• Effects of distal hamstring lengthening on sagittal motion in patients with diplegia

  • FileName: 2009_park.pdf [read-online]
    • Abstract: sagittal kinematics, and to investigate the validity of modeled hamstring length for clinical use. ... Please cite this article in press as: Park MS, et al. Effects of distal hamstring lengthening on sagittal motion in patients with diplegia. ...

Download the ebook

G Model
GAIPOS-2832; No of Pages 5
Gait & Posture xxx (2009) xxx–xxx
Contents lists available at ScienceDirect
Gait & Posture
journal homepage: www.elsevier.com/locate/gaitpost
Effects of distal hamstring lengthening on sagittal motion in patients with diplegia
Hamstring length and its clinical use§
Moon Seok Park a, Chin Youb Chung a, Sang Hyeong Lee a, In Ho Choi b, Tae-Joon Cho b, Won Joon Yoo b,
B.S. Myoung Yl Park c, Kyoung Min Lee a,*
Department of Orthopedic Surgery, Seoul National University Bundang Hospital, 300 Gumi-Dong, Bundang-Gu, Sungnam, Kyungki 463-707, Republic of Korea
Department of Orthopedic Surgery, Seoul National University Children’s Hospital, 28 Yongun-Dong, Chongro-Gu, Seoul 110-744, Republic of Korea
Motion Analysis Korea, 642-1 Yeoksam-Dong, Kangnam-Gu, Seoul 135-080, Republic of Korea
Article history: This study was undertaken to determine the effect of distal hamstring lengthening (DHL) on hip and knee
Received 29 November 2008 sagittal kinematics, and to investigate the validity of modeled hamstring length for clinical use. Patient
Received in revised form 1 July 2009 group consisted of 28 patients (56 limbs, mean age 7.4 years) with spastic diplegia who underwent
Accepted 14 July 2009
bilateral DHL and tendo-Achilles lengthening with/without rectus femoris transfer (RFT) (DHL + RFT
subgroup, 40 limbs; DHL subgroup, 16 limbs). Kinematic data was obtained by gait analysis, and
Keywords: hamstring lengths were obtained using a musculoskeletal modeling technique. Postoperatively, knee
Cerebral palsy
extension improved (p < 0.001) without aggravating anterior pelvic tilt (p = 0.565). However, DHL
Single event multilevel surgery
Hamstring length
aggravated anterior pelvic tilt in the DHL subgroup (2.28, p = 0.011). In terms of concurrent validity,
Distal hamstring lengthening hamstring length was found to be correlated with mean pelvic tilt (r = 0.798, p < 0.001) and popliteal
Rectus femoris transfer angle (r = À0.425, p = 0.001), but the correlation between hamstring length and knee flexion at initial
Pelvic tilt contact was minimal (r = 0.068, p = 0.753). In terms of construct validity, DHL did not lengthen mean
Kinematics hamstring length (p = 0.918). In conclusion, DHL appeared to significantly improve knee motion in
patients with spastic diplegia. Furthermore, DHL did not increase pelvic tilt, when performed with RFT.
Modeled hamstring length is believed to have limited validity in patients with cerebral palsy, because it
does not reflect knee kinematics or postoperative change when DHL was combined with multilevel
ß 2009 Published by Elsevier B.V.
1. Introduction context of single event multilevel surgery, which is a standard
treatment for patients with cerebral palsy. However, although this
Distal hamstring lengthening (DHL) is one of the most common integration makes it difficult to evaluate the effect of DHL in
operations performed in patients with cerebral palsy. Previous isolation, we believed that we could investigate the effect of the
reports have shown that DHL is effective at reducing knee flexion procedure more precisely by controlling for confounding surgeries.
and improving knee motion [1–8]. However, there have been In this study, we hypothesized that rectus femoris transfer (RFT)
concerns that this procedure might aggravate anterior pelvic tilt, importantly confounds the effects of DHL, because the rectus
lumbar hyperlordosis, knee hyperextension, and eventually induce femoris probably counteracts the hamstring muscle on sagittal
crouch gait [2,9,10]. In addition, the need for DHL has been kinematics.
challenged because hamstring length was not found to be shorter Moreover, as far as we are aware, few investigations have been
in patients with a crouch gait [11–13], and because the procedure conducted on changes in hamstring length after DHL [14]. The
can increase the lengths of already long muscles. validity of ‘hamstring length’, as defined by 3D graphic models
However, the majority of studies that have investigated the [15], has been proposed to assist in decision-making regarding
effect of DHL [1–8,10] have included various potentially con- DHL, but this has not been clarified. Furthermore, previous studies
founding surgeries, because DHL is usually performed in the also included a large number of confounding surgical procedures.
Therefore, the primary aim of this study was to determine the
effect of DHL on hip and knee kinematics in a relatively
This study was conducted at Seoul National University Bundang Hospital.
homogeneous group of patients with cerebral palsy after control-
* Corresponding author. Tel.: +82 31 787 6256; fax: +82 31 787 4056. ling for confounding surgeries. The second aim was to clarify the
E-mail address: [email protected] (K.M. Lee). validity of modeled ‘hamstring length’ for clinical usage. In validity
0966-6362/$ – see front matter ß 2009 Published by Elsevier B.V.
Please cite this article in press as: Park MS, et al. Effects of distal hamstring lengthening on sagittal motion in patients with diplegia. Gait
Posture (2009), doi:10.1016/j.gaitpost.2009.07.115
G Model
GAIPOS-2832; No of Pages 5
2 M.S. Park et al. / Gait & Posture xxx (2009) xxx–xxx
test, the concurrent validity was examined by comparing ham-
string length with hip and knee kinematics and popliteal angle, and
the construct validity was determined by analyzing postoperative
hamstring length changes.
2. Materials and methods
2.1. Inclusion/exclusion criteria and operative procedures
This retrospective study was approved by the institutional review board at our
institute (a tertiary referral center for cerebral palsy), which waived the need for
informed consent. The inclusion criteria used were as follows; ambulatory patients
with spastic diplegia (GMFCS level I–III), a history of single event multilevel surgery
between January 1998 and April 2006 with a follow-up period of over a year, and
the availability of preoperative and postoperative 1-year gait analysis results. 288
patients fulfilled the inclusion criteria. The exclusion criteria applied were; a history
of gait correcting surgery or selective dorsal rhizotomy, or a concurrent
neuromuscular disease other than cerebral palsy. To ensure homogeneity, patients
with asymmetric gait patterns who had undergone asymmetric procedures or
procedures other than tendo-Achilles lengthening or DHL with/without RFT were
also excluded. 28 patients (56 limbs) were allocated to patient group and were
divided into two subgroups. Patients who had undergone bilateral DHL and tendo-
Achilles lengthening were allocated to DHL subgroup, and the other patients who
had undergone bilateral DHL, RFT, and tendo-Achilles lengthening were allocated to
DHL + RFT subgroup. In addition, 12 healthy children (24 limbs) and 34 healthy
adult volunteers (68 limbs) were recruited as child and adult control groups,
All surgical procedures were performed by a single surgeon as single event
multilevel surgeries. Preoperative gait analysis revealed that all patients had a jump
gait pattern, as described by Rodda et al. [16]. All patients underwent bilateral
tendo-Achilles lengthening and DHL, the latter of which was composed of gracilis
aponeurotic lengthening, semitendinosus tendon transfer to the adductor magnus,
and aponeurotic lengthening of the semimembranosus. Accordingly, DHL was
performed only on the medial side. For patients with stiff-knee gait patterns by
preoperative gait analysis, RFT to the gracilis was also performed [17]. Following
surgery, all patients underwent 3 weeks of immobilization with a short leg cast.
Standing and weight bearing were resumed with leaf-spring type ankle foot
orthoses, which were worn for the next 3 months. Subsequently, patients were
referred back to a local rehabilitation center to continue muscle-strengthening
exercises and gait training. During this period, ankle foot orthoses were
recommended at night only to prevent the recurrence of Achilles tightness.
2.2. Acquisition of kinematic data
Gait analysis was performed a few days before surgery using a Vicon 370 (Oxford
Metrix, Oxford, UK) equipped with seven cameras and two force plates, and was
repeated after more than 1-year postoperatively. Kinematic data were archived as
patients walked barefoot on a 9-m walkway. Three trials were averaged to determine
the values of index variables. Preoperative kinematic variables were compared
between adult control group and child control group and then between child control
group and the patient group. Pre- and postoperative kinematic variables were
compared to determine the effect of DHL on hip and knee kinematics in the patient
group. Thereafter, subgroup analysis was done on each patient subgroup.
2.3. Validity of hamstring length
‘Hamstring lengths’ were determined using interactive musculoskeletal model-
ing software (SIMM, Motion Analysis Corporation, Santa Rosa, CA) [18] (Fig. 1A and
B). Hamstring length was defined as the distance between its muscular origin and
insertion. To standardize muscle lengths, hamstring lengths were divided by
muscle lengths when the knee and hip were in the anatomic position [13]. The
length of the semimembranosus was used as a representative hamstring length and
charts were prepared of standardized hamstring lengths during the gait cycle.
Concurrent validity of modeled hamstring length was achieved by correlating
hamstring lengths with reference variables, namely, mean pelvic tilt, knee flexion at
initial stance, maximum hip flexion, and popliteal angle. To determine construct
validity preoperative and postoperative hamstring lengths were compared.
2.4. Statistical analysis
Data were analyzed using SPSS Ver. 15.0 (SPSS, Chicago, IL). The normalities of
distributions were tested for each variable using the Kolmogorov–Smirnov test.
Fig. 1. Three dimensional musculoskeletal modeling images depicting hamstring
muscles between their bony origins and insertions (A and B) and a graph of
standardized semimembranosus length which was used as a modeled hamstring by dividing changing hamstring lengths during gait by static hamstring length in
length (C). The three muscles are the gracilis, semimembranosus, and the long head the anatomic position. (C) This graph represents continuous changes in average
of the biceps femoris. (A) Computer model in the anatomic position with the knee standardized semimembranosus length (solid line) and standard deviations (dotted
and hip joint in 08 of extension. Static hamstring lengths were measured in this lines) during the gait cycle in the ‘normal’ control group. The minimum hamstring
position. (B) Hamstring length (distance between its bony origin and insertion) length occurred at maximum knee flexion, and maximum length during the late
changed throughout the gait cycle. Standardized hamstring length was calculated swing phase.
Please cite this article in press as: Park MS, et al. Effects of distal hamstring lengthening on sagittal motion in patients with diplegia. Gait
Posture (2009), doi:10.1016/j.gaitpost.2009.07.115
G Model
GAIPOS-2832; No of Pages 5
M.S. Park et al. / Gait & Posture xxx (2009) xxx–xxx 3
Table 1 than in the adult control group (p < 0.001, p < 0.001). Preoperative
Demographics and preoperative physical examinations of the study group.
mean anterior pelvic tilts, maximum hip flexions in terminal swing
DHL DHL + RFT p-value Total and knee flexions at initial contact in the patient group were
subgroup subgroup significantly larger than in the child control group (p < 0.001,
N (limb) 8 (16) 20 (40) 28 (56) p < 0.001 and p < 0.001, respectively). No significant differences
M:F 7:1 15:5 0.640* 22:6 were found preoperatively between the two patient subgroups in
GMFCS (I/II) 5/3 15/5 0.651* 20/8 terms of anterior pelvic tilt, maximum hip flexion in terminal
Mean age (SD) at 7.3 (2.8) 7.4 (2.4) 0.888 7.4 (2.4)
swing, or knee flexion at initial contact (p = 0.829, p = 0.094, and
gait analysis
p = 0.492, respectively). However, maximum knee flexion in swing
Mean time interval 1.4 (0.8) 1.1 (0.3) 0.270 1.2 (0.5) was significantly smaller in the DHL + RFT subgroup (p = 0.013).
(SD) from surgery Knee flexion at initial contact and maximum knee flexion in
to postoperative swing significantly improved following surgery in patient group
gait analysis
(p < 0.001, p = 0.031) while the anterior pelvic tilt was unchanged
(p = 0.565). Anterior pelvic tilt increased significantly postopera-
Preoperative physical examination tively in the DHL subgroup (p = 0.011) but not in the DHL + RFT
Hip flexion (8) 138.8 (3.4) 138.9 (6.5) 0.909 138.9 (5.7)
Hip extension (8) 0.9 (2.0) 1.9 (3.6) 0.226 1.6 (3.2)
subgroup (p = 0.247). (Table 2).
Knee flexion (8) 145.6 (5.4) 145 (5.7) 0.714 145.2 (5.6)
Knee extension (8) 0.9 (2.7) 0.5 (4.2) 0.718 0.7 (3.8) 3.2. Hamstring length and validity for clinical use
Popliteal angle (8) 47.2 (20.3) 50.4 (13.7) 0.569 49.4 (15.8)
Ankle dorsiflexion 1.0 (19.1) À3.9 (11.5) 0.365 À2.5 (14.0)
Standardized hamstring lengths were charted through gait
(knee extension) (8)
Ankle dorsiflexion 10.3 (17.1) 5.8 (11.8) 0.347 7.2 (13.6) cycles. Maximum length occurred during the terminal swing phase
(knee flexion) (8) and minimum length at maximal knee flexion (Fig. 1C). Length
Ely test 11 Limbs 38 Limbs 0.016** 49 Limbs patterns of hamstring muscles concurred with those found in
Fisher’s exact test. similar studies [11,15].
p < 0.05 by Fisher’s exact test. Maximum hamstring length was significantly longer in child
control group than in adult control group (p = 0.003), while there
Standardized hamstring lengths, knee flexions, hip flexions, and pelvic tilts were was no significant difference in mean hamstring length between the
compared using the t-test. Patients were compared with respect to pre- and
two control groups (p = 0.433). Preoperatively, the mean hamstring
postoperative kinematic variables and hamstring lengths using the paired t-test.
Correlations between hamstring length, popliteal angle, and kinematic variables lengths of patients were longer than those of child control group
were tested using Pearson’s correlation test. p values of

Use: 0.0256