- Email: [email protected]
Altered patterns of pelvic bone motion determined in subjects with posterior pelvic pain using skin markers
Clinical Biomechanics 19 (2004) 456–464 www.elsevier.com/locate/clinbiomech
Altered patterns of pelvic bone motion determined in subjects with posterior pelvic pain using skin markers Barbara Hungerford
a,*
, Wendy Gilleard b, Diane Lee
c
a
b
School of Exercise and Sport Science, University of Sydney, Sydney, Australia School of Exercise Science and Sport Management, Southern Cross University, Lismore, Australia c Diane G. Lee Physiotherapist Corp., BC, Canada Received 14 February 2003; accepted 19 February 2004
Abstract Objective. To determine whether the pattern of pelvic bone motion, determined by skin markers, differs between control subjects and subjects with posterior pelvic pain. Design. Cross-sectional study of three-dimensional angular and translational motion of the innominates relative to the sacrum in two subject groups. Background. Comparative in vivo analysis of the 3D patterning of pelvic motion in subjects with posterior pelvic pain and controls is limited. Methods. Fourteen males with posterior pelvic pain and healthy age and height matched controls were studied. A 6-camera motion analysis system was used to determine 3D angular and translational motion of pelvic skin markers during standing hip flexion. Results. Posterior rotation of the innominate occurred with hip flexion in control subjects and pelvic pain subjects as previously reported in the literature. On the supporting leg, the innominate rotated posteriorly in controls and anteriorly in symptomatic subjects. Conclusion. Posterior rotation of the innominate, as measured using skin markers during weight bearing in controls may reflect activation of optimal lumbo-pelvic stabilisation strategies for load transfer. Anterior rotation occurred in symptomatic subjects, suggesting failure to stabilise intra-pelvic motion for load transfer. Relevance This study found that posterior rotation of the innominate occurred during weight bearing in controls. This movement pattern is thought to optimise stability of the pelvic girdle during increased loading. Conversely, anterior rotation occurred in symptomatic subjects during weight bearing. This is a non-optimal pattern and may indicate abnormal articular or neuromyofascial function during increased vertical loading through the pelvis. 2004 Elsevier Ltd. All rights reserved. Keywords: Sacroiliac joint; Pelvic motion; Pelvic stabilisation; Low back pain; Pelvic pain
1. Introduction A primary function of the lumbar spine and pelvis is to transfer the loads generated by body weight and gravity during standing, walking and sitting (Snijders et al., 1993). How well this load is managed dictates the efficacy of function. A small amount of motion occurs at
*
Corresponding author. Present address: Sydney Spine and Pelvis Centre, 101 Lyons Road, Drummoyne, NSW 2047, Australia. E-mail address: [email protected] (B. Hungerford). 0268-0033/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2004.02.004
the sacroiliac joints (SIJ) and the pubic symphysis during movements of the trunk and lower limbs (Jacob and Kissling, 1995; Walheim and Selvik, 1984). Consequently, during weight bearing activities, control (stabilisation) of intra-pelvic motion is required for transference of loads between the spine and the lower limbs (Snijders et al., 1993; Vleeming et al., 1990). According to Panjabi (1992) stability is achieved when the passive, active and control systems work together. Snijders et al. (1993) suggests that the passive, active and control systems produce approximation of the joint surfaces, essential if stability is to be insured. The
B. Hungerford et al. / Clinical Biomechanics 19 (2004) 456–464
amount of approximation required is variable and difficult to quantify as it is dependent on an individual’s structure (form closure) and the forces they need to control (force closure). The ability to effectively transfer load through the pelvis is dynamic and therefore depends on: (1) optimal function of the bones, joints and ligaments (Vleeming et al., 1989, 1990); (2) optimal function of the muscles and fascia (Hungerford et al., 2003; Richardson et al., 2002; Snijders et al., 1998; Vleeming et al., 1995a,b); (3) appropriate neural function (Hodges and Richardson, 1997; Hungerford et al., 2003). 1.1. Stabilisation and motion control of the pelvic girdle For every joint, there is a position called the selfbraced (close-packed) position in which there is maximum congruence of the articular surfaces and maximum tension on major ligaments. In this position, the joint is under significant compression and the ability to resist shear forces is enhanced by tensioning of the passive structures and increased friction between the articular surfaces (Snijders et al., 1993; Vleeming et al., 1990). The self-braced position of the SIJ is nutation of the sacrum or posterior rotation of the innominate (Vleeming et al., 1989). Studies have shown (Sturesson et al., 2000) that nutation of the sacrum relative to posterior rotation of the innominate occurs bilaterally whenever the lumbo-pelvic spine is loaded vertically (sitting, standing). Counternutation of the sacrum, or anterior rotation of the innominate, is thought to be a relatively less stable position for the SIJ (Vleeming et al., 1995b). The long dorsal ligament becomes taut during this motion, however tension in other ligaments such as sacrotuberous and interosseous ligaments decreases (Vleeming et al., 1996). At present there is only limited research comparing the in vivo pattern of pelvic motion during weight bearing and non-weight bearing activities. This is due to the invasiveness of the most reliable and valid methods of analysis, consequential ethical considerations for using invasive procedures to evaluate large number of subjects and difficulty in evaluating the small amplitude of in vivo SIJ motion. When an individual stands on one leg and flexes the contralateral hip, the non-weight bearing innominate posteriorly rotates relative to the sacrum (range of motion 0.1–5.0) (Jacob and Kissling, 1995; Sturesson et al., 2000). Side flexion and axial rotation of the innominate occur concurrently about sagittal and vertical axes respectively. Translation (motion along the sagittal, vertical and/or coronal axes) also occurs (Jacob and Kissling, 1995; Sturesson et al., 2000), although the specific direction of translation that occurs with angular motion is not fully understood. It has been hypothesised (Lee, 1999) that anterior and superior translation of the innominate, relative to the
457
sacrum, will occur with posterior rotation of the innominate. 1.2. Range of motion versus patterning of pelvic motion Buyruk et al. (1995) and Damen et al. (2002) established that a Doppler imaging system was able to measure stiffness of the SIJ. This research showed that stiffness of the SIJ is variable between subjects and therefore the range of motion is potentially variable. It also revealed that stiffness of right and left SIJs is symmetric in subjects without pelvic pain. The results support putting less emphasis on amplitude of motion and more on the pattern or symmetry of SIJ, or pelvic motion, as range of motion varies between subjects, however within one subject joint stiffness remains symmetric between sides. 1.3. The impact of posterior pelvic pain The SIJs and the posterior SIJ ligaments are a known source of posterior pelvic pain (Fortin et al., 1994; Vleeming et al., 2002). Jacob and Kissling (1995) noted that in the presence of SIJ symptoms, the amplitude of SIJ motion about a coronal axis increased during hip flexion in one subject. Mens et al. (1999) determined increased amplitude of anterior rotation of the innominate in posterior pelvic pain patients. Sturesson et al. (2000) reported no difference in the amplitude or pattern of either angular or translational motion of the innominates when the left and right SIJs were compared in subjects with posterior pelvic pain. No comparison was made with an asymptomatic group. Hungerford et al. (2003) showed that posterior pelvic pain alters the pattern of lumbo-pelvic muscle recruitment, while Buyruk et al. (1999) and Damen et al. (2002) showed that stiffness of the SIJ is asymmetric in subjects with pelvic pain and that asymmetrical stiffness of the SIJs is prognostic for pelvic impairment and pain. It is presently unknown if posterior pelvic pain alters the patterning of bone motion within the pelvis for single leg stance. 1.4. Posterior pelvic pain and the active straight leg raise test The supine active straight leg raise test (ASLR) (Mens et al., 2001) has been validated as a clinical test for measuring effective load transfer between the trunk and lower limbs. When the lumbo-pelvic region is functioning optimally, the leg should rise effortlessly from the table (effort graded from 0 to 5) (Mens et al., 1999). A correlation has been shown between positive ASLR findings and posterior pelvic pain (Mens et al., 1999; O’Sullivan et al., 2002). Similarly, Damen et al.
458
B. Hungerford et al. / Clinical Biomechanics 19 (2004) 456–464
(2001) and Buyruk et al. (1999) showed that the ASLR is positive in the presence of asymmetric stiffness of the SIJ. This suggests that altered pelvic stabilisation strategies may affect pelvic mobility (Buyruk et al., 1999; Damen et al., 2001). 1.5. Limitations of methodology in this study A non-invasive motion analysis system using skin mounted markers was chosen to acquire the kinematic data of pelvic bone motion during a standing hip flexion movement for ethical reasons. Errors in determining the range of motion are likely to occur with an optoelectronic system due to skin marker motion relative to underlying bony landmarks (Maslen and Ackland, 1994). A high resolution motion analysis system has however been reported to provide reliable and consistent in vivo data of lumbar segmental motion patterns (Gracovetsky et al., 1995). In this study, the authors recognise that the movements noted reflect motion of the innominates, sacrum, and femurs in conjunction with overlying skin, and therefore the main emphasis of this study was to investigate the patterns of bone motion rather than the range of motion. The aim of this study was to determine the threedimensional pattern of innominate bone motion occurring in subjects determined to have posterior pelvic pain and impaired pelvic stabilisation strategies during weight bearing and non-weight bearing components of a standing hip flexion movement. These results were compared to age and height matched controls with clinically assessed normal pelvic stabilisation and pelvic motion patterns. It was hypothesised that the pattern of innominate bone motion would alter in subjects with posterior pelvic pain during both components of the movement trial.
2. Methods 2.1. Subjects 2.1.1. Impaired pelvic stabilisation and posterior pelvic pain group Fourteen male subjects with SIJ pain and a mean (range) age, height and weight of 32.7 (24–47) years, 176.8 (168–184) cm, and 77.0 (71–90) kg respectively, volunteered for the study. The criteria for inclusion in this study were: 1. Each subject in the posterior pelvic pain group reported unilateral pain over the posterior pelvic/SI region (Fortin et al., 1994) for greater than two months, and no pain above the lumbo-sacral junction. The pain was consistently and predictably
aggravated by activities that vertically loaded the pelvis (walking, standing or sitting). 2. Positive results on the side of posterior pelvic pain in clinical tests for impaired lumbo-pelvic stabilisation. These tests included: (a) Active straight leg raise test (Mens et al., 1999, 2001): A positive test was indicated when the pelvis failed to remain in neutral alignment, and the subject reported difficulty or inability to elevate a straight leg in supine. The perceived difference of effort, or pain aggravation was scaled from 0 (not difficult to raise the leg) to 5 (unable to perform ASLR). (b) Standing hip flexion test (Mitchell, 1995): During a left standing hip flexion test, the subject stands on their right leg and flexes the left hip towards 90. The left innominate should posteriorly rotate relative to the sacrum (Jacob and Kissling, 1995). A positive test was indicated when superior motion of the posterior superior iliac spine (PSIS) was palpated relative to the sacrum. (c) Neutral zone analysis test (joint play) (Lee, 1999)––this test was used clinically to apply the research of Buyruk et al. (1999) and Damen et al. (2002) and to evaluate motion in the neutral zone of the SIJ. Panjabi (1992) noted that joints have non-linear load–displacement curves and that the size of the neutral zone may increase with injury, articular degeneration and/or weakness of the stabilising musculature and that this is a more sensitive indicator than angular range of motion for detecting instability. All symptomatic subjects demonstrated asymmetric stiffness of the SIJ when the innominate was glided relative to the sacrum (analysis of the neutral zone). As the reliability and predictive ability of the standing hip flexion and neutral zone analysis tests remains uncertain (Carmichael, 1987; Vincent-Smith and Gibbons, 1999) all clinical tests were required to be positive on the side of pain, in conjunction with the ASLR test, for inclusion in the posterior pelvic pain subject group. Subjects were excluded from the study if they could not flex each hip to 90 without pain, if they had undergone spinal surgery, or displayed overt neurological signs such as sensory paraesthesia or motor paresis. 2.1.2. Control group The SIJ pain group were age and height matched to a control group of 14 males with a mean (range) age, height, and weight of 33.5 (22–50) years, 176.0 (168–183) cm, and 72.5 (61–85) kg respectively. The control subjects had no history of low back pain in the last 12 months, no history of congenital lumbar or pelvic anomalies, and tested negative to the ASLR and standing hip flexion and neutral zone analysis tests.
B. Hungerford et al. / Clinical Biomechanics 19 (2004) 456–464
459
Subjects were excluded if no palpable motion between the PSIS and S2 spinous process was observed during unilateral hip flexion in standing, or if they experienced any pain during the clinical assessment. All subjects were assessed by the same experienced physiotherapist to maintain continuity. Informed consent was given by each subject prior to participation in the study, and all rights of the subjects were protected. The study was approved by the institutional Human Ethics Committee. 2.2. Procedure Fifteen lightweight highly reflective 15 mm diameter balls were used to define the bony landmarks of each innominate, both femoral segments, and the sacrum. The pelvic bony landmarks were chosen for their closeness to the skin surface with minimal overlying fascia, and because they reflected palpation points commonly used by therapists. Each innominate was therefore defined by three markers placed on the anterior superior iliac spine (ASIS), the PSIS, and the lateral iliac tubercle. A three armed triangular wand with a single marker attached to each arm was applied to the sacral spinous process of S2. Each arm was 10 mm long, and solidly welded to the triangular base. The apical marker of the wand approximated the S1 spinous process, while a left and right marker formed the base of the triangle at the horizontal level of S2. The left and right femoral segments were defined by markers on the greater trochanter, the lateral femoral condyle, and a mid thigh marker placed 20 cm inferior and 5 cm anterior to the greater trochanter. All markers were applied to the skin while the subjects were standing. A six camera Expert Vision Motion Analysis (Eva)e Hi Res.6.0 System (Motion Analysis Corporation, California, USA) was used to video (60 Hz) the subject motion. Measurements of a known angle showed the six camera Evae system was accurate to 0.25. Force platform data (960 Hz) were used to identify initiation of single leg support during each standing hip flexion trial. Following practice trials, data from one quiet standing trial were collected. The subject then performed six left and six right standing hip flexion trials. For example, during a left standing hip flexion trial the subject was asked to stand on his right leg, and flex his left hip and knee toward 90 hip flexion, then lower the foot back down. 2.3. Data analysis The Evae motion analysis system was used to track the 3D trajectories of each marker over time. These trajectories were then imported into Kintrake (Motion Analysis Corporation) which provided the 3D angular rotation, and translation, of each innominate relative to
Fig. 1. The three axes for angular and translational motion of the innominate relative to the sacral segment. Note the axis of innominate segment motion is centred at the PSIS.
the sacral segment throughout each trial, in respect to neutral position from the quiet standing trial. Calculation of the angular kinematics required definition of individual segment coordinate systems, and a joint coordinate system. The segment coordinate systems were assumed to be embedded within each adjacent segment. The axes of the innominate segment originated at each respective PSIS, and the axes of the sacral segment intercepted at the S2 spinous process (Fig. 1). Pelvic motion was determined by aligning the coronal axis of each innominate and sacral segment to intersect both the PSIS and S2 spinous process. Kintrake subsequently determined relative angular and translational motion of each bone segment about their aligned segment coordinate systems. Hip joint motion was determined by computing the motion of the femur, as defined by linking the three femoral markers, in relation to ipsilateral innominate motion. The hip joint centre was determined using the equation provided by Tylkowski et al. (1982). All angular and translational motion was determined at maximum coronal axis motion of the innominate, relative to the sacrum, during hip flexion. Two tailed paired Student t-tests assuming unequal variance (Domholdt, 1993) were performed for all variables between the left and right side in the control group, and between the symptomatic side and the asymptomatic side in the SIJ pain group. In order to determine if there was a significant alteration to the mean angular and translational intra-pelvic motion measured between the control subjects and the SIJ pain group, independent groups two tailed Student t-tests assuming unequal variance (Domholdt, 1993) were performed for all variables. Further graphical comparisons of coronal axis motion on the side of single leg support were
460
B. Hungerford et al. / Clinical Biomechanics 19 (2004) 456–464 12 10 8
coronal axis rotation(°) posterior anterior
6 4 2 0
-2 -4 -6 1
2
3
4
5
6
7
8
9
10
11
12
13
14
pelvic pain asymptomatic pelvic pain symptomatic
subjects
controls mean Fig. 2. A comparison of coronal axis angular motion at single leg support in each subject with posterior pelvic pain. Mean coronal axis motion for control subjects is depicted as a solid line. Error bars denote 2 · standard error for each subject. Significant at P < 0:05.
was measured during a standing hip flexion movement using a non-invasive method. Possible errors in determining the magnitude of pelvic bone motion may have occurred due to the movement of skin markers over bony landmarks, therefore the emphasis of this study was on changes to the patterns of pelvic bone motion as reflected by the measured skin marker movement.
performed for each subject in the SIJ pain group. Two times the standard error of mean motion for each subject was determined and plotted. As the mean ± 2 SE is approximately equal to the 95% confidence interval (Sim and Reid, 1999), determination of no overlay of each subject’s data was defined as a significant difference (Fig. 2).
3.1. Hip flexion side 3. Results The angular and translational motions of the left and right innominates, during left and right standing hip flexion movements are summarised in Tables 1–3. In the
The pattern of angular and translational motion of each innominate segment, relative to the sacral segment,
Table 1 Comparison of angular and translational motion of the left and right innominates on the side of hip flexion in control subjects and subjects with posterior pelvic pain Motion segment
Axis of motion
A: Control subjects
B: Posterior pelvic pain subjects
Left hip flexion
Right hip flexion
Mean
SD
Mean
P -value
SD
Asymptomatic hip flexion
Symptomatic hip flexion
Mean
SD
Mean
P -value
SD
Femur
Coronal
70.00
5.25
73.00
5.50
0.08
73.25
7.00
74.50
4.50
0.28
Innominate angular ()
Coronal Sagittal Vertical
)8.50 )6.00 4.50
3.50 3.50 4.50
)10.00 )7.75 3.50
3.50 5.00 3.00
0.05 0.18 0.39
)7.50 )5.00 3.75
3.00 5.50 2.50
)10.00 )6.25 4.75
3.25 4.00 1.75
0.04 0.44 0.27
Innominate translation (mm)
Antero-posterior Medio-lateral Vertical
3.50 )5.50 )6.50
2.50 3.00 3.00
4.00 )5.50 )7.50
3.00 2.50 3.50
0.49 0.74 0.46
2.00 )5.50 )4.00
4.00 2.00 3.50
2.00 )6.50 )5.00
3.75 3.00 2.00
0.70 0.36 0.76
Negative coronal value ¼ posterior; negative sagittal ¼ toward flexed hip; negative vertical ¼ toward flexed hip; negative antero-posterior ¼ posterior; negative medio-lateral translation ¼ toward flexed hip; negative vertical translation ¼ superior. * Significant at P < 0:05.
B. Hungerford et al. / Clinical Biomechanics 19 (2004) 456–464
461
Table 2 Comparison of angular and translational motion of the innominates on the side of single leg support in control subjects and subjects with posterior pelvic pain Motion segment
Axis of motion
Femur
Coronal
Innominate angular () Innominate translation (mm)
A: Control subjects
B: Posterior pelvic pain subjects
Left single leg support
Right single leg support
Mean
SD
Mean
SD
1.50
3.00
1.50
2.50
0.79
Coronal Sagittal Vertical
0.00 )6.00 3.50
2.00 2.50 2.75
)0.50 )5.75 4.50
3.00 2.50 4.50
Antero-posterior Medio-lateral Vertical
)4.50 )6.50 )4.50
6.00 3.25 2.50
)2.50 )6.50 )3.50
4.00 3.00 2.50
Axis
P -value
Asymptomatic single leg support
Symptomatic single leg support
P -value
Mean
SD
Mean
SD
1.75
3.50
1.75
3.25
0.86
0.88 0.46 0.17
0.50 )4.75 5.00
1.50 3.25 2.50
2.00 )3.75 3.75
2.00 3.00 2.75
0.02 0.42 0.09
0.13 0.96 0.19
)4.00 )7.00 3.50
2.50 3.25 2.25
)6.00 )5.00 2.00
2.75 2.50 3.00
0.02 0.03 0.04
Negative coronal value ¼ posterior; negative sagittal ¼ toward flexed hip; negative vertical ¼ toward flexed hip; negative A-P translation ¼ posterior; negative medio-lateral ¼ toward flexed hip; negative vertical ¼ superior.
Table 3 Comparison of angular and translational motion of the innominates between control subjects and subjects with posterior pelvic pain Motion segment
Axis of motion
Femur
Coronal
73.00
5.50
74.50
4.50
0.45
Innominate angular ()
Coronal Sagittal Vertical
)10.00 )7.75 3.50
3.50 5.00 3.00
)10.00 )6.25 4.75
3.25 4.00 1.75
Innominate translation (mm)
Antero-posterior Medio-lateral Vertical
4.00 )5.50 )7.50
3.00 2.50 3.50
2.00 )6.50 )5.00
3.75 3.00 2.00
Axis
A: Hip flexion
B: Single leg support
Controls right
Pelvic pain group symptomatic
Mean
Mean
SD
P -value
Controls right
Pelvic pain group symptomatic
Mean
SD
Mean
SD
1.50
2.50
1.75
3.25
0.67
0.16 0.94 0.46
)0.50 )5.75 4.50
3.00 2.50 4.50
2.00 )3.75 3.75
2.00 3.00 2.75
<0.01 0.12 0.90
0.05 0.94 0.04
)2.50 )6.50 )3.50
4.00 3.00 2.50
)6.00 )5.00 2.00
2.75 2.50 3.00
0.02 0.13 <0.001
SD
P -value
Negative coronal value ¼ posterior; negative sagittal ¼ toward flexed hip; negative vertical ¼ toward flexed hip; negative A-P translation ¼ posterior; negative medio-lateral ¼ toward flexed hip; negative vertical ¼ superior. * Significant at P < 0:05.
control subjects, standing hip flexion (Table 1A) produced posterior rotation of the non-weight bearing innominate relative to the sacral segment. On the side of hip flexion, femoral flexion showed a mean (range) 73.0 (10.0) at maximum coronal axis angular motion of the innominate. The innominate concurrently side flexed toward the side of hip flexion, and rotated about the vertical axis away from the side of hip flexion. Translation of the innominate relative to the sacral segment was associated with this angular motion. The innominate translated anteriorly, superiorly, and laterally (toward the side of hip flexion) as the innominate posteriorly rotated (Table 1A). No significant difference was found in the controls between the pattern of left and right innominate motion on the side of hip flexion. On the side of hip flexion for the posterior pelvic pain group, maximum posterior rotation of the innominate occurred at a mean of 73.25 (7.0) femoral flexion on the asymptomatic side, and at a mean of 74.5 (4.5) femoral flexion on the symptomatic side (Table 1B). A pattern of
side flexion of the innominate, and rotation about the vertical axis away from the side of hip flexion, occurred on both the asymptomatic and symptomatic sides during hip flexion. Anterior, superior, and lateral translation of the innominate toward the side of hip flexion, were associated with posterior rotation of the innominate on the side of hip flexion. In the posterior pelvic pain group there was no significant difference between the asymptomatic and symptomatic sides in translational motion, or angular motion of the innominate about the sagittal or vertical axes during hip flexion (Table 1B). 3.2. Single leg support The contralateral limb maintained single leg support during the standing hip flexion movement. Posterior rotation of the weight bearing innominate occurred about the coronal axis in control subjects (Table 2A). A concurrent pattern of side flexion of the innominate
462
B. Hungerford et al. / Clinical Biomechanics 19 (2004) 456–464
toward, and rotation about the vertical axis away from the side of hip flexion also occurred. The weight bearing innominate translated posteriorly, superiorly and medially (toward the side of hip flexion) (Table 2A). No significant difference was found in the pattern of either angular or translational motion of the weight bearing innominate during single leg support in control subjects. The angular and translational motion of the weight bearing innominate on the side of single leg support, in subjects with posterior pelvic pain, is depicted in Table 2B. The weight bearing innominate anteriorly rotated significantly more (P ¼ 0:02) about the coronal axis on the symptomatic in comparison to the asymptomatic side. Concurrently, the innominate side flexed toward, and rotated away from the side of hip flexion on both the symptomatic and asymptomatic sides. The innominate translated inferiorly, posteriorly, and medially (toward the side of hip flexion) relative to the sacral segment (Table 2B). Posterior translation of the innominate relative to the sacral segment was significantly greater (P ¼ 0:02) on the symptomatic side than the asymptomatic side, while inferior translation of the innominate was significantly less (P ¼ 0:04) on the symptomatic side in comparison to the asymptomatic side. Medial translation, or compression, of the innominate was also significantly less (P ¼ 0:03) on the symptomatic side in comparison to the asymptomatic side (Table 2B). Further comparison of coronal axis motion on the symptomatic side and asymptomatic side of single leg support was performed for each subject with posterior pelvic pain (Fig. 2). The mean range of posterior rotation on the side of single leg support in control subjects was depicted as a solid line for comparison to the symptomatic group. A significant change toward anterior rotation of the innominate occurred in 12 of the 14 posterior pelvic pain subjects on the symptomatic side; that is in subjects 1, 2, 4 to 7, 9 to 14 (Fig. 2). 3.3. Comparison of control subjects and subjects with posterior pelvic pain During the hip flexion component of the standing hip flexion movement there was no significant difference in patterning of angular or translational motion between groups on the side of hip flexion (Table 3A). On the side of single leg support, a significant difference in the pattern of angular and translational motion of the innominate between controls and symptomatic subjects was determined (Table 3B). In the control group, posterior rotation of the weight bearing innominate occurred on the side of single leg support; however, in the posterior pelvic pain group, anterior rotation occurred on the symptomatic side (P < 0:01). A pattern of posterior, superior and medial translation occurred concurrently with posterior rotation of the innominate
in control subjects; however, inferior, posterior, and medial translation occurred with anterior rotation of the innominate in the symptomatic subjects.
4. Discussion Significant differences were found in the pattern of pelvic bone motion that occurred during standing hip flexion when intra-subject (between weight bearing and non-weight bearing sides) and inter-subject (between the control group and posterior pelvic pain group) comparisons were made. The range of motion reported between the innominate and sacral segments, as determined using motion analysis, was generally larger than the range of motion determined during analysis of SIJ motion using stereophotogrammetry (Sturesson et al., 2000). This may be due to errors created by skin deformation over bone landmarks, or the movement of muscles close to the placement of the skin markers, such as gluteus maximus or obliquus abdominis internus, as they activated to create the test movement (Hungerford et al., 2003). 4.1. Intra-subject comparisons––control group In the control subjects hip motion toward 90 of femoral flexion produced posterior rotation of the nonweight bearing innominate (side of hip flexion), consistent with previous research (Jacob and Kissling, 1995; Sturesson et al., 2000). A concurrent pattern of side flexion toward the side of hip flexion, and rotation about the vertical axis away from the side of hip flexion was found. In addition, a concurrent translational motion occurred (anterior, superior and lateral) during hip flexion. This pattern of translation is consistent with the model of arthrokinematic motion of the SIJ proposed by Lee (1999). During this same movement, the weight bearing innominate (side of single leg support) posteriorly rotated; a finding consistent with previous research (Sturesson et al., 2000). In addition, this study found a concurrent side flexion of the innominate toward, and axial rotation away from the side of hip flexion. Posterior rotation of the innominate (or sacral nutation) is thought to occur as a consequence of the self-bracing mechanism of the pelvis; essential for optimal load transfer during single leg support (Snijders et al., 1993; Vleeming et al., 1989, 1995b). A concurrent translational motion occurred (posterior, superior and medial) during single leg support. Although both the non-weight bearing and weight bearing innominates posteriorly rotated during standing hip flexion, the pattern of the concurrent translation between the innominate and sacral segments on the non-weight bearing and weight bearing sides differed. This variation of translational motion
B. Hungerford et al. / Clinical Biomechanics 19 (2004) 456–464
may reflect different patterns of muscle activation and different compressive forces acting on the innominate during single leg support. Using the Doppler imaging system, Richardson et al. (2002) noted that a cocontraction of multifidus and transversus abdominis increased the stiffness of the SIJ. An optimal lumbopelvic stabilisation strategy requires recruitment of these muscles and could explain the compression (medial translation of the innominate) noted in the control subjects during single leg support. 4.2. Intra-subject comparisons––posterior pelvic pain group During hip flexion in the subjects with posterior pelvic pain, hip motion toward 90 of femoral flexion also produced posterior rotation of the non-weight bearing innominate. This pattern of motion was found on both the symptomatic and asymptomatic side. The pattern of side flexion, rotation, and translation of the non-weight bearing innominate did not differ between the asymptomatic and symptomatic sides. On the side of single leg support, anterior rotation of the weight bearing innominate occurred in subjects with posterior pelvic pain. This pattern of motion has been noted previously by Mens et al. (1999) during single leg loading in patients with posterior pelvic pain. In addition, this study found that the concurrent translational motion of the weight bearing innominate differed in pelvic pain patients. As the innominate rotated anteriorly it translated inferiorly and posteriorly; a pattern hypothesised to be consistent with less intra-pelvic compression (Lee, 1999). It is interesting to note that in this study, all of the subjects in the posterior pelvic pain group reported increased symptoms with vertical loading through the pelvis (standing, walking). This may suggest they were unable to adequately compress the weight bearing SIJ and maintain self-bracing of the pelvis (Snijders et al., 1998) in order to control vertical shear loads. 4.3. The two groups––inter-subject comparisons A comparison of both the angular and translational motion of the non-weight bearing innominate on the side of hip flexion between control subjects and matched subjects with posterior pelvic pain showed no significant difference. This has important clinical relevance. Clinical assessment of posterior rotation of the innominate on the side of hip flexion (the stork test) as a method of distinguishing normal joint motion from SIJ dysfunction (Mitchell, 1995) has been found to be unreliable and unspecific (Vincent-Smith and Gibbons, 1999). This study further validates such conclusions. A significant difference was noted in the pattern of angular motion of the weight bearing innominate during
463
single leg support when the control group (posterior rotation) was compared to the posterior pelvic pain group (anterior rotation). Vleeming et al. (1995b) suggest that anterior rotation of the innominate (sacral counternutation) disengages the self-bracing mechanism of the pelvis and consequently diminishes the ability to transfer loads between the spine and legs. The positive clinical tests noted in the posterior pelvic pain group may reflect their inability to posteriorly rotate the weight bearing innominate on the symptomatic side. Alternately, subjects may have lacked the ability to create sufficient compression due to altered motor control of the musculature known to stabilise the pelvis, such as transversus abdominis, lumbo-sacral multifidus, and gluteus maximus (Hungerford et al., 2003; Richardson et al., 2002).
5. Conclusions The most significant alteration to the pattern of bone motion between controls and subjects with posterior pelvic pain occurred on the side of single leg support. In the control subjects, the weight bearing innominate posteriorly rotated and translated superiorly, posteriorly and medial relative to the sacral segment. In the subjects with posterior pelvic pain, the weight bearing innominate anteriorly rotated and translated inferiorly. The results of this study suggest that posterior rotation of the innominate is a normal component for optimal stabilisation of the pelvis. Anterior rotation of the innominate is indicative of failure of the self-bracing mechanism and load transfer through the pelvis, with a resultant decrease in the ability to oppose vertical shear loads during weight bearing.
Acknowledgements The technical assistance of Ray Patton and Dr. Richard Smith from the School of Exercise and Sports Science, University of Sydney is acknowledged.
References Buyruk, H.M., Snijders, C.J., Vleeming, A., Lameris, J.S., Holland, W.P.J., Stam, H.J., 1995. The measurements of sacroiliac joint stiffness with colour Doppler imaging: a study on healthy subjects. European Journal of Radiology 21, 117–121. Buyruk, H.M., Snijders, C.J., Vleeming, A., Lameris, J.S., Holland, W.P.J., Stam, H.J., 1999. Measurements of sacroiliac joint stiffness in peripartum pelvic patients with Doppler imaging of vibrations. European Journal of Radiology 83, 159–163. Carmichael, J., 1987. Inter and intra-examiner reliability of palpation of the sacroiliac joint dysfunction. Journal of Manipulative and Physiological Therapeutics 10, 164–171.
464
B. Hungerford et al. / Clinical Biomechanics 19 (2004) 456–464
Damen, L., Buyruk, H.M., Guler-Ulysal, F., Lotgering, F.K., Snijders, C.J., Stam, H.J., 2001. Pelvic pain during pregnancy is associated with asymmetric laxity of the sacroiliac joints. Acta Obstetricia et Gynecologica Scandinavica 80, 1019–1024. Damen, L., Buyruk, H.M., Guler-Uysal, F., Lotgering, F.K., Snijders, C.J., Stam, H.J., 2002. Prognostic value of asymmetric laxity of the sacroiliac joints in pregnancy-related pelvic pain. Spine 27, 2820. Domholdt, E., 1993. Physical Therapy Research: Principles and Applications. W.B Saunders Co, Philadelphia. Fortin, J.D., Dwyer, A., West, S., Pier, J., 1994. Sacroiliac joint referral patterns upon application of a new injection/arthrography technique. I: Asymptomatic volunteers. Spine 19, 1475– 1482. Gracovetsky, S., Newman, N., Pawlowsky, M., Lanzo, V., Davey, B., Robinson, L., 1995. A database for estimating normal spinal motion derived from noninvasive measurements. Spine 20, 1036– 1046. Hodges, P.W., Richardson, C.A.., 1997. Contraction of the abdominal muscles associated with movement of the lower limb. Physical Therapy 77, 132–144. Hungerford, B., Gilleard, W., Hodges, P.W., 2003. Evidence of altered lumbo-pelvic muscle recruitment in the presence of sacroiliac joint pain. Spine 28, 1593–1600. Jacob, H., Kissling, R., 1995. The mobility of the sacroiliac joints in healthy volunteers between 20 and 50 years of age. Clinical Biomechanics 10, 352–361. Lee, D., 1999. The Pelvic Girdle: An Approach to Examination and Treatment of the Lumbo-Pelvic-Hip Region. Churchill Livingstone, Edinburgh. Maslen, B., Ackland, T., 1994. Radiographic study of skin displacement errors in the foot and ankle during standing. Clinical Biomechanics 9, 291–296. Mens, J.M.A., Vleeming, A., Snijders, C.J., Stam, H.J., Ginai, A.Z., 1999. The active straight leg raising test and mobility of the pelvic joints. European Spine Journal 8, 468–473. Mens, J.M., Vleeming, A., Snijders, C.J., Koes, B., Stam, H.J., 2001. Reliability and validity of the active straight leg raise test in posterior pelvic pain since pregnancy. Spine 26, 1167–1171. Mitchell, F.J., 1995. The Muscle Energy Manual, vol. 1. MET Press, East Lansing. O’Sullivan, P.B., Beales, D.J., Beetham, J.A., et al., 2002. Altered motor control strategies in subjects with sacroiliac joint pain during the active straight leg raise test. Spine 27, E1–E8. Panjabi, M.M., 1992. The stabilising system of the spine. Part 2: Neutral zone and stability hypothesis. Journal of Spinal Disorders 5, 390–397. Richardson, C.A., Snijders, C.J., Hides, J.A., Damen, L., Martijn, S.P., Storm, J., 2002. The relationship between transversus
abdominis muscles, sacroiliac joint mechanics, and low back pain. Spine 27, 399–405. Sim, J., Reid, N., 1999. Statistical inference by confidence intervals: issues of interpretation and utilization. Physical Therapy 79, 186– 195. Snijders, C.J., Ribbers, M.T., de Bakker, H.V., Stoeckart, R., Stam, H.J., 1998. EMG recordings of abdominal and back muscles in various standing postures: validation of a biomechanical model on sacroiliac joint stability. Journal of Electromyography and Kinesiology 8, 205–214. Snijders, C.J., Vleeming, A., Stoeckart, R., 1993. Transfer of lumbosacral load to iliac bones and legs. 1: Biomechanics of self-bracing of the sacroiliac joints and its significance for treatment and exercise. Clinical biomechanics 8, 285–294. Sturesson, B., Uden, A., Vleeming, A., 2000. A radiological analysis of movements of the sacroiliac joint during the standing hip flexion test. Spine 25, 364–368. Tylkowski, C.M., Simon, S.R., Mansour, J.M., 1982. Internal rotation gait in spastic cerebral palsy in the hip. Paper presented at the Proceedings of the 10th Open Scientific Meeting of the Hip Society, St Louis. pp. 89–125. Vincent-Smith, B., Gibbons, P., 1999. Inter-examiner and intraexaminer reliability of the standing hip flexion test. Manual Therapy 4, 87–93. Vleeming, A., de Vries, H.J., Mens, J.M., van Wingerden, J.P., 2002. Possible role of the long dorsal sacroiliac ligament in women with peripartum pelvic pain. Acta Obstetrics Gynecology Scandinavia 81, 430. Vleeming, A., Pool-Goudzwaard, A.L., Hammudoghlu, D., Stoeckart, R., Snijders, C.J., Mens, J.M., 1996. The function of the long dorsal sacroiliac ligament: its implication for understanding low back pain. Spine 21, 556–562. Vleeming, A., Pool-Goudzwaard, A.L., Stoeckart, R., van Wingerden, J.P., Snijders, C.J., 1995a. The posterior layer of the thoracolumbar fascia: its function in load transfer from spine to legs. Spine 20, 753–758. Vleeming, A., Snijders, C.J., Stoeckart, R., Mens, J.M., 1995b. A new light on low back pain. Paper presented at the 2nd Interdisciplinary World Congress on Low Back Pain, San Diego, pp. 149–168. Vleeming, A., van Wingerden, J.P., Snijders, C.J., Stoeckart, R., Stijnen, T., 1989. Load application to the sacrotuberous ligament: influences on sacroiliac joint mechanics. Clinical Biomechanics 4, 204–209. Vleeming, A., Volkers, A.C.W., Snijders, C.J., Stoeckart, R., 1990. Relation between form and function in the sacroiliac joint. 2: Biomechanical aspects. Spine 15, 133–136. Walheim, G.G., Selvik, G., 1984. Mobility of the pubic symphysis. Clinical Orthopaedics and related Research 191, 129–135.
Altered patterns of pelvic bone motion determined in subjects with posterior pelvic pain using skin markers Barbara Hungerford
a,*
, Wendy Gilleard b, Diane Lee
c
a
b
School of Exercise and Sport Science, University of Sydney, Sydney, Australia School of Exercise Science and Sport Management, Southern Cross University, Lismore, Australia c Diane G. Lee Physiotherapist Corp., BC, Canada Received 14 February 2003; accepted 19 February 2004
Abstract Objective. To determine whether the pattern of pelvic bone motion, determined by skin markers, differs between control subjects and subjects with posterior pelvic pain. Design. Cross-sectional study of three-dimensional angular and translational motion of the innominates relative to the sacrum in two subject groups. Background. Comparative in vivo analysis of the 3D patterning of pelvic motion in subjects with posterior pelvic pain and controls is limited. Methods. Fourteen males with posterior pelvic pain and healthy age and height matched controls were studied. A 6-camera motion analysis system was used to determine 3D angular and translational motion of pelvic skin markers during standing hip flexion. Results. Posterior rotation of the innominate occurred with hip flexion in control subjects and pelvic pain subjects as previously reported in the literature. On the supporting leg, the innominate rotated posteriorly in controls and anteriorly in symptomatic subjects. Conclusion. Posterior rotation of the innominate, as measured using skin markers during weight bearing in controls may reflect activation of optimal lumbo-pelvic stabilisation strategies for load transfer. Anterior rotation occurred in symptomatic subjects, suggesting failure to stabilise intra-pelvic motion for load transfer. Relevance This study found that posterior rotation of the innominate occurred during weight bearing in controls. This movement pattern is thought to optimise stability of the pelvic girdle during increased loading. Conversely, anterior rotation occurred in symptomatic subjects during weight bearing. This is a non-optimal pattern and may indicate abnormal articular or neuromyofascial function during increased vertical loading through the pelvis. 2004 Elsevier Ltd. All rights reserved. Keywords: Sacroiliac joint; Pelvic motion; Pelvic stabilisation; Low back pain; Pelvic pain
1. Introduction A primary function of the lumbar spine and pelvis is to transfer the loads generated by body weight and gravity during standing, walking and sitting (Snijders et al., 1993). How well this load is managed dictates the efficacy of function. A small amount of motion occurs at
*
Corresponding author. Present address: Sydney Spine and Pelvis Centre, 101 Lyons Road, Drummoyne, NSW 2047, Australia. E-mail address: [email protected] (B. Hungerford). 0268-0033/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2004.02.004
the sacroiliac joints (SIJ) and the pubic symphysis during movements of the trunk and lower limbs (Jacob and Kissling, 1995; Walheim and Selvik, 1984). Consequently, during weight bearing activities, control (stabilisation) of intra-pelvic motion is required for transference of loads between the spine and the lower limbs (Snijders et al., 1993; Vleeming et al., 1990). According to Panjabi (1992) stability is achieved when the passive, active and control systems work together. Snijders et al. (1993) suggests that the passive, active and control systems produce approximation of the joint surfaces, essential if stability is to be insured. The
B. Hungerford et al. / Clinical Biomechanics 19 (2004) 456–464
amount of approximation required is variable and difficult to quantify as it is dependent on an individual’s structure (form closure) and the forces they need to control (force closure). The ability to effectively transfer load through the pelvis is dynamic and therefore depends on: (1) optimal function of the bones, joints and ligaments (Vleeming et al., 1989, 1990); (2) optimal function of the muscles and fascia (Hungerford et al., 2003; Richardson et al., 2002; Snijders et al., 1998; Vleeming et al., 1995a,b); (3) appropriate neural function (Hodges and Richardson, 1997; Hungerford et al., 2003). 1.1. Stabilisation and motion control of the pelvic girdle For every joint, there is a position called the selfbraced (close-packed) position in which there is maximum congruence of the articular surfaces and maximum tension on major ligaments. In this position, the joint is under significant compression and the ability to resist shear forces is enhanced by tensioning of the passive structures and increased friction between the articular surfaces (Snijders et al., 1993; Vleeming et al., 1990). The self-braced position of the SIJ is nutation of the sacrum or posterior rotation of the innominate (Vleeming et al., 1989). Studies have shown (Sturesson et al., 2000) that nutation of the sacrum relative to posterior rotation of the innominate occurs bilaterally whenever the lumbo-pelvic spine is loaded vertically (sitting, standing). Counternutation of the sacrum, or anterior rotation of the innominate, is thought to be a relatively less stable position for the SIJ (Vleeming et al., 1995b). The long dorsal ligament becomes taut during this motion, however tension in other ligaments such as sacrotuberous and interosseous ligaments decreases (Vleeming et al., 1996). At present there is only limited research comparing the in vivo pattern of pelvic motion during weight bearing and non-weight bearing activities. This is due to the invasiveness of the most reliable and valid methods of analysis, consequential ethical considerations for using invasive procedures to evaluate large number of subjects and difficulty in evaluating the small amplitude of in vivo SIJ motion. When an individual stands on one leg and flexes the contralateral hip, the non-weight bearing innominate posteriorly rotates relative to the sacrum (range of motion 0.1–5.0) (Jacob and Kissling, 1995; Sturesson et al., 2000). Side flexion and axial rotation of the innominate occur concurrently about sagittal and vertical axes respectively. Translation (motion along the sagittal, vertical and/or coronal axes) also occurs (Jacob and Kissling, 1995; Sturesson et al., 2000), although the specific direction of translation that occurs with angular motion is not fully understood. It has been hypothesised (Lee, 1999) that anterior and superior translation of the innominate, relative to the
457
sacrum, will occur with posterior rotation of the innominate. 1.2. Range of motion versus patterning of pelvic motion Buyruk et al. (1995) and Damen et al. (2002) established that a Doppler imaging system was able to measure stiffness of the SIJ. This research showed that stiffness of the SIJ is variable between subjects and therefore the range of motion is potentially variable. It also revealed that stiffness of right and left SIJs is symmetric in subjects without pelvic pain. The results support putting less emphasis on amplitude of motion and more on the pattern or symmetry of SIJ, or pelvic motion, as range of motion varies between subjects, however within one subject joint stiffness remains symmetric between sides. 1.3. The impact of posterior pelvic pain The SIJs and the posterior SIJ ligaments are a known source of posterior pelvic pain (Fortin et al., 1994; Vleeming et al., 2002). Jacob and Kissling (1995) noted that in the presence of SIJ symptoms, the amplitude of SIJ motion about a coronal axis increased during hip flexion in one subject. Mens et al. (1999) determined increased amplitude of anterior rotation of the innominate in posterior pelvic pain patients. Sturesson et al. (2000) reported no difference in the amplitude or pattern of either angular or translational motion of the innominates when the left and right SIJs were compared in subjects with posterior pelvic pain. No comparison was made with an asymptomatic group. Hungerford et al. (2003) showed that posterior pelvic pain alters the pattern of lumbo-pelvic muscle recruitment, while Buyruk et al. (1999) and Damen et al. (2002) showed that stiffness of the SIJ is asymmetric in subjects with pelvic pain and that asymmetrical stiffness of the SIJs is prognostic for pelvic impairment and pain. It is presently unknown if posterior pelvic pain alters the patterning of bone motion within the pelvis for single leg stance. 1.4. Posterior pelvic pain and the active straight leg raise test The supine active straight leg raise test (ASLR) (Mens et al., 2001) has been validated as a clinical test for measuring effective load transfer between the trunk and lower limbs. When the lumbo-pelvic region is functioning optimally, the leg should rise effortlessly from the table (effort graded from 0 to 5) (Mens et al., 1999). A correlation has been shown between positive ASLR findings and posterior pelvic pain (Mens et al., 1999; O’Sullivan et al., 2002). Similarly, Damen et al.
458
B. Hungerford et al. / Clinical Biomechanics 19 (2004) 456–464
(2001) and Buyruk et al. (1999) showed that the ASLR is positive in the presence of asymmetric stiffness of the SIJ. This suggests that altered pelvic stabilisation strategies may affect pelvic mobility (Buyruk et al., 1999; Damen et al., 2001). 1.5. Limitations of methodology in this study A non-invasive motion analysis system using skin mounted markers was chosen to acquire the kinematic data of pelvic bone motion during a standing hip flexion movement for ethical reasons. Errors in determining the range of motion are likely to occur with an optoelectronic system due to skin marker motion relative to underlying bony landmarks (Maslen and Ackland, 1994). A high resolution motion analysis system has however been reported to provide reliable and consistent in vivo data of lumbar segmental motion patterns (Gracovetsky et al., 1995). In this study, the authors recognise that the movements noted reflect motion of the innominates, sacrum, and femurs in conjunction with overlying skin, and therefore the main emphasis of this study was to investigate the patterns of bone motion rather than the range of motion. The aim of this study was to determine the threedimensional pattern of innominate bone motion occurring in subjects determined to have posterior pelvic pain and impaired pelvic stabilisation strategies during weight bearing and non-weight bearing components of a standing hip flexion movement. These results were compared to age and height matched controls with clinically assessed normal pelvic stabilisation and pelvic motion patterns. It was hypothesised that the pattern of innominate bone motion would alter in subjects with posterior pelvic pain during both components of the movement trial.
2. Methods 2.1. Subjects 2.1.1. Impaired pelvic stabilisation and posterior pelvic pain group Fourteen male subjects with SIJ pain and a mean (range) age, height and weight of 32.7 (24–47) years, 176.8 (168–184) cm, and 77.0 (71–90) kg respectively, volunteered for the study. The criteria for inclusion in this study were: 1. Each subject in the posterior pelvic pain group reported unilateral pain over the posterior pelvic/SI region (Fortin et al., 1994) for greater than two months, and no pain above the lumbo-sacral junction. The pain was consistently and predictably
aggravated by activities that vertically loaded the pelvis (walking, standing or sitting). 2. Positive results on the side of posterior pelvic pain in clinical tests for impaired lumbo-pelvic stabilisation. These tests included: (a) Active straight leg raise test (Mens et al., 1999, 2001): A positive test was indicated when the pelvis failed to remain in neutral alignment, and the subject reported difficulty or inability to elevate a straight leg in supine. The perceived difference of effort, or pain aggravation was scaled from 0 (not difficult to raise the leg) to 5 (unable to perform ASLR). (b) Standing hip flexion test (Mitchell, 1995): During a left standing hip flexion test, the subject stands on their right leg and flexes the left hip towards 90. The left innominate should posteriorly rotate relative to the sacrum (Jacob and Kissling, 1995). A positive test was indicated when superior motion of the posterior superior iliac spine (PSIS) was palpated relative to the sacrum. (c) Neutral zone analysis test (joint play) (Lee, 1999)––this test was used clinically to apply the research of Buyruk et al. (1999) and Damen et al. (2002) and to evaluate motion in the neutral zone of the SIJ. Panjabi (1992) noted that joints have non-linear load–displacement curves and that the size of the neutral zone may increase with injury, articular degeneration and/or weakness of the stabilising musculature and that this is a more sensitive indicator than angular range of motion for detecting instability. All symptomatic subjects demonstrated asymmetric stiffness of the SIJ when the innominate was glided relative to the sacrum (analysis of the neutral zone). As the reliability and predictive ability of the standing hip flexion and neutral zone analysis tests remains uncertain (Carmichael, 1987; Vincent-Smith and Gibbons, 1999) all clinical tests were required to be positive on the side of pain, in conjunction with the ASLR test, for inclusion in the posterior pelvic pain subject group. Subjects were excluded from the study if they could not flex each hip to 90 without pain, if they had undergone spinal surgery, or displayed overt neurological signs such as sensory paraesthesia or motor paresis. 2.1.2. Control group The SIJ pain group were age and height matched to a control group of 14 males with a mean (range) age, height, and weight of 33.5 (22–50) years, 176.0 (168–183) cm, and 72.5 (61–85) kg respectively. The control subjects had no history of low back pain in the last 12 months, no history of congenital lumbar or pelvic anomalies, and tested negative to the ASLR and standing hip flexion and neutral zone analysis tests.
B. Hungerford et al. / Clinical Biomechanics 19 (2004) 456–464
459
Subjects were excluded if no palpable motion between the PSIS and S2 spinous process was observed during unilateral hip flexion in standing, or if they experienced any pain during the clinical assessment. All subjects were assessed by the same experienced physiotherapist to maintain continuity. Informed consent was given by each subject prior to participation in the study, and all rights of the subjects were protected. The study was approved by the institutional Human Ethics Committee. 2.2. Procedure Fifteen lightweight highly reflective 15 mm diameter balls were used to define the bony landmarks of each innominate, both femoral segments, and the sacrum. The pelvic bony landmarks were chosen for their closeness to the skin surface with minimal overlying fascia, and because they reflected palpation points commonly used by therapists. Each innominate was therefore defined by three markers placed on the anterior superior iliac spine (ASIS), the PSIS, and the lateral iliac tubercle. A three armed triangular wand with a single marker attached to each arm was applied to the sacral spinous process of S2. Each arm was 10 mm long, and solidly welded to the triangular base. The apical marker of the wand approximated the S1 spinous process, while a left and right marker formed the base of the triangle at the horizontal level of S2. The left and right femoral segments were defined by markers on the greater trochanter, the lateral femoral condyle, and a mid thigh marker placed 20 cm inferior and 5 cm anterior to the greater trochanter. All markers were applied to the skin while the subjects were standing. A six camera Expert Vision Motion Analysis (Eva)e Hi Res.6.0 System (Motion Analysis Corporation, California, USA) was used to video (60 Hz) the subject motion. Measurements of a known angle showed the six camera Evae system was accurate to 0.25. Force platform data (960 Hz) were used to identify initiation of single leg support during each standing hip flexion trial. Following practice trials, data from one quiet standing trial were collected. The subject then performed six left and six right standing hip flexion trials. For example, during a left standing hip flexion trial the subject was asked to stand on his right leg, and flex his left hip and knee toward 90 hip flexion, then lower the foot back down. 2.3. Data analysis The Evae motion analysis system was used to track the 3D trajectories of each marker over time. These trajectories were then imported into Kintrake (Motion Analysis Corporation) which provided the 3D angular rotation, and translation, of each innominate relative to
Fig. 1. The three axes for angular and translational motion of the innominate relative to the sacral segment. Note the axis of innominate segment motion is centred at the PSIS.
the sacral segment throughout each trial, in respect to neutral position from the quiet standing trial. Calculation of the angular kinematics required definition of individual segment coordinate systems, and a joint coordinate system. The segment coordinate systems were assumed to be embedded within each adjacent segment. The axes of the innominate segment originated at each respective PSIS, and the axes of the sacral segment intercepted at the S2 spinous process (Fig. 1). Pelvic motion was determined by aligning the coronal axis of each innominate and sacral segment to intersect both the PSIS and S2 spinous process. Kintrake subsequently determined relative angular and translational motion of each bone segment about their aligned segment coordinate systems. Hip joint motion was determined by computing the motion of the femur, as defined by linking the three femoral markers, in relation to ipsilateral innominate motion. The hip joint centre was determined using the equation provided by Tylkowski et al. (1982). All angular and translational motion was determined at maximum coronal axis motion of the innominate, relative to the sacrum, during hip flexion. Two tailed paired Student t-tests assuming unequal variance (Domholdt, 1993) were performed for all variables between the left and right side in the control group, and between the symptomatic side and the asymptomatic side in the SIJ pain group. In order to determine if there was a significant alteration to the mean angular and translational intra-pelvic motion measured between the control subjects and the SIJ pain group, independent groups two tailed Student t-tests assuming unequal variance (Domholdt, 1993) were performed for all variables. Further graphical comparisons of coronal axis motion on the side of single leg support were
460
B. Hungerford et al. / Clinical Biomechanics 19 (2004) 456–464 12 10 8
coronal axis rotation(°) posterior anterior
6 4 2 0
-2 -4 -6 1
2
3
4
5
6
7
8
9
10
11
12
13
14
pelvic pain asymptomatic pelvic pain symptomatic
subjects
controls mean Fig. 2. A comparison of coronal axis angular motion at single leg support in each subject with posterior pelvic pain. Mean coronal axis motion for control subjects is depicted as a solid line. Error bars denote 2 · standard error for each subject. Significant at P < 0:05.
was measured during a standing hip flexion movement using a non-invasive method. Possible errors in determining the magnitude of pelvic bone motion may have occurred due to the movement of skin markers over bony landmarks, therefore the emphasis of this study was on changes to the patterns of pelvic bone motion as reflected by the measured skin marker movement.
performed for each subject in the SIJ pain group. Two times the standard error of mean motion for each subject was determined and plotted. As the mean ± 2 SE is approximately equal to the 95% confidence interval (Sim and Reid, 1999), determination of no overlay of each subject’s data was defined as a significant difference (Fig. 2).
3.1. Hip flexion side 3. Results The angular and translational motions of the left and right innominates, during left and right standing hip flexion movements are summarised in Tables 1–3. In the
The pattern of angular and translational motion of each innominate segment, relative to the sacral segment,
Table 1 Comparison of angular and translational motion of the left and right innominates on the side of hip flexion in control subjects and subjects with posterior pelvic pain Motion segment
Axis of motion
A: Control subjects
B: Posterior pelvic pain subjects
Left hip flexion
Right hip flexion
Mean
SD
Mean
P -value
SD
Asymptomatic hip flexion
Symptomatic hip flexion
Mean
SD
Mean
P -value
SD
Femur
Coronal
70.00
5.25
73.00
5.50
0.08
73.25
7.00
74.50
4.50
0.28
Innominate angular ()
Coronal Sagittal Vertical
)8.50 )6.00 4.50
3.50 3.50 4.50
)10.00 )7.75 3.50
3.50 5.00 3.00
0.05 0.18 0.39
)7.50 )5.00 3.75
3.00 5.50 2.50
)10.00 )6.25 4.75
3.25 4.00 1.75
0.04 0.44 0.27
Innominate translation (mm)
Antero-posterior Medio-lateral Vertical
3.50 )5.50 )6.50
2.50 3.00 3.00
4.00 )5.50 )7.50
3.00 2.50 3.50
0.49 0.74 0.46
2.00 )5.50 )4.00
4.00 2.00 3.50
2.00 )6.50 )5.00
3.75 3.00 2.00
0.70 0.36 0.76
Negative coronal value ¼ posterior; negative sagittal ¼ toward flexed hip; negative vertical ¼ toward flexed hip; negative antero-posterior ¼ posterior; negative medio-lateral translation ¼ toward flexed hip; negative vertical translation ¼ superior. * Significant at P < 0:05.
B. Hungerford et al. / Clinical Biomechanics 19 (2004) 456–464
461
Table 2 Comparison of angular and translational motion of the innominates on the side of single leg support in control subjects and subjects with posterior pelvic pain Motion segment
Axis of motion
Femur
Coronal
Innominate angular () Innominate translation (mm)
A: Control subjects
B: Posterior pelvic pain subjects
Left single leg support
Right single leg support
Mean
SD
Mean
SD
1.50
3.00
1.50
2.50
0.79
Coronal Sagittal Vertical
0.00 )6.00 3.50
2.00 2.50 2.75
)0.50 )5.75 4.50
3.00 2.50 4.50
Antero-posterior Medio-lateral Vertical
)4.50 )6.50 )4.50
6.00 3.25 2.50
)2.50 )6.50 )3.50
4.00 3.00 2.50
Axis
P -value
Asymptomatic single leg support
Symptomatic single leg support
P -value
Mean
SD
Mean
SD
1.75
3.50
1.75
3.25
0.86
0.88 0.46 0.17
0.50 )4.75 5.00
1.50 3.25 2.50
2.00 )3.75 3.75
2.00 3.00 2.75
0.02 0.42 0.09
0.13 0.96 0.19
)4.00 )7.00 3.50
2.50 3.25 2.25
)6.00 )5.00 2.00
2.75 2.50 3.00
0.02 0.03 0.04
Negative coronal value ¼ posterior; negative sagittal ¼ toward flexed hip; negative vertical ¼ toward flexed hip; negative A-P translation ¼ posterior; negative medio-lateral ¼ toward flexed hip; negative vertical ¼ superior.
Table 3 Comparison of angular and translational motion of the innominates between control subjects and subjects with posterior pelvic pain Motion segment
Axis of motion
Femur
Coronal
73.00
5.50
74.50
4.50
0.45
Innominate angular ()
Coronal Sagittal Vertical
)10.00 )7.75 3.50
3.50 5.00 3.00
)10.00 )6.25 4.75
3.25 4.00 1.75
Innominate translation (mm)
Antero-posterior Medio-lateral Vertical
4.00 )5.50 )7.50
3.00 2.50 3.50
2.00 )6.50 )5.00
3.75 3.00 2.00
Axis
A: Hip flexion
B: Single leg support
Controls right
Pelvic pain group symptomatic
Mean
Mean
SD
P -value
Controls right
Pelvic pain group symptomatic
Mean
SD
Mean
SD
1.50
2.50
1.75
3.25
0.67
0.16 0.94 0.46
)0.50 )5.75 4.50
3.00 2.50 4.50
2.00 )3.75 3.75
2.00 3.00 2.75
<0.01 0.12 0.90
0.05 0.94 0.04
)2.50 )6.50 )3.50
4.00 3.00 2.50
)6.00 )5.00 2.00
2.75 2.50 3.00
0.02 0.13 <0.001
SD
P -value
Negative coronal value ¼ posterior; negative sagittal ¼ toward flexed hip; negative vertical ¼ toward flexed hip; negative A-P translation ¼ posterior; negative medio-lateral ¼ toward flexed hip; negative vertical ¼ superior. * Significant at P < 0:05.
control subjects, standing hip flexion (Table 1A) produced posterior rotation of the non-weight bearing innominate relative to the sacral segment. On the side of hip flexion, femoral flexion showed a mean (range) 73.0 (10.0) at maximum coronal axis angular motion of the innominate. The innominate concurrently side flexed toward the side of hip flexion, and rotated about the vertical axis away from the side of hip flexion. Translation of the innominate relative to the sacral segment was associated with this angular motion. The innominate translated anteriorly, superiorly, and laterally (toward the side of hip flexion) as the innominate posteriorly rotated (Table 1A). No significant difference was found in the controls between the pattern of left and right innominate motion on the side of hip flexion. On the side of hip flexion for the posterior pelvic pain group, maximum posterior rotation of the innominate occurred at a mean of 73.25 (7.0) femoral flexion on the asymptomatic side, and at a mean of 74.5 (4.5) femoral flexion on the symptomatic side (Table 1B). A pattern of
side flexion of the innominate, and rotation about the vertical axis away from the side of hip flexion, occurred on both the asymptomatic and symptomatic sides during hip flexion. Anterior, superior, and lateral translation of the innominate toward the side of hip flexion, were associated with posterior rotation of the innominate on the side of hip flexion. In the posterior pelvic pain group there was no significant difference between the asymptomatic and symptomatic sides in translational motion, or angular motion of the innominate about the sagittal or vertical axes during hip flexion (Table 1B). 3.2. Single leg support The contralateral limb maintained single leg support during the standing hip flexion movement. Posterior rotation of the weight bearing innominate occurred about the coronal axis in control subjects (Table 2A). A concurrent pattern of side flexion of the innominate
462
B. Hungerford et al. / Clinical Biomechanics 19 (2004) 456–464
toward, and rotation about the vertical axis away from the side of hip flexion also occurred. The weight bearing innominate translated posteriorly, superiorly and medially (toward the side of hip flexion) (Table 2A). No significant difference was found in the pattern of either angular or translational motion of the weight bearing innominate during single leg support in control subjects. The angular and translational motion of the weight bearing innominate on the side of single leg support, in subjects with posterior pelvic pain, is depicted in Table 2B. The weight bearing innominate anteriorly rotated significantly more (P ¼ 0:02) about the coronal axis on the symptomatic in comparison to the asymptomatic side. Concurrently, the innominate side flexed toward, and rotated away from the side of hip flexion on both the symptomatic and asymptomatic sides. The innominate translated inferiorly, posteriorly, and medially (toward the side of hip flexion) relative to the sacral segment (Table 2B). Posterior translation of the innominate relative to the sacral segment was significantly greater (P ¼ 0:02) on the symptomatic side than the asymptomatic side, while inferior translation of the innominate was significantly less (P ¼ 0:04) on the symptomatic side in comparison to the asymptomatic side. Medial translation, or compression, of the innominate was also significantly less (P ¼ 0:03) on the symptomatic side in comparison to the asymptomatic side (Table 2B). Further comparison of coronal axis motion on the symptomatic side and asymptomatic side of single leg support was performed for each subject with posterior pelvic pain (Fig. 2). The mean range of posterior rotation on the side of single leg support in control subjects was depicted as a solid line for comparison to the symptomatic group. A significant change toward anterior rotation of the innominate occurred in 12 of the 14 posterior pelvic pain subjects on the symptomatic side; that is in subjects 1, 2, 4 to 7, 9 to 14 (Fig. 2). 3.3. Comparison of control subjects and subjects with posterior pelvic pain During the hip flexion component of the standing hip flexion movement there was no significant difference in patterning of angular or translational motion between groups on the side of hip flexion (Table 3A). On the side of single leg support, a significant difference in the pattern of angular and translational motion of the innominate between controls and symptomatic subjects was determined (Table 3B). In the control group, posterior rotation of the weight bearing innominate occurred on the side of single leg support; however, in the posterior pelvic pain group, anterior rotation occurred on the symptomatic side (P < 0:01). A pattern of posterior, superior and medial translation occurred concurrently with posterior rotation of the innominate
in control subjects; however, inferior, posterior, and medial translation occurred with anterior rotation of the innominate in the symptomatic subjects.
4. Discussion Significant differences were found in the pattern of pelvic bone motion that occurred during standing hip flexion when intra-subject (between weight bearing and non-weight bearing sides) and inter-subject (between the control group and posterior pelvic pain group) comparisons were made. The range of motion reported between the innominate and sacral segments, as determined using motion analysis, was generally larger than the range of motion determined during analysis of SIJ motion using stereophotogrammetry (Sturesson et al., 2000). This may be due to errors created by skin deformation over bone landmarks, or the movement of muscles close to the placement of the skin markers, such as gluteus maximus or obliquus abdominis internus, as they activated to create the test movement (Hungerford et al., 2003). 4.1. Intra-subject comparisons––control group In the control subjects hip motion toward 90 of femoral flexion produced posterior rotation of the nonweight bearing innominate (side of hip flexion), consistent with previous research (Jacob and Kissling, 1995; Sturesson et al., 2000). A concurrent pattern of side flexion toward the side of hip flexion, and rotation about the vertical axis away from the side of hip flexion was found. In addition, a concurrent translational motion occurred (anterior, superior and lateral) during hip flexion. This pattern of translation is consistent with the model of arthrokinematic motion of the SIJ proposed by Lee (1999). During this same movement, the weight bearing innominate (side of single leg support) posteriorly rotated; a finding consistent with previous research (Sturesson et al., 2000). In addition, this study found a concurrent side flexion of the innominate toward, and axial rotation away from the side of hip flexion. Posterior rotation of the innominate (or sacral nutation) is thought to occur as a consequence of the self-bracing mechanism of the pelvis; essential for optimal load transfer during single leg support (Snijders et al., 1993; Vleeming et al., 1989, 1995b). A concurrent translational motion occurred (posterior, superior and medial) during single leg support. Although both the non-weight bearing and weight bearing innominates posteriorly rotated during standing hip flexion, the pattern of the concurrent translation between the innominate and sacral segments on the non-weight bearing and weight bearing sides differed. This variation of translational motion
B. Hungerford et al. / Clinical Biomechanics 19 (2004) 456–464
may reflect different patterns of muscle activation and different compressive forces acting on the innominate during single leg support. Using the Doppler imaging system, Richardson et al. (2002) noted that a cocontraction of multifidus and transversus abdominis increased the stiffness of the SIJ. An optimal lumbopelvic stabilisation strategy requires recruitment of these muscles and could explain the compression (medial translation of the innominate) noted in the control subjects during single leg support. 4.2. Intra-subject comparisons––posterior pelvic pain group During hip flexion in the subjects with posterior pelvic pain, hip motion toward 90 of femoral flexion also produced posterior rotation of the non-weight bearing innominate. This pattern of motion was found on both the symptomatic and asymptomatic side. The pattern of side flexion, rotation, and translation of the non-weight bearing innominate did not differ between the asymptomatic and symptomatic sides. On the side of single leg support, anterior rotation of the weight bearing innominate occurred in subjects with posterior pelvic pain. This pattern of motion has been noted previously by Mens et al. (1999) during single leg loading in patients with posterior pelvic pain. In addition, this study found that the concurrent translational motion of the weight bearing innominate differed in pelvic pain patients. As the innominate rotated anteriorly it translated inferiorly and posteriorly; a pattern hypothesised to be consistent with less intra-pelvic compression (Lee, 1999). It is interesting to note that in this study, all of the subjects in the posterior pelvic pain group reported increased symptoms with vertical loading through the pelvis (standing, walking). This may suggest they were unable to adequately compress the weight bearing SIJ and maintain self-bracing of the pelvis (Snijders et al., 1998) in order to control vertical shear loads. 4.3. The two groups––inter-subject comparisons A comparison of both the angular and translational motion of the non-weight bearing innominate on the side of hip flexion between control subjects and matched subjects with posterior pelvic pain showed no significant difference. This has important clinical relevance. Clinical assessment of posterior rotation of the innominate on the side of hip flexion (the stork test) as a method of distinguishing normal joint motion from SIJ dysfunction (Mitchell, 1995) has been found to be unreliable and unspecific (Vincent-Smith and Gibbons, 1999). This study further validates such conclusions. A significant difference was noted in the pattern of angular motion of the weight bearing innominate during
463
single leg support when the control group (posterior rotation) was compared to the posterior pelvic pain group (anterior rotation). Vleeming et al. (1995b) suggest that anterior rotation of the innominate (sacral counternutation) disengages the self-bracing mechanism of the pelvis and consequently diminishes the ability to transfer loads between the spine and legs. The positive clinical tests noted in the posterior pelvic pain group may reflect their inability to posteriorly rotate the weight bearing innominate on the symptomatic side. Alternately, subjects may have lacked the ability to create sufficient compression due to altered motor control of the musculature known to stabilise the pelvis, such as transversus abdominis, lumbo-sacral multifidus, and gluteus maximus (Hungerford et al., 2003; Richardson et al., 2002).
5. Conclusions The most significant alteration to the pattern of bone motion between controls and subjects with posterior pelvic pain occurred on the side of single leg support. In the control subjects, the weight bearing innominate posteriorly rotated and translated superiorly, posteriorly and medial relative to the sacral segment. In the subjects with posterior pelvic pain, the weight bearing innominate anteriorly rotated and translated inferiorly. The results of this study suggest that posterior rotation of the innominate is a normal component for optimal stabilisation of the pelvis. Anterior rotation of the innominate is indicative of failure of the self-bracing mechanism and load transfer through the pelvis, with a resultant decrease in the ability to oppose vertical shear loads during weight bearing.
Acknowledgements The technical assistance of Ray Patton and Dr. Richard Smith from the School of Exercise and Sports Science, University of Sydney is acknowledged.
References Buyruk, H.M., Snijders, C.J., Vleeming, A., Lameris, J.S., Holland, W.P.J., Stam, H.J., 1995. The measurements of sacroiliac joint stiffness with colour Doppler imaging: a study on healthy subjects. European Journal of Radiology 21, 117–121. Buyruk, H.M., Snijders, C.J., Vleeming, A., Lameris, J.S., Holland, W.P.J., Stam, H.J., 1999. Measurements of sacroiliac joint stiffness in peripartum pelvic patients with Doppler imaging of vibrations. European Journal of Radiology 83, 159–163. Carmichael, J., 1987. Inter and intra-examiner reliability of palpation of the sacroiliac joint dysfunction. Journal of Manipulative and Physiological Therapeutics 10, 164–171.
464
B. Hungerford et al. / Clinical Biomechanics 19 (2004) 456–464
Damen, L., Buyruk, H.M., Guler-Ulysal, F., Lotgering, F.K., Snijders, C.J., Stam, H.J., 2001. Pelvic pain during pregnancy is associated with asymmetric laxity of the sacroiliac joints. Acta Obstetricia et Gynecologica Scandinavica 80, 1019–1024. Damen, L., Buyruk, H.M., Guler-Uysal, F., Lotgering, F.K., Snijders, C.J., Stam, H.J., 2002. Prognostic value of asymmetric laxity of the sacroiliac joints in pregnancy-related pelvic pain. Spine 27, 2820. Domholdt, E., 1993. Physical Therapy Research: Principles and Applications. W.B Saunders Co, Philadelphia. Fortin, J.D., Dwyer, A., West, S., Pier, J., 1994. Sacroiliac joint referral patterns upon application of a new injection/arthrography technique. I: Asymptomatic volunteers. Spine 19, 1475– 1482. Gracovetsky, S., Newman, N., Pawlowsky, M., Lanzo, V., Davey, B., Robinson, L., 1995. A database for estimating normal spinal motion derived from noninvasive measurements. Spine 20, 1036– 1046. Hodges, P.W., Richardson, C.A.., 1997. Contraction of the abdominal muscles associated with movement of the lower limb. Physical Therapy 77, 132–144. Hungerford, B., Gilleard, W., Hodges, P.W., 2003. Evidence of altered lumbo-pelvic muscle recruitment in the presence of sacroiliac joint pain. Spine 28, 1593–1600. Jacob, H., Kissling, R., 1995. The mobility of the sacroiliac joints in healthy volunteers between 20 and 50 years of age. Clinical Biomechanics 10, 352–361. Lee, D., 1999. The Pelvic Girdle: An Approach to Examination and Treatment of the Lumbo-Pelvic-Hip Region. Churchill Livingstone, Edinburgh. Maslen, B., Ackland, T., 1994. Radiographic study of skin displacement errors in the foot and ankle during standing. Clinical Biomechanics 9, 291–296. Mens, J.M.A., Vleeming, A., Snijders, C.J., Stam, H.J., Ginai, A.Z., 1999. The active straight leg raising test and mobility of the pelvic joints. European Spine Journal 8, 468–473. Mens, J.M., Vleeming, A., Snijders, C.J., Koes, B., Stam, H.J., 2001. Reliability and validity of the active straight leg raise test in posterior pelvic pain since pregnancy. Spine 26, 1167–1171. Mitchell, F.J., 1995. The Muscle Energy Manual, vol. 1. MET Press, East Lansing. O’Sullivan, P.B., Beales, D.J., Beetham, J.A., et al., 2002. Altered motor control strategies in subjects with sacroiliac joint pain during the active straight leg raise test. Spine 27, E1–E8. Panjabi, M.M., 1992. The stabilising system of the spine. Part 2: Neutral zone and stability hypothesis. Journal of Spinal Disorders 5, 390–397. Richardson, C.A., Snijders, C.J., Hides, J.A., Damen, L., Martijn, S.P., Storm, J., 2002. The relationship between transversus
abdominis muscles, sacroiliac joint mechanics, and low back pain. Spine 27, 399–405. Sim, J., Reid, N., 1999. Statistical inference by confidence intervals: issues of interpretation and utilization. Physical Therapy 79, 186– 195. Snijders, C.J., Ribbers, M.T., de Bakker, H.V., Stoeckart, R., Stam, H.J., 1998. EMG recordings of abdominal and back muscles in various standing postures: validation of a biomechanical model on sacroiliac joint stability. Journal of Electromyography and Kinesiology 8, 205–214. Snijders, C.J., Vleeming, A., Stoeckart, R., 1993. Transfer of lumbosacral load to iliac bones and legs. 1: Biomechanics of self-bracing of the sacroiliac joints and its significance for treatment and exercise. Clinical biomechanics 8, 285–294. Sturesson, B., Uden, A., Vleeming, A., 2000. A radiological analysis of movements of the sacroiliac joint during the standing hip flexion test. Spine 25, 364–368. Tylkowski, C.M., Simon, S.R., Mansour, J.M., 1982. Internal rotation gait in spastic cerebral palsy in the hip. Paper presented at the Proceedings of the 10th Open Scientific Meeting of the Hip Society, St Louis. pp. 89–125. Vincent-Smith, B., Gibbons, P., 1999. Inter-examiner and intraexaminer reliability of the standing hip flexion test. Manual Therapy 4, 87–93. Vleeming, A., de Vries, H.J., Mens, J.M., van Wingerden, J.P., 2002. Possible role of the long dorsal sacroiliac ligament in women with peripartum pelvic pain. Acta Obstetrics Gynecology Scandinavia 81, 430. Vleeming, A., Pool-Goudzwaard, A.L., Hammudoghlu, D., Stoeckart, R., Snijders, C.J., Mens, J.M., 1996. The function of the long dorsal sacroiliac ligament: its implication for understanding low back pain. Spine 21, 556–562. Vleeming, A., Pool-Goudzwaard, A.L., Stoeckart, R., van Wingerden, J.P., Snijders, C.J., 1995a. The posterior layer of the thoracolumbar fascia: its function in load transfer from spine to legs. Spine 20, 753–758. Vleeming, A., Snijders, C.J., Stoeckart, R., Mens, J.M., 1995b. A new light on low back pain. Paper presented at the 2nd Interdisciplinary World Congress on Low Back Pain, San Diego, pp. 149–168. Vleeming, A., van Wingerden, J.P., Snijders, C.J., Stoeckart, R., Stijnen, T., 1989. Load application to the sacrotuberous ligament: influences on sacroiliac joint mechanics. Clinical Biomechanics 4, 204–209. Vleeming, A., Volkers, A.C.W., Snijders, C.J., Stoeckart, R., 1990. Relation between form and function in the sacroiliac joint. 2: Biomechanical aspects. Spine 15, 133–136. Walheim, G.G., Selvik, G., 1984. Mobility of the pubic symphysis. Clinical Orthopaedics and related Research 191, 129–135.