Are Altered Kinematics in Runners With Patellofemoral Pain Sex Specific?

Abstract

Background:

Altered kinematics have been frequently observed in runners with patellofemoral pain (PFP), and few studies have aimed to understand the influence of sex on kinematics of this population. The aim of this study was to investigate whether altered hip and knee kinematics in runners with PFP are sex specific.

Hypothesis:

Kinematics will be different between female and male runners with and without PFP.

Study Design:

Case-control study.

Level of Evidence:

Level 2.

Methods:

Eighty-four runners were divided into 4 groups: 42 runners with PFP (20 women, 22 men) and 42 asymptomatic runners (21 women, 21 men). Three-dimensional gait analyses of the hip in the frontal and transverse plane and the knee in the frontal plane were analyzed at self-selected running speed on a treadmill. One-way analysis of covariance was used to test for differences in kinematic variables between groups.

Results:

Women with PFP ran with a significantly greater peak hip adduction compared with men with PFP (mean difference [MD] = 4.45°; P = 0.00; effect size [ES] = 0.58) and male control subjects (MD = 4.2°; P = 0.01; ES = 0.54) and greater hip adduction range of motion (ROM) than men with PFP (MD = 3.44°; P = 0.01; ES = 0.49). No significant differences were identified between women with and without PFP. Female control subjects ran with greater peak hip adduction than men with PFP (MD = 5.46°; P < 0.01; ES = 0.58) and male control subjects (MD = 5.21°; P < 0.01; ES = 0.55); greater hip adduction ROM than men with PFP (MD = 4.02°; P = 0.00; ES = 0.52) and male control subjects (MD = 2.91°;P = 0.04; ES = 0.36); and greater peak knee abduction than men with PFP (MD = 3.35°; P = 0.02; ES = 0.44) and male control subjects (MD = 3.69°; P = 0.01; ES = 0.4).

Conclusion:

Women have greater hip adduction than men regardless of the presence of PFP. There were no kinematics difference between women with and without PFP. Comparisons of hip internal rotation between all groups were nonsignificant.

Clinical Relevance:

Altered hip and knee kinematics does not appear to be sex specific in runners with PFP.

Patellofemoral pain (PFP) is the most-frequent running-related injury ²⁰ and is characterized by pain localized around or behind the patella, ⁶ with a worsening of symptoms during activities that incur high patellofemoral joint (PFJ) loads, such as running. ¹⁸ Some factors are considered to be associated with PFP, including excessive hip adduction, hip internal rotation, and knee abduction. It has been proposed that these factors increase patellofemoral stress and, thus, contribute to the persistence of PFP. ²⁷

Despite these factors, which have been previously widely studied in several PFP populations and during specific activities, including running, studies that compared hip and knee kinematics between runners with and without PFP showed divergent results. In mixed-sex cohorts, greater peak hip adduction,^13,25 greater hip internal rotation range of motion (ROM), ¹³ and greater knee abduction ROM ¹³ have been observed in those with PFP compared with healthy control subjects. However, other studies have found that peak hip adduction and peak hip internal rotation in those with PFP is lower than ⁷ or not different from ⁹ healthy control subjects. The number of participants of each sex in these studies can provide an explanation for these divergent findings. Although Neal et al ²⁵ included the same number of women and men, Dierks et al ⁷ and Esculier et al ⁹ included a far greater number of women and men in their samples, and Fox et al, ¹³ in turn, included a slightly higher number of women than men.

Despite all these studies having contributed to the understanding of kinematics of runners with PFP, few studies have aimed to specifically understand the influence of sex on the altered kinematics of runners with PFP. Willy et al ²⁹ compared the lower extremity mechanics of men with PFP with women with PFP during running, and they observed that women with PFP ran with greater peak hip adduction and less peak knee adduction than men with PFP. Only one study compared women and men with and without PFP in separate groups and found that women with PFP ran with greater peak hip adduction angle compared with female controls and with a trend toward having a significantly greater peak hip adduction angle when compared with men with and without PFP. ²⁵ Although the participants were assessed at their self-selected speed in the study by Neal et al, ²⁵ this variable (running speed) was not considered during data analysis and may have influenced the results. Furthermore, these studies considered the entire running stance phase rather than analyzing it in subphases (deceleration and acceleration). It is possible that there are significant kinematic differences between male and female runners with and without PFP at different times of the running stance phase, because in the subphases, the kinematics of lower limbs in the sagittal, frontal, and transverse planes are different.^15,26

Therefore, a better understanding of sex influence in runners with PFP could help in the specificity of treatment based on sex. Thus, the purpose of the study is to investigate whether altered hip and knee kinematics in runners with PFP are sex specific. We hypothesized that kinematics will be different between woman and men with and without PFP and women with PFP will demonstrate greater hip adduction, hip internal rotation, and knee abduction.

Methods

Participants

All participants read and signed an informed consent form approved by the University Ethics Committee for Human Investigations. A total of 84 participants were recruited for this case-control study and divided into 4 groups: 42 runners with PFP (20 women, 22 men) and 42 asymptomatic runners (21 women, 21 men). Participants were recruited through advertisements at São Carlos Federal University, running clubs, running events, and social network websites between September 2016 and March 2019. An a priori sample size calculation for 1-way, fixed-effects analysis of variance was conducted using data from a pilot work for this study (n = 8 for each group). Using the variable with the highest standard deviation, peak knee abduction angle, a required minimum sample size of 19 participants per group was calculated based on a 90% predicted power and an alpha of 0.05 with an effect size (ES) of 0.44.

Inclusion/Exclusion Criteria

All participants were included according the following criteria: age between 18 and 35 years; run a minimum of 15 km per week in the last 3 months; and used a rearfoot strike (RFS) pattern. ⁴ The RFS pattern was confirmed by a real-time qualitative camera (120 Hz) positioned lateral to the foot (1 m away). ¹⁶ To be included in the PFP group, participants were required to have presence of peripatellar/retropatellar pain during running and in at least 2 of the following functional activities ⁸ : stair ascent or descent, kneeling, squatting, prolonged sitting, jumping, and isometric quadriceps contraction; to present the worst knee pain experienced in the previous week of minimum 3 of 10 points on the visual analog scale (VAS); and self-report PFP symptoms during and/or after their running training for at least the past 3 months unrelated to any traumatic event. Participants were excluded if they reported: previous history of knee surgery; current back, hip, or ankle joint injury or pain; patellar instability; signs or symptoms of meniscal or knee ligament involvement; or any neurological condition that would affect movement.

Procedures

After confirmation of the diagnosis of PFP by a clinical physiotherapist based on previously published diagnostic criteria, participants in PFP group answered the Anterior Knee Pain Scale to characterize their functional capacity. ¹⁷ Assessments were performed in a single session. The symptomatic lower limb was assessed for runners with PFP. In cases of bilateral symptoms, the most affected lower limb (highest self-reported pain) was chosen for analysis. ²⁴ For the asymptomatic runners, the dominant lower limb was assessed, and determined by asking which leg they would use to kick a ball as far as possible. ³⁰ A neutral running shoe (Asics Gel-Equation 5, ASICS, Kobe, Japan) was provided for all participants in order to reduce the influence of shoe condition. ¹

Kinematic Data Collection

Kinematic data were collected using a 7-camera 3-dimensional (3D) motion analysis system (Qualisys Motion-Capture System, Qualisys Medical-AB) at 240 Hz. Sixteen passive reflective anatomical markers of 14 mm in diameter were placed on each participant at the following landmarks as described in a previous study ²¹ in accordance with a modified model Helen-Hayes ⁸ : the highest points of the iliac crests, the anterior superior and posterior superior iliac spines, the base of the sacrum, the femoral greater trochanter, the medial and lateral femoral epicondyles, the medial and lateral malleoli, the head of the first metatarsal, the head of the fifth right metatarsal, and the base of second metatarsal. Three trace markers (clusters) were fixed on the posterior side of the thigh and leg and the posterior region of the shoe (posterior aspect of the calcaneus). A static trial with the participant standing in neutral position was performed to obtain the global (laboratory) coordinate system and to determine the reference for subsequent kinematic analysis.

All participants warmed up on a treadmill at a constant speed of 4.5 km/h for 5 min. Subsequently, the self-selected running speed was determined according to the comfort report of each participant by asking what pace (minutes per kilometer) they would select for an easy 20-min run. ¹² This speed could be adjusted, if requested by the subject, during a period of 5 min. After the running speed was established, the participant ran for approximately 2 min, and at least 30 consecutive steps of the evaluated lower limb were recorded.

Data Reduction

The average of 10 consecutive steps was analyzed to obtain the kinematic variables. ⁸ Visual 3D software (C-Motion, Inc.) was used to calculate anatomical joint coordinate system and filter marker trajectory (fourth-order, zero-lag, low-pass Butterworth at 12 Hz). The Cardan angles were calculated relative to the static standing trial using the joint coordinate system definitions recommended by the International Society of Biomechanics.^14,31 The knee joint center was defined as the midpoint between the reference markers positioned at medial and lateral femoral epicondyle, and the hip joint center was estimated as one-quarter of the distance from the ipsilateral to the contralateral greater trochanter.^11,22,28

Initial contact was identified as the point in time when the calcaneus marker moved from positive to negative velocity in the anteroposterior direction ³² and toe-off was determined by the second peak knee extension. ¹⁰ Stance phase was considered as initial ground contact to toe-off. In addition, the stance phase was divided into two subphases: (1) deceleration, defined as initial contact to peak knee flexion; (2) acceleration, defined as peak knee flexion to toe-off.

Kinematic variables were determined using the Matlab software (Mathworks). Variables of interest were: peak hip adduction (HADD), peak hip internal rotation (HIR), and peak knee abduction (KABD); and hip adduction ROM, hip internal rotation ROM, and knee abduction ROM. All variables were analyzed during the stance phase and in stance subphases (deceleration and acceleration).

A pilot study was conducted to verify the test-retest reliability of the peaks of kinematics measurements. Nine participants were evaluated on 2 separate occasions for 7 days. The intraclass correlation coefficient (ICC_1,1) and standard error of measurement were, respectively, 0.79º and 1.46° for HADD, 0.81º and 1.73° for HIR, and 0.91º and 1.44° for KABD. The minimal detectable difference (MDD₉₅) was 4.03º for HADD, 4.78º for HIR, and 3.97º for KABD.

Statistical Analysis

All statistical testing was performed using SPSS software (version 25, IBM SPSS Statistics). The Shapiro-Wilk and Levene tests were used to analyze the data with respect to their statistical distribution and variance homogeneity, respectively. One-way analysis of covariance with running speed as a covariate was used to test for differences in kinematic variables between the 4 groups in total stance phase and in each subphase. In the event of a significant main effect, post hoc analysis using Bonferroni test was performed to compare groups using P < 0.05 for the level of significance. ES of all identified intergroup differences were calculated using the Cohen’s d and interpreted as 0.2, 0.5, and 0.8 as small, medium, and large, respectively. ⁵

Results

Results of the comparison between groups for participant demographics, running characteristics, pain level, and self-reported functional score are presented in Table 1. Men with PFP and male control subjects ran with a statistically higher running speed than women with PFP and female control subjects.

Table 1.

Mean (standard deviation) participant characteristics

Variable	Women with PFP (n = 20)	Men with PFP (n = 22)	Female Control subjects (n = 21)	Male Control subjects (n = 21)
Age (years)	27.4 (3.4)	28.3 (4.5)	27 (6.1)	28.5 (4.8)
Height (m)	1.66 (0.04)	1.77 (0.08) a	1.63 (0.09)	1.7 (0.07) a
Mass (kg)	61 (7.3)	80 (11.4) a	60.5 (10.5)	79.2 (11.5) a
BMI (kg/m²)	22.1 (2.3)	25.3 (2.5) a	22.4 (2.5	24.5 (2.6) a
Weekly distance (km)	20.05 (8.4)	22.5 (5.8)	27.8 (12.6)	39.7 (20.8) b
Running experience (years)	1.6 (1.7)	2.5 (2.3)	4.1 (5)	4.7 (3.2) c
Running Speed (km/h)	8 (0.8)	9.6 (1.3) d	9 (0.9)	10 (1.5) d
Symptom duration (months)	14.5 (17.6)	14.3 (15.9)	N/A	N/A
AKPS score	79.3 (9)	78.4 (8)	N/A	N/A
VAS	4.8 (1.5)	4.4 (1.2)	N/A	N/A

AKPS, Anterior Knee Pain Scale; BMI, body mass index; N/A, not applicable; PFP, patellofemoral pain; VAS, visual analog scale.

^aMen with PFP and male control subjects significantly different from women with PFP and female control subjects (P < 0.01).

^bMale control subjects significantly different from women and men with PFP and female control subjects (P < 0.05).

^cMale control subjects significantly different from women with PFP (P < 0.05).

^dMen with PFP and male control subjects significantly different from women with PFP (P < 0.01).

Average time normalized curves of hip and knee kinematics during running stance phase for all groups are presented in Appendix Figure A1 (available in the online version of this article).

Total Stance Phase

Comparing the groups in total stance phase, women with PFP ran with a significantly greater peak HADD than men with PFP (mean difference [MD] = 4.45°; 95% confidence interval [CI] 2.5-6.3; P = 0.00; ES = 0.58) and male controls (MD = 4.2°; 95% CI 2.1-6.2; P = 0.01; ES = 0.54), but there was no difference when compared with female control subjects (Table 2). Female control subjects ran with a significantly greater peak HADD than men with PFP (MD = 5.46°; 95% CI 3.1-7.7; P = 0.00; ES = 0.58) and male control subjects (MD = 5.2°; 95% CI 2.7-7.6; P = 0.00; ES = 0.55). In addition, female control subjects ran with a significantly greater peak KABD than men with PFP (MD = 3.35°; 95% CI 1.2-5.4; P = 0.02; ES = 0.44) and male control subjects (MD = 3.69°; 95% CI 1.09-6.2; P = 0.01; ES = 0.4). No differences were detected in peak HIR between all groups. With respect to ROM, women with PFP presented significantly greater HADD ROM than men with PFP (MD = 3.44°; 95% CI 1.5-5.3; P = 0.01; ES = 0.49), but there was no significant difference between female and male control subjects. Female control subjects also ran with greater HADD ROM than men with PFP (MD = 4.02°; 95% CI 2.01-6.03; P = 0.00; ES = 0.52) and male control subjects (MD = 2.9°; 95% CI 0.5-5.2; P = 0.04; ES = 0.36). There was no difference between all groups for the HIR and KABD ROM (Table 2).

Table 2.

Mean (standard deviation) joint angle measure for the men and women with and without PFP during entire stance phase of running

	Women with PFP (n = 20)	Men with PFP (n = 22)	Female Control subjects (n = 21)	Male Control subjects (n = 21)
Peak
HADD (°)	13.75 (3.4) a	9.3 (2.79)	14.76 (4.49) c	9.55 (3.14)
HIR (°)	18.5 (4.38)	15.68 (4.72)	16.9 (5.13)	15.16 (3.65)
KABD (°)	4.52 (2.55)	2.39 (3.03)	5.74 (3.76) c	2.05 (4.54)
ROM
HADD (°)	10.39 (3.6) b	6.95 (2.29)	10.97 (4.04) c	8.06 (3.46)
HIR (°)	18.62 (4.1)	18.67 (4.81)	17.46 (5.46)	16.95 (5.22)
KABD (°)	10.13 (2.59)	9.96 (3.53)	9.9 (3.54)	9.31 (2.77)

HADD, hip adduction; HIR, hip internal rotation; KABD, knee abduction; PFP, patellofemoral pain; ROM, range of motion.

^aWomen with PFP significantly greater than men with PFP and male control subjects (P < 0.05).

^bWomen with PFP significantly greater than men with PFP (P < 0.05).

^cFemale Control subjects significantly greater than men with PFP and male control subjects (P < 0.05).

Deceleration Phase

In deceleration phase, women with PFP presented significantly greater peak HADD compared with men with PFP (MD = 4.32°; 95% CI 2.3-6.2; P = 0.00; ES = 0.56) and male control subjects (MD = 4.23°; 95% CI 2.1-6.3; P = 0.01; ES = 0.54) and also presented significantly greater HADD ROM than men with PFP (MD = 2.07°; 95% CI 1.2-2.9; P = 0.01; ES = 0.6) and male control subjects (MD = 1.88°; 95% CI 0.9-2.8; P = 0.02; ES = 0.52) (Table 3). There was no significant difference between women with PFP and female control subjects for HADD (peak and ROM). Female control subjects presented significantly greater peak HADD than men with PFP (MD = 5.4°; 95% CI 3.1-7.6; P = 0.00; ES = 0.58) and male control subjects (MD = 5.31°; 95% CI 2.9-7.7; P = 0.00; ES = 0.56) and also presented significantly greater HADD ROM than men with PFP (MD = 1.61°; 95% CI 0.4-2.7; P = 0.02; ES = 0.38). No differences were detected in HIR and KABD (peak and ROM) between all groups (Table 3).

Table 3.

Mean (standard deviation) joint angle measure for the men and women with and without PFP in the stance subphases (deceleration and acceleration)

	Women with PFP (n = 20)	Men with PFP (n = 22)	Female Control subjects (n = 21)	Male Control subjects (n = 21)
Peak deceleration
HADD (°)	13.59 (3.46) a	9.27 (2.8)	14.67 (4.46) b	9.36 (3.09)
HIR (°)	18.5 (4.38)	15.63 (4.76)	16.87 (5.19)	15 (3.69)
KABD (°)	1.25 (3.4)	0.32 (2.81)	2.58 (4.32)	0.13 (4.24)
ROM deceleration
HADD (°)	5.86 (1.49) a	3.79 (1.26)	5.4 (2.43) c	3.98 (1.55)
HIR (°)	7.3 (3.45)	6.55 (2.88)	8.18 (3.54)	5.67 (2.15)
KABD (°)	6.77 (2.48)	7.64 (3.08)	6.58(2.5)	7.04 (2.55)
Peak acceleration
HADD (°)	13.22 (3.4) a	8.2 (2.86)	14.08 (4.75) b	8.32 (3.51)
HIR (°)	12.74 (6.42)	10.33 (4.99)	10.8 (5.17)	11.44 (4.24)
KABD (°)	4.85 (2.52)	−2.31 (3.01)	5.64 (3.9) b	1.18 (4.71)
ROM acceleration
HADD (°)	9.79 (3.67) d	5.78 (2.25)	10.07 (4.18) b	6.5 (4.12)
HIR (°)	12.92 (3.33)	13.19 (4.75)	11.26 (4.63)	13.04 (4.85)
KABD (°)	8.94 (2.75)	9.01 (3.97)	8.68 (3.56)	7.61 (2.5)

HADD, hip adduction; HIR, hip internal rotation; KABD, knee abduction; PFP, patellofemoral pain; ROM, range of motion.

^aWomen with PFP significantly greater than men with PFP and male control subjects (P < 0.05).

^bFemale Control subjects significantly greater than men with PFP and male control subjects (P < 0.05).

^cFemale control subjects significantly greater than men PFP (P < 0.05).

^dWomen with PFP significantly greater than men with PFP (P < 0.05).

Acceleration Phase

In the acceleration phase, women with PFP presented significantly greater peak HADD than men with PFP (MD = 5.02°; 95% CI 3.07-6.9; P = 0.00; ES = 0.62) and male control subjects (MD = 4.9°; 95% CI 2.72-7.08; P = 0.00; ES = 0.57) (Table 3). Female control subjects presented significantly greater peak HADD than men with PFP (MD = 5.88°; 95% CI 3.4-8.2; P = 0.00; ES = 0.59) and male control subjects (MD = 5.76°; 95% CI 3.1-8.3; P = 0.00; ES = 0.56) and greater peak KABD than men with PFP (MD = 7.95°; 95% CI 5.85-10.09; P = 0.02; ES = 0.75) and male control subjects (MD = 4.46°; 95% CI 1.76-7.16; P = 0.00; ES = 0.45). Women with PFP ran with a significantly greater HADD ROM than men with PFP (MD = 4.01°; 95% CI 2.1-5.8; P = 0.01; ES = 0.55) and female control subjects also presented significantly greater HADD ROM compared with men with PFP (MD = 4.29°; 95% CI 2.2-6.3; P = 0.00; ES = 0.53) and male control subjects (MD = 3.57°; 95% CI 0.9-6.1; P = 0.01; ES = 0.39). There was no significant difference between women with PFP and female control subjects for HADD (peak and ROM). No differences were detected in HIR (peak and ROM) between all groups (Table 3).

Discussion

The aim of this case-control study was to investigate whether altered hip and knee kinematics in runners with PFP are sex specific. The results of this study partially support our hypothesis. As no very significant differences were found in the analysis, considering the entire stance phase in relation to the analysis considering the subphases (deceleration and acceleration), we discuss the results in a more general way. We found that female runners have greater hip adduction (peak and ROM) than male runners with and without PFP, but there is no difference between female runners with and without PFP. Furthermore, we found no differences in hip internal rotation among groups and only female controls showed a significant greater peak knee abduction compared with men with and without PFP.

Our findings indicate that women have greater hip adduction than men regardless of the presence of PFP in all running phases analysis (total stance phase, deceleration, and acceleration). This was previously shown in part by Willy et al ²⁹ who demonstrated that female participants with PFP had greater peak hip adduction than male participants with and without PFP. However, healthy women were not included in this study. Partially agreeing with our findings, although Neal et al ²⁵ did not find statistically significant difference in hip adduction between female participants with PFP and male participants with and without PFP, a statistical trend was observed for female participants with PFP to have greater hip adduction than male participants with and without PFP.

However, in contrast to our findings, Neal et al ²⁵ also reported that female participants with PFP ran with a significantly greater hip adduction angle compared with female control subjects. As previously pointed out, when analyzing the possible reasons for the contradictory results between Neal et al ²⁵ and the present study, two important differences emerged. First, considering the influence that running speed has on joint kinematics, ³ kinematic analysis in the study by Neal et al ²⁵ was performed with participants running at their preferred speed, but possible differences in this variable (running speed) between groups were not considered in the data analysis. In our study, we included running speed as a covariate in the statistical analysis, as it was statistically different between groups. The second reason can be explained by the duration of symptoms. Fox et al ¹³ found that runners with chronic PFP demonstrated greater peak hip adduction compared with runners with acute PFP. In the study conducted by Neal et al, ²⁵ female participants with PFP had a much longer duration of symptoms than female participants with PFP included in the present study. However, there are no studies that show that hip adduction in women can increase with increasing duration of symptoms. Therefore, future studies are encouraged to investigate this relationship between the influence of pain duration and hip adduction in female runners with PFP, contributing to a better treatment strategy in this population.

Excessive hip internal rotation can increase PFJ stress. Liao et al ¹⁹ reported a significant increase in mean hydrostatic pressure and mean octahedral shear stress in the PFJ when the femur was internally rotated 5° and 10°. No differences were found between groups for hip internal rotation (peak and ROM). This result is in agreement with the result by Neal et al ²⁵ that also compared female and male participants with and without PFP. Thus, increased hip internal rotation also does not seem to be sex specific in runners with PFP.

We found greater peak knee abduction in female control subjects compared with men with and without PFP during total stance phase and acceleration. However, there was no difference between women with and without PFP. Although women with PFP trended toward greater knee abduction angles than men with and without PFP, this difference was not enough to be statistically significant. A hypothesis for not having found differences between women with and without PFP, as with hip adduction, may also be related to the influence of the duration of symptoms in knee abduction, because it is possible that greater hip adduction position contributes to greater knee abduction angle in the stance phase. ¹¹ However, in order to determine whether this relationship is true or not, further studies are needed.

Limitations

The authors recognize that the present study has some limitations. First, owing to the cross-sectional design of this study, although we were able to detect differences between women and men, with and without PFP, we could not define the cause and consequence of kinematic alterations. In addition, we evaluated runners with an average duration of 14 months of symptoms, and it may be that a longer period of duration of these symptoms may have altered the knee and hip kinematics of these runners with PFP. Considering that the groups of runners with PFP ran a shorter weekly distance, future studies could assess the influence of this variable on the kinematics of these runners. Finally, knowing that pelvis and trunk movements can influence hip adduction during running ²³ and that muscle fatigue influences hip adduction in other pathological conditions of the knee (iliotibial band syndrome), ² it is important that future studies investigate the relationship of these factors in runners with PFP.

Conclusion

The results of this study indicate that women have greater hip adduction (peak and ROM) than men regardless of the presence of PFP in all running stance phases. There was no difference between women with and without PFP for this variable. Thus, although women have greater hip adduction during running, this increase is not sex specific in runners with PFP. In addition, women and men with and without PFP showed no difference in hip internal rotation during running. Last, healthy women run with greater knee abduction than men with and without PFP. However, no difference between the other groups was shown. Thus, greater knee abduction, as well as hip internal rotation, are also not sex specific in runners with PFP.

Therefore, proposing a treatment that is sex specific for runners with PFP, focused only on altered knee and hip kinematics, does not seem to be ideal for this population.