- Email: [email protected]
Asymmetrical shoulder kinematics in children with brachial plexus birth palsy
Clinical Biomechanics 22 (2007) 630–638 www.elsevier.com/locate/clinbiomech
Asymmetrical shoulder kinematics in children with brachial plexus birth palsy Susan V. Duff a
a,*
, Sudarshan Dayanidhi b, Scott H. Kozin
b
Department of Physical Therapy, Thomas Jefferson University, 130 S. 9th Street, Suite 830, Philadelphia, PA 19107, USA b Shriners Hospitals for Children, PA, USA Received 1 July 2006; accepted 10 February 2007
Abstract Background. Shoulder movement patterns differ between limbs of children with unilateral brachial plexus birth palsy. To better understand the interlimb differences we examined the glenohumeral and scapulothoracic joint contributions to arm elevation. Methods. Sixteen children with brachial plexus birth palsy, 4–12 years of age participated. Shoulder 3D kinematic data were collected using a magnetic tracking device during arm elevation with the involved and non-involved limbs for three trials each at a fixed rate. Based on maximum arm elevation in the involved limb the children were divided into two groups: group one 675; and group two >75. Findings. During arm elevation from 15 to 75 the involved limb of group one displayed lower glenohumeral joint excursion than the non-involved and both limbs of group two. Scapular upward rotation was higher in the involved limb of both groups. For group one, the glenohumeral:scapulothoracic ratio for 15–75 arm elevation was lower in the involved (0.6:1) than the non-involved (2.2:1) limb and both limbs of group two: involved (1.7:1); non-involved (1.9:1). During 15–135 arm elevation for group two, the glenohumeral:scapulothoracic ratio was more similar between limbs: involved (1.5:1) and non-involved (2:1). Interpretation. The scapulothoracic joint made a greater contribution to arm elevation than the glenohumeral joint only in the involved limb of group one, altering the scapulohumeral rhythm. Musculoskeletal and neural factors may account for the group and limb differences. Routine 3D kinematic analysis of shoulder joint rotation may aid treatment planning and better quantify outcomes in this group. 2007 Elsevier Ltd. All rights reserved. Keywords: Scapula; Kinematics; Brachial plexus; Shoulder; Scapulohumeral; Interlimb; Children
1. Introduction Scapulohumeral rhythm or the relationship between the glenohumeral (GH) and scapulothoracic (ST) joints during arm elevation underlies typical shoulder function (Codman, 1934; Inman et al., 1944; McClure et al., 2001). Based on 2D radiographic analysis, Inman and colleagues (1944) reported that GH joint rotation dominates the first 30 of arm elevation in the saggital plane, while the scapula moves minimally. During elevation beyond 30, they found GH
*
Corresponding author. E-mail address: susan.duff@jefferson.edu (S.V. Duff).
0268-0033/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2007.02.002
joint rotation to be twice that of ST joint rotation, resulting in a GH:ST ratio of 2:1 (Inman et al., 1944). Recent 3D analyses of arm elevation beyond 30 in the scapular plane found the GH:ST ratio to be lower than 2:1 (McClure et al., 2001; Poppen and Walker, 1976) and to vary with age and fatigue (Dayanidhi et al., 2005; McQuade et al., 1998). Thus, the scapular contribution to arm elevation seems to be greater than previously reported and may be even greater in select clinical populations (Mell et al., 2005). Based on clinical observation, children with brachial plexus birth palsy (BPBP) are one population that appear to use greater than average scapular motion to elevate the involved arm. BPBP is a condition caused by a unilateral traction injury to the brachial plexus (C5–T1) experienced
S.V. Duff et al. / Clinical Biomechanics 22 (2007) 630–638
in utero or during childbirth (Dodds and Wolfe, 2000; Waters, 1997), and occurs in up to 4.6 of 1000 live births (Hoeksma et al., 2000; Mollberg et al., 2005). According to Waters (2005), the most common form of BPBP is Erb’s palsy (C5–C6) followed by Extended Erb’s palsy (C5–C7). Partial or complete muscle denervation often results in limited arm elevation, humeral external rotation, elbow flexion, forearm supination and occasionally wrist extension (Price and Grossman, 1995; Sundholm et al., 1998; Waters, 2005). Spontaneous recovery from neonatal brachial plexus injury varies from 33% (Waters, 1999) to 94% (Michelow et al., 1994). In children who do not fully recover, the impact on the musculoskeletal integrity of the shoulder and upper limb function is substantial (Strombeck et al., 2000; Waters, 2005). Even children with complete neurological recovery may still exhibit soft-tissue contractures and skeletal deformities (Hoeksma et al., 2003). Although early surgical intervention minimizes the progression of shoulder-joint deformity, existing changes cannot be reversed (Kozin et al., 2006; Waters and Bae, 2005). Because this population has a high probability for developing musculoskeletal impairments and movement dysfunction, quantification of the contributing factors is warranted. Clinical measures such as the Active Movement Scale (Curtis et al., 2002) and Mallet Classification (Mallet, 1972) provide information on general strength but little on the specific joint contributions to movement. Kinematic analysis may be a valuable adjunct to clinical tools, as suggested by Mosqueda and colleagues (2004), who examined arm elevation during activities of daily living in children with BPBP. Although the methods and findings from that study are valid, it may be more informative to quantify the specific shoulder-joint contributions to arm elevation as has been done in typical/atypical adults and children (Dayanidhi et al., 2005; Karduna et al., 2001; Graichen et al., 2000; Ludewig and Cook, 2000; Ludewig et al., 1996; Moriwaki, 1992). Verification of interlimb differences based on the contributions from the GH, ST, acromioclavicular, and sternoclavicular joints could add to our understanding of how shoulder impairments influence movement dysfunction. Documentation of differences may also aid clinical decision-making, and provide a quantitative measure of outcomes. The purpose of this study was to examine the contributions of the GH and ST joints to arm elevation in the involved and non-involved limbs of children with BPBP who have not received surgical intervention. We hypothesized that the GH and ST joint contributions to arm elevation would vary between the involved and non-involved limbs. Furthermore, based on clinical observation we also proposed that children with limited arm elevation would display greater rotation through the ST joint than the GH joint, further altering the pattern of scapulohumeral rhythm.
631
2. Methods 2.1. Subjects Sixteen children with brachial plexus birth palsy participated in this study (Table 1). Children were excluded from participating if they: (1) had undergone microsurgical nerve repair in infancy; (2) had received shoulder reconstructive surgery in the form of tendon transfers or joint releases; (3) presented with bilateral brachial plexus birth injuries; or (4) were unable to follow directions. Temple University’s Institutional Review Board approved the study for children. Signed assent from participants and signed consent from a parent/guardian were received prior to participation. 2.2. Experimental set-up Kinematic data were collected using a magnetic tracking device (Polhemus 3 space Fastrak, Colchester, VT, USA). Prior to testing each limb, the three Polhemus sensors were secured in the following locations: over the T3 spinous process, just below the deltoid insertion with an armband, and on the flat portion of the acromion. This shoulder model and methodology has been used previously in adults and children, and is described in detail (Dayanidhi et al., 2005; Karduna et al., 2000; McClure et al., 2001). Although the method used for examining scapular kinematics during elevation has been validated in adults (Karduna et al., 2001), it has not yet been validated in children. 2.3. Procedure For testing, subjects stood with their feet at a comfortable width apart just behind a line on the floor, with the Table 1 Demographic data Group
Age (years) Mean = 7.8
Gender
Involved limb
Nerve roots affected
Diagnosis Extended Erb’s Palsy Extended Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy Extended Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy
1
8
M
R
C5,6,7
1
4
M
R
C5,6,7
1 1 1 1 1
8 12 4 10 9
M F F F M
L R R R R
C5,6 C5,6 C5,6 C5,6 C5,6,7
1 2 2 2 2 2 2 2 2
5 6 9 5 12 4 12 7 10
F M M M M F F F F
R R L L R R L R R
C5,6 C5,6 C5,6 C5,6 C5,6 C5,6 C5,6 C5,6 C5,6
632
S.V. Duff et al. / Clinical Biomechanics 22 (2007) 630–638
global transmitter aligned immediately posterior at the midthoracic level. Following a demonstration, the children actively lifted each limb separately from a resting position of partial shoulder abduction against the trunk. Each limb was lifted as high as possible anterior to a board placed in the scapular plane (40 anterior to the frontal plane). During arm elevation the elbow was extended and the thumb pointed upward (as feasible given impairments). The order of the starting limb was counterbalanced between the involved and non-involved limbs. Because kinematic differences have been reported based on movement velocity (Michiels and Grevenstein, 1995; Sugamoto et al., 2002), each lift was performed at a fixed rate of about 45 per second per verbal count by the experimenter. Practice trials were provided to familiarize subjects with lifting their limbs at the target velocity. Subjects were asked to look forward while moving, not at their limbs or the computer. All subjects were given verbal feedback after each trial; younger children also received visual cues during each trial.
(group · limb) ANOVAs with repeated measures on the second factor were calculated. To analyze overall scapulohumeral rhythm, we calculated the ratio between GH joint excursion and ST joint excursion. We used scapular UR to represent ST joint excursion for this calculation. We analyzed the GH:ST ratio through 15–75 arm elevation using a 2 · 2 (group · limb) ANOVA with repeated measures on the second factor. Neuman–Keuls post-hoc tests were conducted for significant interactions. A paired t-test was calculated across limbs for the GH:ST ratio of group two through 15–135 arm elevation. Significance was placed at the P < 0.05 level for all statistical analyses. To examine the variability among subjects, we calculated a z-score for GH:ST ratio for each limb of both groups through 15–75 and through 15–135 for group two only. The z-score reveals how many units an individual measure is above or below the mean for a particular variable. 3. Results
2.4. Data analysis 3.1. Joint excursions Arm elevation was calculated in 5 increments by linear interpolation for three repetitions and averaged for each limb for statistical analysis. Previous work has suggested that the root mean squared error (RMSE) for measures of scapular motion increases after 135 of elevation (Karduna et al., 2001). Thus, only data up to 135 of arm elevation was used for analysis. The dependent variables included GH elevation, scapular upward rotation (UR), clavicular elevation, clavicular plane motion (protraction– retraction), scapular external rotation (ER), and scapular posterior tilt (PT). Because the starting and ending values differed between subjects, we calculated the excursion or the difference between the initial and final orientation angle achieved during arm elevation for each dependent variable. Based on the maximum arm elevation in the involved limb, subjects were placed into one of two groups: group one (N = 8) included those whose maximum arm elevation was 75 or below; and group two (N = 8) included those whose maximum arm elevation was above 75. This division of subjects corresponded with clinical differences in active abduction observed in children with BPBP based on the Mallet Scale (Mallet, 1972); with group one associated with grade III and group two associated with grade IV classification. To examine the group and limb differences for the excursions displayed during 15–75 of arm elevation, 2 · 2 (group · limb) Analysis of Variance (ANOVA) with repeated measures on the second factor were calculated for each dependent variable listed above. We also calculated the percent contribution to arm elevation by subject, for each dependent variable, through 15–75 or an arc of 60. Therefore, the percent contribution = degree of excursion/60. To assess group and limb differences in the percent contribution for each dependent variable, 2 · 2
Fig. 1a–f shows the mean excursions for each dependent variable during 15–75 of arm elevation. Table 2 lists the statistical findings. As shown for GH elevation (Fig. 1a), a group · limb interaction was found (P < 0.0001) with main effects for group and limb (both, P < 0.0001). Posthoc tests indicated that GH joint excursion in the involved limb of group one was lower than the non-involved and both limbs of group two (all, P < 0.05). Excursion in the clavicular plane (Fig. 1d), was greater in group one (P < 0.01) and the involved limb (P < 0.0002). Excursion into scapular ER (Fig. 1e), was greater for group one (P < 0.05) and for scapular UR (Fig. 1b), it was greater in the involved limb (P < 0.05). Excursion for clavicular elevation (Fig. 1c) and scapular PT (Fig. 1f) were not significantly different for group or limb. Fig. 2 shows the percent contribution each dependent variable made toward 15–75 of arm elevation among four subgroups clustered by group and limb (Fig. 2a–d). Statistical findings are listed in Table 3. In general, the percent contribution from GH joint elevation was lower and for most ST variables was greater in the involved limb of group one (Fig. 2a) when compared to the non-involved (Fig. 2b), and both limbs of group two (Fig. 2c and d). For GH elevation we found a group · limb interaction (P < 0.001) and main effects for group and limb (both, P < 0.0001). Post-hoc tests indicated that the contribution from GH elevation was lower in the involved limb of group one than the non-involved and both limbs of group two (all, P < 0.05). For clavicular elevation (Fig. 2d), we found a group · limb interaction (P < 0.05) and a main effect for limb (P < 0.01) only. Post-hoc tests revealed that clavicular elevation made a greater contribution to arm elevation in
S.V. Duff et al. / Clinical Biomechanics 22 (2007) 630–638
633
Fig. 1. Mean excursion (Exc) (standard error of the mean or SEM) during 15–75 arm elevation for the involved (Inv) limb (left) and non-involved (NI) limb (right) of group 1 (solid line) and group 2 (dashed line) for all dependent variables: (a) GH elevation; (b) scapular UR; (c) clavicular elevation; (d) clavicular plane motion; (e) scapular ER; and (f) scapular PT. See Table 2 for F ratios and significant P-values.
Table 2 F ratios and significant P-values (0.05*, P0.01**) for differences between group and limb in the mean excursion of dependent variables during 15–75 arm elevation Excursion
GH elevation Scapular UR Clavicular elevation Clavicular plane motion Scapular ER Scapular PT
Group · limb interaction
Main effect group (1 vs. 2)
Main effect limb (Inv vs. NI)
F ratio
P-value
F ratio
P-value
F ratio
P-value
28.4 .49 1.37 1.34 1.55 1.53
0.0001**
40.77 .71 .50 9.6 7.32 .68
0.0001**
28.6 8.05 3.08 23.96 1.34 4.4
0.0001** 0.01** 0.10 0.0002** 0.27 0.06
0.50 0.26 0.27 0.23 0.24
0.41 0.49 0.01** 0.02* 0.42
See Fig. 1.
the involved limb than the non-involved and both limbs of group two (all, P < 0.05). For clavicular plane motion (Fig. 2d), we found a group · limb interaction (P < 0.05) and main effects for group and limb (both, P < 0.0001). Post-hoc tests revealed that clavicular plane motion made a greater contribution in the involved limb of group one than the non-involved and both limbs of group two (all, P < 0.05). The contribution from scapular ER was greater in group one (P < 0.005). Scapular UR (P < 0.002) and scapular PT (P < 0.05) made greater contributions in the involved limb. 3.2. Scapulohumeral rhythm Fig. 3a displays the GH:ST ratio for all subjects across group and limb for 15–75 arm elevation. As shown, the
ratio in the involved limb of most subjects in group one was lower than the non-involved limb. The involved limb of group one was also lower than both limbs of group two. Individual ratios were supported by the mean GH:ST ratio (Fig. 3b and Table 4), based on a group · limb interaction (P < 0.001) and a main effect for limb (P < 0.0001). Post-hoc tests indicated that the GH:ST ratio in the involved limb of group one (0.6:1) was lower than the non-involved (2.2:1) (P < 0.05) and both the involved (1.7:1) (P < 0.05) and non-involved (1.9:1) (P < 0.05) limbs of group two. Z-scores for individual GH:ST ratios ranged from 1.5 above and below the mean (Fig. 3c), suggesting the distributions were generally consistent and stable. Fig. 4a shows the GH:ST ratio for 15–135 arm elevation for each subject across limbs for group two only. Overall, the excursions in the involved limb skewed slightly
634
S.V. Duff et al. / Clinical Biomechanics 22 (2007) 630–638
Fig. 2. Percent (%) contribution to 15–75 arm elevation (excursion/60), mean (SEM) of dependent variables within 4 clusters including: (a) Involved limb – Group one; (b) Non-involved limb – Group two; (c) Involved limb – Group two; and (d) Non-involved – Group two. The dependent variables in each of the four clusters are shown from left to right as follows: GH elevation (black); scapular UR (dark gray); clavicular elevation (medium-dark gray); clavicular plane motion (medium-light gray); scapular ER (light gray); and scapular posterior tilt (white). See Table 3 for F ratios and significant P-values.
Table 3 F ratios and significant P-values (0.05*, P0.01**) for differences between group and limb in the percent contribution of dependent variables during 15–75 arm elevation Percent contribution
Group · limb interaction F ratio
P-value
F ratio
P-value
F ratio
P-value
GH Elevation Scapular UR Clavicular elevation Clavicular plane motion Scapular ER Scapular PT
46.11 3.99 5.14 7.9 2.3 2.64
0.0001** 0.07 0.05* 0.05* 0.16 0.13
16.04 2.2 1.24 29.65 11.64 2.84
0.001** 0.16 .28 0.0001** 0.005** 0.11
38.9 15.22 8.97 44.3 2.22 5.62
0.0001** 0.002** 0.01** 0.0001** 0.16 0.05*
Main effect group (1 vs. 2)
Main effect limb (Inv vs. NI)
See Fig. 2.
lower and the range was narrower than the non-involved limb. As shown in Fig. 4b, the mean GH:ST ratio for 15–135 of arm elevation did not significantly differ (F = 2.88; P > 0.05) between the involved (1.5:1) and non-involved (2:1) limbs. Individual z-scores for the GH:ST ratio ranged from 1.5 above and below the mean for both limbs (Fig. 4c), suggesting the distributions were generally consistent and stable. 4. Discussion 4.1. Altered movement patterns Kinematic findings from this study support the clinical observation that children with BPBP employ greater than average scapular motion in the involved limb during arm elevation. The results also suggest that the relative contributions from the GH and ST joints through 15–75 depend on the amount of active arm elevation available. For the children with limited elevation (group one), the ST joint in the involved limb made a larger contribution than the GH joint, resulting in a much lower GH:ST ratio. In children with greater elevation (group two), the GH joint in the involved limb made a larger contribution than the ST joint, resulting in a higher ratio. Interestingly, for 15–135 arm elevation, the GH:ST ratio in the involved limb (1.5:1)
resembled the ratio reported for the dominant limb of typical children (1.3:1) during 25–125 arm elevation (Dayanidhi et al., 2005). A greater contribution from the ST joint during arm elevation has also been observed in adults with rotator cuff tears (Mell et al., 2005) and induced shoulder muscle fatigue (Ebaugh et al., 2006; McQuade et al., 1998). As in the case of adults with rotator cuff tears, the altered scapulohumeral pattern found in children with BPBP in group one may have allowed the intact muscles to function in a more effective portion of the length-tension curve, enhancing the strength of the muscle contractions used to lift the arm. During this study, we controlled the velocity by pacing the lift. It is possible that this demand prevented the children with BPBP from using a common strategy used to enhance muscle power; that of increasing the velocity of a lift to compensate for weakness (Burkholder and Leiber, 2001). In particular, children in group one, with less active elevation, may routinely lift the involved arm quickly when the demand arises or they may avoid lifting it because they cannot generate sufficient muscle power to do so. 4.2. Achieving dynamic shoulder stability Dynamic shoulder stability can be defined as centering of the humeral head on the glenoid fossa accompanied by
S.V. Duff et al. / Clinical Biomechanics 22 (2007) 630–638
635
humeral head internally rotates and shifts posteriorly in the glenoid fossa (Waters et al., 1998), altering the line of muscle pull, or action line. This biomechanical alteration could diminish the contraction strength of the external rotators even if they become re-innervated. The variation in muscle length and strength may partly account for group differences in the scapulohumeral pattern found in this study (Langenderfer et al., 2006). The children in group two may be able to effectively activate the shoulder external rotators needed to stabilize and rotate the humerus, whereas, the children in group one may be unable to do so. Obligatory resting postures and habitual use patterns contribute to the evolution of skeletal deformities (Poyhia et al., 2005; Waters, 1997). These deformities include osseous changes in the glenoid fossa and humeral head, progressive posterior dislocation of the humeral head, and shortening of the anterior GH joint capsule (Hoeksma et al., 2000, 2003; Kozin, 2004; Pearl and Edgerton, 1998; Poyhia et al., 2005; Soldado and Kozin, 2005; Waters et al., 1998). Typically, if the humerus cannot externally rotate or glide inferiorly, it may be able to achieve only 60–90 of elevation before the glenoid tubercle hits the acromion (Schenkman and Rugo De Cartaya, 1987). Because of muscle weakness and skeletal deformities in children with BPBP, the greater tubercle remains vulnerable,
Fig. 3. GH:ST ratio during 15–75 arm elevation for both groups: 15(a) individual subject GH:ST ratios for the involved (solid circles) and noninvolved (open circles) from group 1 (left) and group 2 (right); (b) mean (SEM) GH:ST ratio for the involved (Inv) limb (left) and non-involved (NI) limb (right) of group 1 (solid line) and group 2 (dashed line); and (c) z-scores of all subjects for GH:ST ratio for the involved (solid circles) and non-involved (open circles) limbs from group 1 (bottom) and group 2 (top). See Table 4 for F ratios and significant P values.
Table 4 F ratios and significant P-values (0.05*, P0.01**) for differences between group and limb in the GH:ST ratio during 15–75 arm elevation Arm elevation 15–75
Group · limb interaction
Main effect group (1 vs. 2)
Main effect limb (Inv vs. NI)
F ratio
P-value
F ratio
P-value
F ratio
P-value
GH:ST ratio
17.35
0.001*
3.07
0.1
30.44
0.0001*
See Fig. 3.
inferior translation via the rotator cuff muscles to facilitate humeral rotation. The shoulder weakness and atrophy found in children with BPBP seems to alter this stability. Because of weak shoulder external rotators, children with BPBP often present with an obligatory resting posture of humeral internal rotation. As posturing into internal rotation progresses, the external rotators (e.g., infraspinatous and teres minor) lengthen, and the internal rotators (e.g., subscapularis) shorten. With muscle length changes, the
Fig. 4. GH:ST ratio during 15–135 arm elevation for group 2 only: (a) individual subject GH:ST ratios for the involved (solid circles, bottom) and non-involved (open circles, top) limbs; (b) mean (SEM) GH:ST ratio for the involved (Inv) limb (left) and non-involved (NI) limb (right) which were not significantly different; and (c) z-scores of all subjects in group 2 for GH:ST ratio with the involved (solid circles, bottom) and non-involved (open circles, top) limbs.
636
S.V. Duff et al. / Clinical Biomechanics 22 (2007) 630–638
and repetitive contact with the acromion could increase the risk for impingement syndrome (McClure et al., 2006). Enhanced scapular UR and clavicular elevation, particularly in group two, may in part keep the acromion out of the way of the glenoid tubercle allowing for greater humeral elevation and dynamic stability. Alterations in muscle power and musculoskeletal integrity may differentially impact the dynamic joint stability and movement patterns displayed by each group. To further elucidate the group and limb differences, imaging techniques and EMG analysis would need to be employed (Illyes and Kiss, 2006). 4.3. Neural adaptation Neonatal nerve injury may be more detrimental than adult nerve injury because peripheral lesions may deprive central neurons of important trophic factors, leading to cell death and smaller neural representations (Cusick, 1996). Group differences in this study may stem from the degree of initial nerve injury along with variable neural recovery and adaptation. As shown in Table 1, 3 out of 8 children in group one had been given an initial diagnosis of Extended Erb’s palsy, suggesting they had greater nerve injury at birth and/or less neural recovery post-natally. The central nervous system adapts to alterations in deafferentation and use (Florence et al., 1998; Garraghty and Kaas, 1991; Merzenich et al., 1983; Merzenich and Jenkins, 1993). Thus, if the involved limb of a child with BPBP is weak, and sensibility is reduced, he or she may be less apt to use it, contributing to disuse atrophy and further loss of neural representations (Cusick, 1996; Taub et al., 1975). The degree of initial injury and subsequent recovery as well as differences in neural representations would likely influence shoulder movement patterns, thus may be partly responsible for group differences. 4.4. Clinical relevance Despite muscle imbalances and skeletal changes, children in both groups are reportedly quite functional. Findings from this study suggest that the movement strategy of increased scapular UR observed in children who have not had surgical intervention might provide functional benefit because of the increased active range it affords. Key goals for treatment of this population are to limit the development of musculoskeletal deformity, to strengthen recovering muscles, and to promote active use of the involved limb. From early infancy, appropriate stretching and mobilization programs for scapulohumeral muscles, such as the shoulder internal rotators and the GH joint capsule, should be emphasized to prevent the development of joint contractures (Ramos and Zell, 2000; Waters, 2005). Along with passive range-of-motion, strengthening of recovering and potential transfer muscles such as the latissimus dorsi (Waters and Bae, 2005) should be encouraged. If passive range is not maintained or activation of partially innervated muscles is not promoted mus-
culoskeletal deformities will worsen (Ramos and Zell, 2000). Furthermore, active use and progression through motor milestones should be reinforced. If unimanual function is limited, bimanual task practice may be fostered. As with other neurological injuries, it is important to prevent disuse muscle atrophy and reinforce neural representations needed for adequate motor control (Thomas et al., 1997). Criteria for surgical intervention are linked to muscle/ skeletal integrity and functional needs. Growth spurts in adolescence have been found to increase skeletal deformities, including a progressive increase in glenoscapular angle and posterior subluxation of the humeral head (Waters et al., 1998). Because the progression in skeletal deformity may further impact movement patterns and function, surgical intervention is now being conducted in younger patients including infants and toddlers. One such surgery is the release of the joint capsule and tendon transfer of the latissimus dorsi and teres major to a posterior/lateral position high on the rotator cuff tendon(s) (Waters and Bae, 2005). This transfer allows former internal rotators to act as external rotators and improves the degree of functional arm elevation. Severe deformities in older children may require an osteotomy of the humerus with external rotation into a more neutral alignment to improve arm elevation for overhead tasks such as hair grooming (Waters, 1997). Differences in muscle activation, skeletal integrity and subsequent movement patterns between the two groups examined in this study might provide a basis for the selection of surgical candidates and the type of surgical procedure required. As suggested by Mosqueda and colleagues (2004), the inclusion of kinematic analysis in our study allowed for close examination of the interlimb differences in joint contribution to arm elevation. Clinicians and researchers may be able to utilize the magnetic tracking device as an adjunct to clinical scales such as the Mallet Scale (Mallet, 1972) to document baseline performance along with outcomes from therapeutic and surgical intervention. The additional use of EMG and imaging techniques would allow tracking of sequential muscle activation and GH and ST joint rotation during various phases of arm elevation. 5. Conclusion This study provided support for the clinical observation of interlimb differences in shoulder movement patterns in the involved and non-involved limbs of children with brachial plexus birth palsy. Children with limited arm elevation (675) in the involved limb displayed a greater contribution from the ST joint than the GH joint when compared to children with greater arm elevation (>75). However, both groups displayed greater scapular UR in the involved limb. Although enhanced scapular mobility is likely a consequence of muscle imbalance, skeletal deformity, and neural adaptation, it serves to enhance function during the performance of unimanual and bimanual tasks. Given better tools for evaluating the impact of nerve injury
S.V. Duff et al. / Clinical Biomechanics 22 (2007) 630–638
on shoulder movement patterns, we may be able to refine methods of therapeutic and surgical intervention to enhance neuromuscular resources and movement strategies for those with brachial plexus birth palsy. Acknowledgements We would like to thank Marcus Besser, PhD, Andrew Karduna, PhD, and Patricia A. Shewokis, PhD for their thoughtful reviews and suggestions. References Burkholder, T.J., Leiber, R.L., 2001. Sarcomere length operating range of vertebrate muscles during movement. J Exp. Biol. 204, 1529– 1536. Codman, E.A., 1934. The shoulder. Thomas Todd, Boston. Curtis, C., Stephens, D., Clarke, H.M., Andrews, D., 2002. The active movement scale: an evaluative tool for infants with obstetrical brachial plexus palsy. J. Hand Surg. [Am]. 27 (3), 470–478. Cusick, C.G., 1996. Extensive cortical reorganization following sciatic nerve injury in adult rats versus restricted reorganization after neonatal injury: implications for spatial and temporal limits on somatosensory plasticity. Prog. Brain Res. 108, 379–390. Dayanidhi, S., Orlin, M., Kozin, S., Duff, S., Karduna, A., 2005. Scapular kinematics during humeral elevation in adults and children. Clin. Biomech. 20 (6), 600–606. Dodds, S.D., Wolfe, S.W., 2000. Perinatal brachial plexus palsy. Curr. Opin. Pediatr. 12 (1), 40–47. Ebaugh, D.D., McClure, P.W., Karduna, A.R., 2006. Three-dimensional scapulothoracic motion during active and passive arm elevation. Clin. Biomech. 20, 700–709. Florence, S.L., Taub, H.B., Kaas, J.H., 1998. Large-scale sprouting of cortical connections after peripheral nerve injury in adult macaque monkeys. Science 282 (5391), 1117–1121. Garraghty, P.E., Kaas, J.H., 1991. Large-scale functional reorganization in adult monkey cortex after peripheral nerve injury. Proc. Natl. Acad. Sci. USA 88 (16), 6976–6980. Graichen, H., Stammberger, T., Bonel, H., Haubner, M., Englmeier, K.H., Reiser, M., Eckstein, F., 2000. Magnetic resonance-based motion analysis of the shoulder during elevation. Clin. Orthop. Relat. Res. 370, 154–163. Hoeksma, A.F., Ter Steeg, A.M., Dijkstra, P., Nelissen, R.G., Beelen, A., de Jong, B.A., 2003. Shoulder contracture and osseous deformity in obstetrical brachial plexus injuries. J. Bone Joint Surg. 85A (2), 316– 322. Hoeksma, A.F., Wolf, H., Oei, S.L., 2000. Obstetrical brachial plexus injuries: incidence, natural course and shoulder contracture. Clin. Rehab. 14 (5), 523–526. Illyes, A., Kiss, R.M., 2006. [Epub]. Shoulder joint kinematics during elevation measured by ultrasound-based measuring system. J. Electromyogr. Kinesiol. Apr 16. Inman, V.T., Saunders, J.B., Abbott, L.C., 1944. Observations on the function of the shoulder joint. J. Bone Joint Surg. 26, 1–30. Karduna, R.A., McClure, P.W., Michener, L.A., Sennett, B., 2001. Dynamic measurement of three-dimensional scapular kinematics: a validation study. J. Biomech. Eng. 123, 184–190. Karduna, R.A., McClure, P.W., Michener, L.A., 2000. Scapular kinematics: effects of altering the Euler angle sequence of rotations. J. Biomech. 33, 1063–1068. Kozin, S.H., 2004. Correlation between external rotation of the glenohumeral joint and deformity after brachial plexus birth palsy. J. Ped. Orthop. 24 (2), 189–193. Kozin, S.H., Chafetz, R.S., Barus, D., Filipone, L., 2006. Magnetic resonance imaging and clinical findings before and after tendon
637
transfers about the shoulder in children with residual brachial plexus birth palsy. J. Shoulder Elbow Surg. 15 (5), 554–561. Langenderfer, J.E., Patthanacharenphon, C., Carpenter, J.E., Hughes, R.E., 2006. Variability in isometric force and moment during generating capability of glenohumeral external rotator muscles. Clin. Biomech. 21 (7), 701–709. Ludewig, P.M., Cook, T.M., 2000. Alterations in shoulder kinematics and associated muscle activity in people with symptoms of shoulder impingement. Phys. Ther. 80 (3), 276–291. Ludewig, P.M., Cook, T.M., Nawoczenski, D.A., 1996. Three-dimensional scapular orientation and muscle activity at selected positions of humeral elevation. J. Orthop. Sports Phys. Ther. 24 (2), 57–65. Mallet, J., 1972. Primaute du traitement de l’epaule – methode d’expresion des resultats. Rev. Chir. Orthop. 58S, 166–168. McClure, P., Michener, L.A., Karduna, A.R., 2006. Shoulder function and 3-dimensional scapular kinematics in people with and without shoulder impingement syndrome. Phys. Ther. 86 (8), 1075–1090. McClure, P., Michener, L.A., Sennett, B.J., Karduna, A.R., 2001. Direct 3-dimensional measurement of scapular kinematics during dynamic movements during in vivo. J. Shoulder Elb. Surg. 10, 269–277. McQuade, K.J., Dawson, J., Smidt, G.L., 1998. Scapulothoracic muscle fatigue associated with alterations in scapulohumeral rhythm kinematics during maximum resistive shoulder elevation. J. Orthop. Sports Phys. Ther. 28 (2), 74–80. Mell, A.G., LaScalza, S., Guffey, P., Ray, J., Maciejewski, M., Carpenter, J.E., Hughes, R.E., 2005. Effect of rotator cuff pathology on shoulder rhythm. J. Shoulder Elbow Surg. 14 (1S), 58S–64S. Merzenich, M.M., Jenkins, W.M., 1993. Reorganization of cortical representations of the hand following alterations of skin inputs induced by nerve injury, skin island transfers, and experience. J. Hand Ther. 6 (2), 89–104. Merzenich, M.M., Kaas, J.H., Wall, J., Nelson, R.J., Sur, M., Felleman, D., 1983. Topographic reorganization of somatosensory cortical areas 3b and 1 in adult monkeys following restricted deafferentation. Neuroscience 8 (1), 33–55. Michelow, B.J., Clarke, H.M., Curtis, C.G., Zuker, R.M., Seifu, Y., Andrews, D.F., 1994. The natural history of obstetrical brachial plexus palsy. Plast. Reconstr. Surg. 93 (4), 675–680. Michiels, I., Grevenstein, J., 1995. Kinematics of shoulder arm elevation in the scapular plane. On the influence of arm elevation velocity and external load. Clin. Biomech. 10 (3), 137–143. Mollberg, M., Hagberg, H., Bager, B., Lilja, H., Ladfors, L., 2005. High birthweight and shoulder dystonia: the strongest risk factors for obstetrical brachial plexus palsy in a Swedish population-based study. Acta Obstet. Gynecol. Scand. 84 (7), 654–659. Moriwaki, M., 1992. Analysis of three dimensional motion of the scapula and the glenohumeral joint. J. Japanese Orthop. Assoc. 66, 675–687. Mosqueda, T., James, M.A., Petuskey, K., Bagley, A., Abdala, E., Rab, G., 2004. Kinematic assessment of the upper extremity in brachial plexus birth palsy. J. Ped. Orthop. 24 (6), 695–699. Pearl, M.L., Edgerton, B.W., 1998. Glenoid deformity secondary brachial plexus birth palsy. J. Bone Joint Surg. [Am] 80 (5), 659–667. Poppen, N., Walker, P., 1976. Normal and abnormal motion of the shoulder. J. Bone Joint Surg. [Am] 58 (2), 195–201. Poyhia, T.H., Nietosvaara, Y.A., Remes, V.M., Kirjavainen, M.O., Peltonen, J.I., Lamminen, A.E., 2005. MRI of rotator cuff muscle atrophy in relation to glenohumeral joint incongruence in brachial plexus birth injury. Ped. Radiol. 35 (4), 402–409. Price, A.E., Grossman, J.A., 1995. A management approach for secondary shoulder and forelimb deformities following obstetrical brachial plexus injury. Hand Clin. 11 (4), 607–617. Ramos, L.E., Zell, J.P., 2000. Rehabilitation program for children with brachial plexus and peripheral nerve injury. Semin Pediatr Neurol. 7 (1), 52–57. Schenkman, M., Rugo De Cartaya, V., 1987. Kinesiology of the shoulder complex. J. Ortho. Sports Phys. Ther. 8 (9), 438–450. Strombeck, C., Krumlinde-Sundholm, L., Forssberg, H., 2000. Functional outcome at 5 years in children with obstetrical brachial plexus palsy
638
S.V. Duff et al. / Clinical Biomechanics 22 (2007) 630–638
with and without microsurgical reconstruction. Dev. Med. Child Neurol. 42 (3), 148–157. Sugamoto, K., Harada, T., Machida, A., Inui, H., Miyamota, T., Takeuchi, E., Yoshikawa, H., Ochi, T., 2002. Scapulohumeral rhythm: relationship between motion velocity and rhythm. Clin. Orthop. Relat. Res. 401, 119–124. Sundholm, L.K., Eliasson, A.C., Forssberg, H., 1998. Obstetric brachial plexus injuries: assessment protocol and functional outcome at age 5 years. Dev. Med. Child Neurol. 40 (1), 4–11. Soldado, F., Kozin, S.H., 2005. The relationship between the coracoid and glenoid after brachial plexus birth palsy. J. Ped. Orthop. 25 (5), 666–670. Taub, E., Goldberg, I.A., Taub, P., 1975. Deafferentation in monkeys: pointing at a target without visual feedback. Exp. Neurol. 46, 176–186. Thomas, C.K., Zaidner, E.Y., Calancie, B., Broton, J.G., Bigland-Ritchie, B.R., 1997. Muscle weakness, paralysis, and atrophy after human cervical spinal cord injury. Exp. Neurol. 148 (2), 414–423.
Waters, P.M., 1997. Obstetric brachial plexus injuries: evaluation and management. J. Am. Acad. Orthop. Surg. 5, 205–214. Waters, P.M., 1999. Comparison of the natural history, the outcome of microsurgical repair, and the outcome of operative reconstruction in brachial plexus birth palsy. J. Bone Joint Surg. [Am] 81 (5), 649– 659. Waters, P.M., 2005. Update on management of pediatric brachial plexus palsy. J. Ped. Orthop. 25 (1), 116–126. Waters, P.M., Bae, D.S., 2005. Effect of tendon transfers and extraarticular soft-tissue balancing on glenohumeral development in brachial plexus birth palsy. J. Bone Joint Surg. [Am] 87 (2), 320– 325. Waters, P.M., Smith, G.R., Jaramillo, D., 1998. Glenohumeral deformity secondary to brachial plexus birth palsy. J. Bone Joint Surg. [Am] 80 (5), 668–677.
Asymmetrical shoulder kinematics in children with brachial plexus birth palsy Susan V. Duff a
a,*
, Sudarshan Dayanidhi b, Scott H. Kozin
b
Department of Physical Therapy, Thomas Jefferson University, 130 S. 9th Street, Suite 830, Philadelphia, PA 19107, USA b Shriners Hospitals for Children, PA, USA Received 1 July 2006; accepted 10 February 2007
Abstract Background. Shoulder movement patterns differ between limbs of children with unilateral brachial plexus birth palsy. To better understand the interlimb differences we examined the glenohumeral and scapulothoracic joint contributions to arm elevation. Methods. Sixteen children with brachial plexus birth palsy, 4–12 years of age participated. Shoulder 3D kinematic data were collected using a magnetic tracking device during arm elevation with the involved and non-involved limbs for three trials each at a fixed rate. Based on maximum arm elevation in the involved limb the children were divided into two groups: group one 675; and group two >75. Findings. During arm elevation from 15 to 75 the involved limb of group one displayed lower glenohumeral joint excursion than the non-involved and both limbs of group two. Scapular upward rotation was higher in the involved limb of both groups. For group one, the glenohumeral:scapulothoracic ratio for 15–75 arm elevation was lower in the involved (0.6:1) than the non-involved (2.2:1) limb and both limbs of group two: involved (1.7:1); non-involved (1.9:1). During 15–135 arm elevation for group two, the glenohumeral:scapulothoracic ratio was more similar between limbs: involved (1.5:1) and non-involved (2:1). Interpretation. The scapulothoracic joint made a greater contribution to arm elevation than the glenohumeral joint only in the involved limb of group one, altering the scapulohumeral rhythm. Musculoskeletal and neural factors may account for the group and limb differences. Routine 3D kinematic analysis of shoulder joint rotation may aid treatment planning and better quantify outcomes in this group. 2007 Elsevier Ltd. All rights reserved. Keywords: Scapula; Kinematics; Brachial plexus; Shoulder; Scapulohumeral; Interlimb; Children
1. Introduction Scapulohumeral rhythm or the relationship between the glenohumeral (GH) and scapulothoracic (ST) joints during arm elevation underlies typical shoulder function (Codman, 1934; Inman et al., 1944; McClure et al., 2001). Based on 2D radiographic analysis, Inman and colleagues (1944) reported that GH joint rotation dominates the first 30 of arm elevation in the saggital plane, while the scapula moves minimally. During elevation beyond 30, they found GH
*
Corresponding author. E-mail address: susan.duff@jefferson.edu (S.V. Duff).
0268-0033/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2007.02.002
joint rotation to be twice that of ST joint rotation, resulting in a GH:ST ratio of 2:1 (Inman et al., 1944). Recent 3D analyses of arm elevation beyond 30 in the scapular plane found the GH:ST ratio to be lower than 2:1 (McClure et al., 2001; Poppen and Walker, 1976) and to vary with age and fatigue (Dayanidhi et al., 2005; McQuade et al., 1998). Thus, the scapular contribution to arm elevation seems to be greater than previously reported and may be even greater in select clinical populations (Mell et al., 2005). Based on clinical observation, children with brachial plexus birth palsy (BPBP) are one population that appear to use greater than average scapular motion to elevate the involved arm. BPBP is a condition caused by a unilateral traction injury to the brachial plexus (C5–T1) experienced
S.V. Duff et al. / Clinical Biomechanics 22 (2007) 630–638
in utero or during childbirth (Dodds and Wolfe, 2000; Waters, 1997), and occurs in up to 4.6 of 1000 live births (Hoeksma et al., 2000; Mollberg et al., 2005). According to Waters (2005), the most common form of BPBP is Erb’s palsy (C5–C6) followed by Extended Erb’s palsy (C5–C7). Partial or complete muscle denervation often results in limited arm elevation, humeral external rotation, elbow flexion, forearm supination and occasionally wrist extension (Price and Grossman, 1995; Sundholm et al., 1998; Waters, 2005). Spontaneous recovery from neonatal brachial plexus injury varies from 33% (Waters, 1999) to 94% (Michelow et al., 1994). In children who do not fully recover, the impact on the musculoskeletal integrity of the shoulder and upper limb function is substantial (Strombeck et al., 2000; Waters, 2005). Even children with complete neurological recovery may still exhibit soft-tissue contractures and skeletal deformities (Hoeksma et al., 2003). Although early surgical intervention minimizes the progression of shoulder-joint deformity, existing changes cannot be reversed (Kozin et al., 2006; Waters and Bae, 2005). Because this population has a high probability for developing musculoskeletal impairments and movement dysfunction, quantification of the contributing factors is warranted. Clinical measures such as the Active Movement Scale (Curtis et al., 2002) and Mallet Classification (Mallet, 1972) provide information on general strength but little on the specific joint contributions to movement. Kinematic analysis may be a valuable adjunct to clinical tools, as suggested by Mosqueda and colleagues (2004), who examined arm elevation during activities of daily living in children with BPBP. Although the methods and findings from that study are valid, it may be more informative to quantify the specific shoulder-joint contributions to arm elevation as has been done in typical/atypical adults and children (Dayanidhi et al., 2005; Karduna et al., 2001; Graichen et al., 2000; Ludewig and Cook, 2000; Ludewig et al., 1996; Moriwaki, 1992). Verification of interlimb differences based on the contributions from the GH, ST, acromioclavicular, and sternoclavicular joints could add to our understanding of how shoulder impairments influence movement dysfunction. Documentation of differences may also aid clinical decision-making, and provide a quantitative measure of outcomes. The purpose of this study was to examine the contributions of the GH and ST joints to arm elevation in the involved and non-involved limbs of children with BPBP who have not received surgical intervention. We hypothesized that the GH and ST joint contributions to arm elevation would vary between the involved and non-involved limbs. Furthermore, based on clinical observation we also proposed that children with limited arm elevation would display greater rotation through the ST joint than the GH joint, further altering the pattern of scapulohumeral rhythm.
631
2. Methods 2.1. Subjects Sixteen children with brachial plexus birth palsy participated in this study (Table 1). Children were excluded from participating if they: (1) had undergone microsurgical nerve repair in infancy; (2) had received shoulder reconstructive surgery in the form of tendon transfers or joint releases; (3) presented with bilateral brachial plexus birth injuries; or (4) were unable to follow directions. Temple University’s Institutional Review Board approved the study for children. Signed assent from participants and signed consent from a parent/guardian were received prior to participation. 2.2. Experimental set-up Kinematic data were collected using a magnetic tracking device (Polhemus 3 space Fastrak, Colchester, VT, USA). Prior to testing each limb, the three Polhemus sensors were secured in the following locations: over the T3 spinous process, just below the deltoid insertion with an armband, and on the flat portion of the acromion. This shoulder model and methodology has been used previously in adults and children, and is described in detail (Dayanidhi et al., 2005; Karduna et al., 2000; McClure et al., 2001). Although the method used for examining scapular kinematics during elevation has been validated in adults (Karduna et al., 2001), it has not yet been validated in children. 2.3. Procedure For testing, subjects stood with their feet at a comfortable width apart just behind a line on the floor, with the Table 1 Demographic data Group
Age (years) Mean = 7.8
Gender
Involved limb
Nerve roots affected
Diagnosis Extended Erb’s Palsy Extended Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy Extended Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy Erb’s Palsy
1
8
M
R
C5,6,7
1
4
M
R
C5,6,7
1 1 1 1 1
8 12 4 10 9
M F F F M
L R R R R
C5,6 C5,6 C5,6 C5,6 C5,6,7
1 2 2 2 2 2 2 2 2
5 6 9 5 12 4 12 7 10
F M M M M F F F F
R R L L R R L R R
C5,6 C5,6 C5,6 C5,6 C5,6 C5,6 C5,6 C5,6 C5,6
632
S.V. Duff et al. / Clinical Biomechanics 22 (2007) 630–638
global transmitter aligned immediately posterior at the midthoracic level. Following a demonstration, the children actively lifted each limb separately from a resting position of partial shoulder abduction against the trunk. Each limb was lifted as high as possible anterior to a board placed in the scapular plane (40 anterior to the frontal plane). During arm elevation the elbow was extended and the thumb pointed upward (as feasible given impairments). The order of the starting limb was counterbalanced between the involved and non-involved limbs. Because kinematic differences have been reported based on movement velocity (Michiels and Grevenstein, 1995; Sugamoto et al., 2002), each lift was performed at a fixed rate of about 45 per second per verbal count by the experimenter. Practice trials were provided to familiarize subjects with lifting their limbs at the target velocity. Subjects were asked to look forward while moving, not at their limbs or the computer. All subjects were given verbal feedback after each trial; younger children also received visual cues during each trial.
(group · limb) ANOVAs with repeated measures on the second factor were calculated. To analyze overall scapulohumeral rhythm, we calculated the ratio between GH joint excursion and ST joint excursion. We used scapular UR to represent ST joint excursion for this calculation. We analyzed the GH:ST ratio through 15–75 arm elevation using a 2 · 2 (group · limb) ANOVA with repeated measures on the second factor. Neuman–Keuls post-hoc tests were conducted for significant interactions. A paired t-test was calculated across limbs for the GH:ST ratio of group two through 15–135 arm elevation. Significance was placed at the P < 0.05 level for all statistical analyses. To examine the variability among subjects, we calculated a z-score for GH:ST ratio for each limb of both groups through 15–75 and through 15–135 for group two only. The z-score reveals how many units an individual measure is above or below the mean for a particular variable. 3. Results
2.4. Data analysis 3.1. Joint excursions Arm elevation was calculated in 5 increments by linear interpolation for three repetitions and averaged for each limb for statistical analysis. Previous work has suggested that the root mean squared error (RMSE) for measures of scapular motion increases after 135 of elevation (Karduna et al., 2001). Thus, only data up to 135 of arm elevation was used for analysis. The dependent variables included GH elevation, scapular upward rotation (UR), clavicular elevation, clavicular plane motion (protraction– retraction), scapular external rotation (ER), and scapular posterior tilt (PT). Because the starting and ending values differed between subjects, we calculated the excursion or the difference between the initial and final orientation angle achieved during arm elevation for each dependent variable. Based on the maximum arm elevation in the involved limb, subjects were placed into one of two groups: group one (N = 8) included those whose maximum arm elevation was 75 or below; and group two (N = 8) included those whose maximum arm elevation was above 75. This division of subjects corresponded with clinical differences in active abduction observed in children with BPBP based on the Mallet Scale (Mallet, 1972); with group one associated with grade III and group two associated with grade IV classification. To examine the group and limb differences for the excursions displayed during 15–75 of arm elevation, 2 · 2 (group · limb) Analysis of Variance (ANOVA) with repeated measures on the second factor were calculated for each dependent variable listed above. We also calculated the percent contribution to arm elevation by subject, for each dependent variable, through 15–75 or an arc of 60. Therefore, the percent contribution = degree of excursion/60. To assess group and limb differences in the percent contribution for each dependent variable, 2 · 2
Fig. 1a–f shows the mean excursions for each dependent variable during 15–75 of arm elevation. Table 2 lists the statistical findings. As shown for GH elevation (Fig. 1a), a group · limb interaction was found (P < 0.0001) with main effects for group and limb (both, P < 0.0001). Posthoc tests indicated that GH joint excursion in the involved limb of group one was lower than the non-involved and both limbs of group two (all, P < 0.05). Excursion in the clavicular plane (Fig. 1d), was greater in group one (P < 0.01) and the involved limb (P < 0.0002). Excursion into scapular ER (Fig. 1e), was greater for group one (P < 0.05) and for scapular UR (Fig. 1b), it was greater in the involved limb (P < 0.05). Excursion for clavicular elevation (Fig. 1c) and scapular PT (Fig. 1f) were not significantly different for group or limb. Fig. 2 shows the percent contribution each dependent variable made toward 15–75 of arm elevation among four subgroups clustered by group and limb (Fig. 2a–d). Statistical findings are listed in Table 3. In general, the percent contribution from GH joint elevation was lower and for most ST variables was greater in the involved limb of group one (Fig. 2a) when compared to the non-involved (Fig. 2b), and both limbs of group two (Fig. 2c and d). For GH elevation we found a group · limb interaction (P < 0.001) and main effects for group and limb (both, P < 0.0001). Post-hoc tests indicated that the contribution from GH elevation was lower in the involved limb of group one than the non-involved and both limbs of group two (all, P < 0.05). For clavicular elevation (Fig. 2d), we found a group · limb interaction (P < 0.05) and a main effect for limb (P < 0.01) only. Post-hoc tests revealed that clavicular elevation made a greater contribution to arm elevation in
S.V. Duff et al. / Clinical Biomechanics 22 (2007) 630–638
633
Fig. 1. Mean excursion (Exc) (standard error of the mean or SEM) during 15–75 arm elevation for the involved (Inv) limb (left) and non-involved (NI) limb (right) of group 1 (solid line) and group 2 (dashed line) for all dependent variables: (a) GH elevation; (b) scapular UR; (c) clavicular elevation; (d) clavicular plane motion; (e) scapular ER; and (f) scapular PT. See Table 2 for F ratios and significant P-values.
Table 2 F ratios and significant P-values (0.05*, P0.01**) for differences between group and limb in the mean excursion of dependent variables during 15–75 arm elevation Excursion
GH elevation Scapular UR Clavicular elevation Clavicular plane motion Scapular ER Scapular PT
Group · limb interaction
Main effect group (1 vs. 2)
Main effect limb (Inv vs. NI)
F ratio
P-value
F ratio
P-value
F ratio
P-value
28.4 .49 1.37 1.34 1.55 1.53
0.0001**
40.77 .71 .50 9.6 7.32 .68
0.0001**
28.6 8.05 3.08 23.96 1.34 4.4
0.0001** 0.01** 0.10 0.0002** 0.27 0.06
0.50 0.26 0.27 0.23 0.24
0.41 0.49 0.01** 0.02* 0.42
See Fig. 1.
the involved limb than the non-involved and both limbs of group two (all, P < 0.05). For clavicular plane motion (Fig. 2d), we found a group · limb interaction (P < 0.05) and main effects for group and limb (both, P < 0.0001). Post-hoc tests revealed that clavicular plane motion made a greater contribution in the involved limb of group one than the non-involved and both limbs of group two (all, P < 0.05). The contribution from scapular ER was greater in group one (P < 0.005). Scapular UR (P < 0.002) and scapular PT (P < 0.05) made greater contributions in the involved limb. 3.2. Scapulohumeral rhythm Fig. 3a displays the GH:ST ratio for all subjects across group and limb for 15–75 arm elevation. As shown, the
ratio in the involved limb of most subjects in group one was lower than the non-involved limb. The involved limb of group one was also lower than both limbs of group two. Individual ratios were supported by the mean GH:ST ratio (Fig. 3b and Table 4), based on a group · limb interaction (P < 0.001) and a main effect for limb (P < 0.0001). Post-hoc tests indicated that the GH:ST ratio in the involved limb of group one (0.6:1) was lower than the non-involved (2.2:1) (P < 0.05) and both the involved (1.7:1) (P < 0.05) and non-involved (1.9:1) (P < 0.05) limbs of group two. Z-scores for individual GH:ST ratios ranged from 1.5 above and below the mean (Fig. 3c), suggesting the distributions were generally consistent and stable. Fig. 4a shows the GH:ST ratio for 15–135 arm elevation for each subject across limbs for group two only. Overall, the excursions in the involved limb skewed slightly
634
S.V. Duff et al. / Clinical Biomechanics 22 (2007) 630–638
Fig. 2. Percent (%) contribution to 15–75 arm elevation (excursion/60), mean (SEM) of dependent variables within 4 clusters including: (a) Involved limb – Group one; (b) Non-involved limb – Group two; (c) Involved limb – Group two; and (d) Non-involved – Group two. The dependent variables in each of the four clusters are shown from left to right as follows: GH elevation (black); scapular UR (dark gray); clavicular elevation (medium-dark gray); clavicular plane motion (medium-light gray); scapular ER (light gray); and scapular posterior tilt (white). See Table 3 for F ratios and significant P-values.
Table 3 F ratios and significant P-values (0.05*, P0.01**) for differences between group and limb in the percent contribution of dependent variables during 15–75 arm elevation Percent contribution
Group · limb interaction F ratio
P-value
F ratio
P-value
F ratio
P-value
GH Elevation Scapular UR Clavicular elevation Clavicular plane motion Scapular ER Scapular PT
46.11 3.99 5.14 7.9 2.3 2.64
0.0001** 0.07 0.05* 0.05* 0.16 0.13
16.04 2.2 1.24 29.65 11.64 2.84
0.001** 0.16 .28 0.0001** 0.005** 0.11
38.9 15.22 8.97 44.3 2.22 5.62
0.0001** 0.002** 0.01** 0.0001** 0.16 0.05*
Main effect group (1 vs. 2)
Main effect limb (Inv vs. NI)
See Fig. 2.
lower and the range was narrower than the non-involved limb. As shown in Fig. 4b, the mean GH:ST ratio for 15–135 of arm elevation did not significantly differ (F = 2.88; P > 0.05) between the involved (1.5:1) and non-involved (2:1) limbs. Individual z-scores for the GH:ST ratio ranged from 1.5 above and below the mean for both limbs (Fig. 4c), suggesting the distributions were generally consistent and stable. 4. Discussion 4.1. Altered movement patterns Kinematic findings from this study support the clinical observation that children with BPBP employ greater than average scapular motion in the involved limb during arm elevation. The results also suggest that the relative contributions from the GH and ST joints through 15–75 depend on the amount of active arm elevation available. For the children with limited elevation (group one), the ST joint in the involved limb made a larger contribution than the GH joint, resulting in a much lower GH:ST ratio. In children with greater elevation (group two), the GH joint in the involved limb made a larger contribution than the ST joint, resulting in a higher ratio. Interestingly, for 15–135 arm elevation, the GH:ST ratio in the involved limb (1.5:1)
resembled the ratio reported for the dominant limb of typical children (1.3:1) during 25–125 arm elevation (Dayanidhi et al., 2005). A greater contribution from the ST joint during arm elevation has also been observed in adults with rotator cuff tears (Mell et al., 2005) and induced shoulder muscle fatigue (Ebaugh et al., 2006; McQuade et al., 1998). As in the case of adults with rotator cuff tears, the altered scapulohumeral pattern found in children with BPBP in group one may have allowed the intact muscles to function in a more effective portion of the length-tension curve, enhancing the strength of the muscle contractions used to lift the arm. During this study, we controlled the velocity by pacing the lift. It is possible that this demand prevented the children with BPBP from using a common strategy used to enhance muscle power; that of increasing the velocity of a lift to compensate for weakness (Burkholder and Leiber, 2001). In particular, children in group one, with less active elevation, may routinely lift the involved arm quickly when the demand arises or they may avoid lifting it because they cannot generate sufficient muscle power to do so. 4.2. Achieving dynamic shoulder stability Dynamic shoulder stability can be defined as centering of the humeral head on the glenoid fossa accompanied by
S.V. Duff et al. / Clinical Biomechanics 22 (2007) 630–638
635
humeral head internally rotates and shifts posteriorly in the glenoid fossa (Waters et al., 1998), altering the line of muscle pull, or action line. This biomechanical alteration could diminish the contraction strength of the external rotators even if they become re-innervated. The variation in muscle length and strength may partly account for group differences in the scapulohumeral pattern found in this study (Langenderfer et al., 2006). The children in group two may be able to effectively activate the shoulder external rotators needed to stabilize and rotate the humerus, whereas, the children in group one may be unable to do so. Obligatory resting postures and habitual use patterns contribute to the evolution of skeletal deformities (Poyhia et al., 2005; Waters, 1997). These deformities include osseous changes in the glenoid fossa and humeral head, progressive posterior dislocation of the humeral head, and shortening of the anterior GH joint capsule (Hoeksma et al., 2000, 2003; Kozin, 2004; Pearl and Edgerton, 1998; Poyhia et al., 2005; Soldado and Kozin, 2005; Waters et al., 1998). Typically, if the humerus cannot externally rotate or glide inferiorly, it may be able to achieve only 60–90 of elevation before the glenoid tubercle hits the acromion (Schenkman and Rugo De Cartaya, 1987). Because of muscle weakness and skeletal deformities in children with BPBP, the greater tubercle remains vulnerable,
Fig. 3. GH:ST ratio during 15–75 arm elevation for both groups: 15(a) individual subject GH:ST ratios for the involved (solid circles) and noninvolved (open circles) from group 1 (left) and group 2 (right); (b) mean (SEM) GH:ST ratio for the involved (Inv) limb (left) and non-involved (NI) limb (right) of group 1 (solid line) and group 2 (dashed line); and (c) z-scores of all subjects for GH:ST ratio for the involved (solid circles) and non-involved (open circles) limbs from group 1 (bottom) and group 2 (top). See Table 4 for F ratios and significant P values.
Table 4 F ratios and significant P-values (0.05*, P0.01**) for differences between group and limb in the GH:ST ratio during 15–75 arm elevation Arm elevation 15–75
Group · limb interaction
Main effect group (1 vs. 2)
Main effect limb (Inv vs. NI)
F ratio
P-value
F ratio
P-value
F ratio
P-value
GH:ST ratio
17.35
0.001*
3.07
0.1
30.44
0.0001*
See Fig. 3.
inferior translation via the rotator cuff muscles to facilitate humeral rotation. The shoulder weakness and atrophy found in children with BPBP seems to alter this stability. Because of weak shoulder external rotators, children with BPBP often present with an obligatory resting posture of humeral internal rotation. As posturing into internal rotation progresses, the external rotators (e.g., infraspinatous and teres minor) lengthen, and the internal rotators (e.g., subscapularis) shorten. With muscle length changes, the
Fig. 4. GH:ST ratio during 15–135 arm elevation for group 2 only: (a) individual subject GH:ST ratios for the involved (solid circles, bottom) and non-involved (open circles, top) limbs; (b) mean (SEM) GH:ST ratio for the involved (Inv) limb (left) and non-involved (NI) limb (right) which were not significantly different; and (c) z-scores of all subjects in group 2 for GH:ST ratio with the involved (solid circles, bottom) and non-involved (open circles, top) limbs.
636
S.V. Duff et al. / Clinical Biomechanics 22 (2007) 630–638
and repetitive contact with the acromion could increase the risk for impingement syndrome (McClure et al., 2006). Enhanced scapular UR and clavicular elevation, particularly in group two, may in part keep the acromion out of the way of the glenoid tubercle allowing for greater humeral elevation and dynamic stability. Alterations in muscle power and musculoskeletal integrity may differentially impact the dynamic joint stability and movement patterns displayed by each group. To further elucidate the group and limb differences, imaging techniques and EMG analysis would need to be employed (Illyes and Kiss, 2006). 4.3. Neural adaptation Neonatal nerve injury may be more detrimental than adult nerve injury because peripheral lesions may deprive central neurons of important trophic factors, leading to cell death and smaller neural representations (Cusick, 1996). Group differences in this study may stem from the degree of initial nerve injury along with variable neural recovery and adaptation. As shown in Table 1, 3 out of 8 children in group one had been given an initial diagnosis of Extended Erb’s palsy, suggesting they had greater nerve injury at birth and/or less neural recovery post-natally. The central nervous system adapts to alterations in deafferentation and use (Florence et al., 1998; Garraghty and Kaas, 1991; Merzenich et al., 1983; Merzenich and Jenkins, 1993). Thus, if the involved limb of a child with BPBP is weak, and sensibility is reduced, he or she may be less apt to use it, contributing to disuse atrophy and further loss of neural representations (Cusick, 1996; Taub et al., 1975). The degree of initial injury and subsequent recovery as well as differences in neural representations would likely influence shoulder movement patterns, thus may be partly responsible for group differences. 4.4. Clinical relevance Despite muscle imbalances and skeletal changes, children in both groups are reportedly quite functional. Findings from this study suggest that the movement strategy of increased scapular UR observed in children who have not had surgical intervention might provide functional benefit because of the increased active range it affords. Key goals for treatment of this population are to limit the development of musculoskeletal deformity, to strengthen recovering muscles, and to promote active use of the involved limb. From early infancy, appropriate stretching and mobilization programs for scapulohumeral muscles, such as the shoulder internal rotators and the GH joint capsule, should be emphasized to prevent the development of joint contractures (Ramos and Zell, 2000; Waters, 2005). Along with passive range-of-motion, strengthening of recovering and potential transfer muscles such as the latissimus dorsi (Waters and Bae, 2005) should be encouraged. If passive range is not maintained or activation of partially innervated muscles is not promoted mus-
culoskeletal deformities will worsen (Ramos and Zell, 2000). Furthermore, active use and progression through motor milestones should be reinforced. If unimanual function is limited, bimanual task practice may be fostered. As with other neurological injuries, it is important to prevent disuse muscle atrophy and reinforce neural representations needed for adequate motor control (Thomas et al., 1997). Criteria for surgical intervention are linked to muscle/ skeletal integrity and functional needs. Growth spurts in adolescence have been found to increase skeletal deformities, including a progressive increase in glenoscapular angle and posterior subluxation of the humeral head (Waters et al., 1998). Because the progression in skeletal deformity may further impact movement patterns and function, surgical intervention is now being conducted in younger patients including infants and toddlers. One such surgery is the release of the joint capsule and tendon transfer of the latissimus dorsi and teres major to a posterior/lateral position high on the rotator cuff tendon(s) (Waters and Bae, 2005). This transfer allows former internal rotators to act as external rotators and improves the degree of functional arm elevation. Severe deformities in older children may require an osteotomy of the humerus with external rotation into a more neutral alignment to improve arm elevation for overhead tasks such as hair grooming (Waters, 1997). Differences in muscle activation, skeletal integrity and subsequent movement patterns between the two groups examined in this study might provide a basis for the selection of surgical candidates and the type of surgical procedure required. As suggested by Mosqueda and colleagues (2004), the inclusion of kinematic analysis in our study allowed for close examination of the interlimb differences in joint contribution to arm elevation. Clinicians and researchers may be able to utilize the magnetic tracking device as an adjunct to clinical scales such as the Mallet Scale (Mallet, 1972) to document baseline performance along with outcomes from therapeutic and surgical intervention. The additional use of EMG and imaging techniques would allow tracking of sequential muscle activation and GH and ST joint rotation during various phases of arm elevation. 5. Conclusion This study provided support for the clinical observation of interlimb differences in shoulder movement patterns in the involved and non-involved limbs of children with brachial plexus birth palsy. Children with limited arm elevation (675) in the involved limb displayed a greater contribution from the ST joint than the GH joint when compared to children with greater arm elevation (>75). However, both groups displayed greater scapular UR in the involved limb. Although enhanced scapular mobility is likely a consequence of muscle imbalance, skeletal deformity, and neural adaptation, it serves to enhance function during the performance of unimanual and bimanual tasks. Given better tools for evaluating the impact of nerve injury
S.V. Duff et al. / Clinical Biomechanics 22 (2007) 630–638
on shoulder movement patterns, we may be able to refine methods of therapeutic and surgical intervention to enhance neuromuscular resources and movement strategies for those with brachial plexus birth palsy. Acknowledgements We would like to thank Marcus Besser, PhD, Andrew Karduna, PhD, and Patricia A. Shewokis, PhD for their thoughtful reviews and suggestions. References Burkholder, T.J., Leiber, R.L., 2001. Sarcomere length operating range of vertebrate muscles during movement. J Exp. Biol. 204, 1529– 1536. Codman, E.A., 1934. The shoulder. Thomas Todd, Boston. Curtis, C., Stephens, D., Clarke, H.M., Andrews, D., 2002. The active movement scale: an evaluative tool for infants with obstetrical brachial plexus palsy. J. Hand Surg. [Am]. 27 (3), 470–478. Cusick, C.G., 1996. Extensive cortical reorganization following sciatic nerve injury in adult rats versus restricted reorganization after neonatal injury: implications for spatial and temporal limits on somatosensory plasticity. Prog. Brain Res. 108, 379–390. Dayanidhi, S., Orlin, M., Kozin, S., Duff, S., Karduna, A., 2005. Scapular kinematics during humeral elevation in adults and children. Clin. Biomech. 20 (6), 600–606. Dodds, S.D., Wolfe, S.W., 2000. Perinatal brachial plexus palsy. Curr. Opin. Pediatr. 12 (1), 40–47. Ebaugh, D.D., McClure, P.W., Karduna, A.R., 2006. Three-dimensional scapulothoracic motion during active and passive arm elevation. Clin. Biomech. 20, 700–709. Florence, S.L., Taub, H.B., Kaas, J.H., 1998. Large-scale sprouting of cortical connections after peripheral nerve injury in adult macaque monkeys. Science 282 (5391), 1117–1121. Garraghty, P.E., Kaas, J.H., 1991. Large-scale functional reorganization in adult monkey cortex after peripheral nerve injury. Proc. Natl. Acad. Sci. USA 88 (16), 6976–6980. Graichen, H., Stammberger, T., Bonel, H., Haubner, M., Englmeier, K.H., Reiser, M., Eckstein, F., 2000. Magnetic resonance-based motion analysis of the shoulder during elevation. Clin. Orthop. Relat. Res. 370, 154–163. Hoeksma, A.F., Ter Steeg, A.M., Dijkstra, P., Nelissen, R.G., Beelen, A., de Jong, B.A., 2003. Shoulder contracture and osseous deformity in obstetrical brachial plexus injuries. J. Bone Joint Surg. 85A (2), 316– 322. Hoeksma, A.F., Wolf, H., Oei, S.L., 2000. Obstetrical brachial plexus injuries: incidence, natural course and shoulder contracture. Clin. Rehab. 14 (5), 523–526. Illyes, A., Kiss, R.M., 2006. [Epub]. Shoulder joint kinematics during elevation measured by ultrasound-based measuring system. J. Electromyogr. Kinesiol. Apr 16. Inman, V.T., Saunders, J.B., Abbott, L.C., 1944. Observations on the function of the shoulder joint. J. Bone Joint Surg. 26, 1–30. Karduna, R.A., McClure, P.W., Michener, L.A., Sennett, B., 2001. Dynamic measurement of three-dimensional scapular kinematics: a validation study. J. Biomech. Eng. 123, 184–190. Karduna, R.A., McClure, P.W., Michener, L.A., 2000. Scapular kinematics: effects of altering the Euler angle sequence of rotations. J. Biomech. 33, 1063–1068. Kozin, S.H., 2004. Correlation between external rotation of the glenohumeral joint and deformity after brachial plexus birth palsy. J. Ped. Orthop. 24 (2), 189–193. Kozin, S.H., Chafetz, R.S., Barus, D., Filipone, L., 2006. Magnetic resonance imaging and clinical findings before and after tendon
637
transfers about the shoulder in children with residual brachial plexus birth palsy. J. Shoulder Elbow Surg. 15 (5), 554–561. Langenderfer, J.E., Patthanacharenphon, C., Carpenter, J.E., Hughes, R.E., 2006. Variability in isometric force and moment during generating capability of glenohumeral external rotator muscles. Clin. Biomech. 21 (7), 701–709. Ludewig, P.M., Cook, T.M., 2000. Alterations in shoulder kinematics and associated muscle activity in people with symptoms of shoulder impingement. Phys. Ther. 80 (3), 276–291. Ludewig, P.M., Cook, T.M., Nawoczenski, D.A., 1996. Three-dimensional scapular orientation and muscle activity at selected positions of humeral elevation. J. Orthop. Sports Phys. Ther. 24 (2), 57–65. Mallet, J., 1972. Primaute du traitement de l’epaule – methode d’expresion des resultats. Rev. Chir. Orthop. 58S, 166–168. McClure, P., Michener, L.A., Karduna, A.R., 2006. Shoulder function and 3-dimensional scapular kinematics in people with and without shoulder impingement syndrome. Phys. Ther. 86 (8), 1075–1090. McClure, P., Michener, L.A., Sennett, B.J., Karduna, A.R., 2001. Direct 3-dimensional measurement of scapular kinematics during dynamic movements during in vivo. J. Shoulder Elb. Surg. 10, 269–277. McQuade, K.J., Dawson, J., Smidt, G.L., 1998. Scapulothoracic muscle fatigue associated with alterations in scapulohumeral rhythm kinematics during maximum resistive shoulder elevation. J. Orthop. Sports Phys. Ther. 28 (2), 74–80. Mell, A.G., LaScalza, S., Guffey, P., Ray, J., Maciejewski, M., Carpenter, J.E., Hughes, R.E., 2005. Effect of rotator cuff pathology on shoulder rhythm. J. Shoulder Elbow Surg. 14 (1S), 58S–64S. Merzenich, M.M., Jenkins, W.M., 1993. Reorganization of cortical representations of the hand following alterations of skin inputs induced by nerve injury, skin island transfers, and experience. J. Hand Ther. 6 (2), 89–104. Merzenich, M.M., Kaas, J.H., Wall, J., Nelson, R.J., Sur, M., Felleman, D., 1983. Topographic reorganization of somatosensory cortical areas 3b and 1 in adult monkeys following restricted deafferentation. Neuroscience 8 (1), 33–55. Michelow, B.J., Clarke, H.M., Curtis, C.G., Zuker, R.M., Seifu, Y., Andrews, D.F., 1994. The natural history of obstetrical brachial plexus palsy. Plast. Reconstr. Surg. 93 (4), 675–680. Michiels, I., Grevenstein, J., 1995. Kinematics of shoulder arm elevation in the scapular plane. On the influence of arm elevation velocity and external load. Clin. Biomech. 10 (3), 137–143. Mollberg, M., Hagberg, H., Bager, B., Lilja, H., Ladfors, L., 2005. High birthweight and shoulder dystonia: the strongest risk factors for obstetrical brachial plexus palsy in a Swedish population-based study. Acta Obstet. Gynecol. Scand. 84 (7), 654–659. Moriwaki, M., 1992. Analysis of three dimensional motion of the scapula and the glenohumeral joint. J. Japanese Orthop. Assoc. 66, 675–687. Mosqueda, T., James, M.A., Petuskey, K., Bagley, A., Abdala, E., Rab, G., 2004. Kinematic assessment of the upper extremity in brachial plexus birth palsy. J. Ped. Orthop. 24 (6), 695–699. Pearl, M.L., Edgerton, B.W., 1998. Glenoid deformity secondary brachial plexus birth palsy. J. Bone Joint Surg. [Am] 80 (5), 659–667. Poppen, N., Walker, P., 1976. Normal and abnormal motion of the shoulder. J. Bone Joint Surg. [Am] 58 (2), 195–201. Poyhia, T.H., Nietosvaara, Y.A., Remes, V.M., Kirjavainen, M.O., Peltonen, J.I., Lamminen, A.E., 2005. MRI of rotator cuff muscle atrophy in relation to glenohumeral joint incongruence in brachial plexus birth injury. Ped. Radiol. 35 (4), 402–409. Price, A.E., Grossman, J.A., 1995. A management approach for secondary shoulder and forelimb deformities following obstetrical brachial plexus injury. Hand Clin. 11 (4), 607–617. Ramos, L.E., Zell, J.P., 2000. Rehabilitation program for children with brachial plexus and peripheral nerve injury. Semin Pediatr Neurol. 7 (1), 52–57. Schenkman, M., Rugo De Cartaya, V., 1987. Kinesiology of the shoulder complex. J. Ortho. Sports Phys. Ther. 8 (9), 438–450. Strombeck, C., Krumlinde-Sundholm, L., Forssberg, H., 2000. Functional outcome at 5 years in children with obstetrical brachial plexus palsy
638
S.V. Duff et al. / Clinical Biomechanics 22 (2007) 630–638
with and without microsurgical reconstruction. Dev. Med. Child Neurol. 42 (3), 148–157. Sugamoto, K., Harada, T., Machida, A., Inui, H., Miyamota, T., Takeuchi, E., Yoshikawa, H., Ochi, T., 2002. Scapulohumeral rhythm: relationship between motion velocity and rhythm. Clin. Orthop. Relat. Res. 401, 119–124. Sundholm, L.K., Eliasson, A.C., Forssberg, H., 1998. Obstetric brachial plexus injuries: assessment protocol and functional outcome at age 5 years. Dev. Med. Child Neurol. 40 (1), 4–11. Soldado, F., Kozin, S.H., 2005. The relationship between the coracoid and glenoid after brachial plexus birth palsy. J. Ped. Orthop. 25 (5), 666–670. Taub, E., Goldberg, I.A., Taub, P., 1975. Deafferentation in monkeys: pointing at a target without visual feedback. Exp. Neurol. 46, 176–186. Thomas, C.K., Zaidner, E.Y., Calancie, B., Broton, J.G., Bigland-Ritchie, B.R., 1997. Muscle weakness, paralysis, and atrophy after human cervical spinal cord injury. Exp. Neurol. 148 (2), 414–423.
Waters, P.M., 1997. Obstetric brachial plexus injuries: evaluation and management. J. Am. Acad. Orthop. Surg. 5, 205–214. Waters, P.M., 1999. Comparison of the natural history, the outcome of microsurgical repair, and the outcome of operative reconstruction in brachial plexus birth palsy. J. Bone Joint Surg. [Am] 81 (5), 649– 659. Waters, P.M., 2005. Update on management of pediatric brachial plexus palsy. J. Ped. Orthop. 25 (1), 116–126. Waters, P.M., Bae, D.S., 2005. Effect of tendon transfers and extraarticular soft-tissue balancing on glenohumeral development in brachial plexus birth palsy. J. Bone Joint Surg. [Am] 87 (2), 320– 325. Waters, P.M., Smith, G.R., Jaramillo, D., 1998. Glenohumeral deformity secondary to brachial plexus birth palsy. J. Bone Joint Surg. [Am] 80 (5), 668–677.