation for this finding may be that frontal and transverse plane movement at the knee is more “fixed” than at the hip. Women with PFP may be able to alter their hip kinematics more readily, choosing slightly different movement strategies at the hip joint depending on the task, whereas the knee strategies are less variable (e.g. the knee exhibits a similar kinematic pattern regardless of the task). Reduced variability of movement patterns may contribute to injury, pain and poor task performance.³² Perhaps this is one explanation for why pain manifests at the patellofemoral joint.

Among the five tasks, stair descent had the lowest correlations among the hip kinematic variables, with stair ascent having the only positive association with stair descent. This finding is difficult to interpret, as stair descent was not the only single-limb task performed nor was it the only eccentric task (e.g. single limb squat). However, stair descent and single-limb squat differed in the location of the non-weight bearing limb as the task was performed. During stair descent, the non-weight bearing limb was in front of the stance limb, whereas during the squat task it was behind the stance limb. Khuu and Lewis³³ have shown that the position of the non-weight bearing limb influences the kinematics of the weight bearing limb during a single-limb squat. Perhaps the anterior position of the non-weight bearing limb during stair descent contributed to the lack of associations among hip kinematics during the other tasks.

There are several clinical implications of the current findings. If routine low-level daily tasks are performed throughout the day with an impaired movement pattern that might contribute to PFP, rehabilitation interventions might need to focus on improving limb alignment during repeated practice of various daily tasks, and emphasize education on the importance of maintaining improved limb alignment throughout the day. Results from the larger primary feasibility study investigating a task-specific movement training intervention in women with PFP²⁰ supports this strategy; wherein hip and knee kinematics as well as pain improved following the intervention.²⁰ Another implication is that movement during a single-limb squat, a commonly used clinical screening test, may be an indicator of movement during daily activities, thus providing clinicians with a representative and informative screening tool for dynamic knee valgus. Further, while the magnitude of pain might be higher during the single limb squat than in less demanding tasks, the level of pain a person has during the squat may be indicative of the level of pain they have during daily activities, again supporting the single limb squat as an informative clinical screening tool. It is possible a different task such as bilateral partial squat, which may cause less pain, could be used as a screening tool. Additional study to explore alternative, less painful screening tools that correlate to functional tasks would be beneficial.

There are several limitations of the current study. First, there was no pain-free comparison group to confirm that the participants with PFP demonstrated a greater degree of dynamic knee valgus than pain-free individuals during the functional tasks. The primary study from which this subset of data was derived from was an intervention study where pain-free participants were not included. However, it has been previously reported in the literature that females with PFP, compared to pain-free individuals, demonstrate greater dynamic knee valgus components during stair descent,³⁴ single-limb squat^35-37 and across progressively higher intensity activities including single-limb squatting, running, and single-limb jumping.¹⁰ Further study would be necessary to determine whether the level of dynamic knee valgus demonstrated during all functional tasks is greater than in pain-free individuals. A second limitation of this study is that only females were included. While females with PFP and dynamic knee valgus may represent an important phenotype of people with PFP, the information collected here may not be generalizable to men; further study would be necessary to assess this same question in men.

CONCLUSION

Females with PFP demonstrated a positive relationship in dynamic knee valgus kinematics and task-elicited pain across five functional tasks. The clinical test of single-limb squat was representative of movement patterns and pain during the functional tasks.