Clinical Biomechanics 25 (2010) 867872

Contents lists available at ScienceDirect

Clinical Biomechanics
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c l i n b i o m e c h

Biomechanical effects of the extent of sacrectomy on the stability of lumbo-iliac

reconstruction using iliac screw techniques: What level of sacrectomy requires the
bilateral dual iliac screw technique?
Bin-Sheng Yu a,, Xin-Ming Zhuang a, Ze-Min Li a, Zhao-Min Zheng a, Zhi-Yu Zhou a,
Xue-Nong Zou a, William W. Lu b,
a
Department of Spine Surgery, the First Afliated Hospital of Sun Yat-sen University, 183 Huangpu East Road, Guangzhou, China, 510700
b
Department of Orthopaedics and Traumatology, the University of Hong Kong, Hong Kong, China

a r t i c l e i n f o a b s t r a c t

Article history: Background: Although both single and dual iliac screw techniques are used in spino-pelvic reconstruction
Received 26 November 2009 following sacrectomy for treating sacral tumors, the basis for choosing between the two techniques for
Accepted 16 June 2010 different instability types remains undetermined. The purpose of this study was to evaluate the effects of the
extent of sacrectomy on the stability of the lumbo-iliac xation construct using single and dual iliac screw
Keywords: techniques.
Sacral tumor
Methods: Nine human L2-pelvic specimens were tested for their intact condition simulated by L3L5 pedicle
Sacrectomy
Lumbo-iliac xation
screw xation. Sequential partial sacrectomies and L3-iliac xation using bilateral single and dual iliac
Iliac screw screws were conducted on the same specimens as follows: under-S1 sacrectomy + single screw, under-S1
Biomechanics sacrectomy + single screw, one-side sacroiliac joint resection + single screw, total sacrectomy + single
screw, and total sacrectomy + dual screw. Biomechanical testing was performed on a material testing
machine for evaluating the stiffness of the L3-iliac xation construct in compression and torsion.
Findings: Single iliac screw technique was found to effectively restore the local stability in under-S1
sacrectomy. However, it could not provide adequate stability for further resection of one-side sacroiliac joint
in torsion and total sacrectomy in compression (P b 0.05). On the other hand, dual iliac screw technique could
restore the stability to the intact condition after total sacrectomy in both compression and torsion.
Interpretation: The single iliac screw technique for L3-iliac xation could effectively restore the local stability
for under-S1 sacrectomy. However, for instabilities of the under-S1 sacrectomy with one-side sacroiliac
joint resection or total sacrectomy, the dual iliac screw technique should be considered.
2010 Elsevier Ltd. All rights reserved.

1. Introduction Therefore, the extent of sacral resection may be an important factor

Spino-pelvic fusion with screw-rod instrumentation is commonly
adopted in the surgical treatment of sacral tumors requiring high level
sacrectomy (Acharya et al., 2008; Fujibayashi et al., 2007; Dickey et al.,
2005; Doita et al., 2003; Mindea et al., 2003; Xiao et al., 2003; Zhang et
al., 2003). Compared with the Galveston technique, the iliac screw is
able to increase the initial stability and simplify the operative
manipulation (Schwend et al., 2003). Although providing good clinical
and functional results for patients with partial sacrectomy and
preserved total (or partial) S1 (Xiao et al., 2003), the single iliac
screw technique frequently results in a considerable rate of iliac screw
failure (loosening and breakage) and nonunion when it is used for
total sacrectomy condition (Doita et al., 2003; Mindea et al., 2003).
Corresponding authors.
E-mail addresses: hpyubinsheng@hotmail.com (B.-S. Yu), wwlu@hku.hk (W.W. Lu).
that affects the stability of the lumbo-iliac xation construct.
To overcome the biomechanical limitations faced by the single iliac
screw technique, the dual iliac screw technique was developed
(Zhang et al., 2003). Yu et al. (2010) have conrmed the biomechan-
ical advantage of dual over single iliac screw in restoring the stability
of the lumbo-iliac xation construct in vertical and rotational planes.
Furthermore, several clinical studies (Acharya et al., 2008; Fujibayashi
et al., 2007; Dickey et al., 2005) have demonstrated that the dual iliac
screw technique provides good clinical results for patients with the
partial or total sacrectomy with no iliac screw failure. However,
clinical practices caution that the dual iliac screw technique may
increase bone stock loss, prominence of the instrumentation, and
screw-rod connection difculty as compared with the single iliac
screw technique (Moshirfar et al., 2005; Doita et al., 2003).
Consequently, single iliac screw technique should be preferred to
avoid these problems as long as sufcient stability can be maintained
by this simpler procedure. Therefore, the extent of sacrectomy that

0268-0033/$ see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.clinbiomech.2010.06.012
868 B.-S. Yu et al. / Clinical Biomechanics 25 (2010) 867872
demands for a dual iliac screw xation in lumbo-iliac reconstruction
should be determined.
The sacrum, serving as the foundation of spine, transmits the stress
between spine and pelvis through sacroiliac joints. Depending upon
the tumor nature and location, the surgical procedure may include
complete or partial sacrectomy, which induces various instabilities.
Biomechanical studies have conrmed that the distal partial sacrect-
omy involving S1 may notably damage the biomechanical functions of
sacrum and sacroiliac joints (Gunterberg et al., 1976; Hugate et al.,
2006; Yu et al., 2009). Up till now, previous studies have been focusing
on whether the spino-pelvic reconstruction should be used and which
reconstruction technique could improve the local stability. However,
no previous study has investigated how to select the single and dual
iliac screw techniques for different instability types caused by various
sacrectomies.
Therefore, the purpose of this study was to biomechanically
compare the single and dual iliac screw xation techniques in lumbo-
iliac reconstruction for sequential partial sacrectomies. By making this
comparison, this study aimed to investigate the extent of partial
sacrectomy requiring the application of dual iliac screw technique.
2. Methods
2.1. Specimen preparation
Nine fresh human cadavers (5 male and 4 female) of age ranging
from 25 to 75 years (mean: 49.3 years) were used in this study. The
medical history of each cadaver was inspected and their radiographs
were taken to exclude specimen with metastatic or primary
neoplastic bone lesions or evidence of any other gross abnormalities.
For each specimen, the bone mineral density (BMD) was measured at
L1L4 by CT scan (DEXA, Densiscan 1000, Scanco Medical,
Switzerland) with a slice thickness of 2 mm, and the average value
was used as the specimen's BMD (Yu et al., 2009; Zheng et al., 2009).
L2-pelves were harvested from the cadavers and stored in double-
wrapped plastic bags at 20 C. Before the experiment, the specimens
were thawed for 24 h at room temperature. The residual musculatures
Fig. 1. Photograph and radiograph showing the dual iliac screw technique after total sacrectomy. A: the cranial and caudal sites of the specimen were mounted by polyester resin, by
which the specimen was rmly clamped between the upper and lower grips of the MTS machine; B: two short iliac screws were placed in the lower iliac column.
were carefully removed, leaving the spinal ligaments, osseous and
articular structures intact. Keeping the anterio-superior iliac spines
and pubic tubercles in the vertical plane to simulate standing, the
cranial (L2 and partial L3 vertebrae) and caudal (both sides of
ischiadic tuberosities) ends of the specimens were embedded with
polyester resin to mount the specimen to the mechanical testing
machine (Yu et al., 2009) (Fig. 1A).
2.2. Partial sacrectomies and lumbo-iliac reconstructions
The specimens were rmly clamped to the upper and lower
xtures of a material testing machine (MTS 858 Bionix, MTS System
Inc, Minneapolis, MN) (Fig. 1A). Under the displacement control
mode, ensuring that the specimen is held rmly in place without any
movement, the L3, L4, and L5 pedicle screws (6.5-mm diameter and
45-mm long), and iliac screws (7.5-mm diameter and 70-mm length)
were bilaterally inserted with the use of a posterior spino-pelvic
xation system (Medtronic-WeiGao Inc., Weihai, China). The pedicle
screws were laterally straight and parallel to the vertebral endplate as
described in a previous study (Zheng et al., 2009). A trapezoid-shaped
tricortical iliac bone block, of which the bottom reached the surface of
sacral ala, was cut around the center of the postero-superior iliac
spine (PSIS) (Acharya et al., 2008; Fujibayashi et al., 2007). Two iliac
screws were inserted from the two centers of the superior and inferior
portions of the exposed cancellous bone to the antero-inferior iliac
spine (AIIS) (Fig. 2). In this procedure, after pilot holes were made
using a straight surgical probe under direct vision, a 7.0-mm surgical
tap was used to tap the holes. Then the iliac screws were inserted into
the superior and inferior screw tracts, respectively (Zheng et al.,
2009). Radiography was used to conrm the correct insertion of
screws.
Two rods were pre-bent based on the lumbo-iliac alignment and
L3L5 xation was conducted in situ. Thereafter, two cross-links were
set at the L3L4 and L5-S1 levels respectively to connect the two
longitudinal rods. This state was dened as the lumbo-iliac intact
condition in this study. Following intact testing, L3-iliac xation using
bilateral single iliac screw technique was carried out by further
 B.-S. Yu et al. / Clinical Biomechanics 25 (2010) 867872
Fig. 2. Photograph showing two iliac screws inserted into the lower iliac column. For
the single iliac screw technique, the inferior iliac screw was connected to the rod.
connecting the inferior iliac screws to the rods. With the lumbo-iliac
xation, sequential sacrectomies (Yu et al., 2009) and biomechanical
testing were performed in each specimen as follows (Fig. 3):
(1) Under-S1 sacrectomy + single screw: sacrum amputation
along with the attached ligaments below the inferior border
of the rst sacral foramen was performed using a surgical
T-saw. The cut-line corresponded to the level of S1/2 disc;
(2) Under-S1 sacrectomy + single screw: the cut-line passed
through the superior border of the rst posterior sacral
foramen at the level of 1 cm below sacral promontory. Below
this cut-line, the sacrum bone was osteotomized;
Fig. 3. Sketch showing the patterns of partial sacrectomy.
869
(3) One-side (left) sacroiliac joint resection + single screw: the left
side of residual sacroiliac joint and part of sacral ala from 1 cm
to the articular surface were removed;
(4) Total sacrectomy + single screw: the residual sacrum was
completely cut leaving the bilateral ilio-lumbar ligaments
intact (Fig. 4A);
(5) Total sacrectomy + dual screw: the bilateral superior iliac
screws were additionally connected to the rods (Fig. 4B).
All the instrumentation xation and destabilization procedures
were performed under MTS displacement control mode to prevent
any possible change in position of the specimens. All the screw-rod
junctions were tightened to 60 inch-pounds of torque using a torque
wrench.
2.3. Biomechanical testing
Nondestructive biomechanical testing was rst performed on the
intact condition. After each reconstruction, the testing was repeated
to evaluate the lumbo-iliac stability. The testing protocol included
compression (0 to 800 N) and axial rotation (7 Nm to + 7 Nm
with 100 N compressive preload). These maximum loading levels
chosen are similar to those in previous in vitro studies (Yu et al., 2009;
Zheng et al., 2009). Each testing mode was repeated 5 cycles at a rate
of 0.5 Hz, the data from the fourth cycle of compression and torsion
were used for analysis. The linear, angular range of motion (ROM) of
the xation constructs and the loads were measured by the inbuilt
displacement transducers and load cells of MTS. During the experi-
ments, a saline humidier was used to keep the cadaveric specimens
moist. Each specimen was tested within 4 h after thawing. After the
nal testing, antero-posterior and lateral radiographies were taken for
each specimen.
2.4. Data analysis
Two typical curves, load vs linear displacement and torque vs
angular displacement, were plotted from the data of the 4th cycle. The
maximal compressive (linear) and rotational (angular) displacements
between the peak and trough points were obtained to calculate the
stiffness of xation construct (Yu et al., 2009). The construct stiffness
was furthermore normalized to the corresponding data from the
870 B.-S. Yu et al. / Clinical Biomechanics 25 (2010) 867872
Fig. 4. Diagrams showing the single and dual iliac screw technique after total sacrectomy. A: single iliac screw technique; B: dual iliac screw technique.
intact condition of the same specimen. Statistical analyses were
performed with SPSS software (version 11.5, Chicago, IL). One-way
ANOVA and post hoc LSD test were used for statistical analysis of the
data at a signicance level of P b 0.05.
3. Results
The average BMD value of the 9 specimens was 1.07 (standard
deviation: 0.14) g/cm2, ranging from 0.83 to 1.26 g/cm2. None of the
specimens was osteoporotic. All the specimens successfully went
through the biomechanical testing after reconstruction. No radiolu-
cent halo around the screws was observed in radiographs, and high
torque was still required for removing the iliac screws after each
testing. The mean values and standard deviations of all biomechanical
parameters are summarized in Table 1.
3.1. Compressive stiffness of the L3-iliac xation construct
Signicant differences were detected among the six groups under
compressive loading condition (F = 21.55; P = 0.001). No signicant
differences were demonstrated when the three groups of under-S1
sacrectomy, under-S1 sacrectomy, and one-side sacroiliac joint
resection with the single iliac screw xation were compared with the
intact state (P N 0.05); however, the above four groups showed
signicantly higher compressive stiffness than total sacrectomy +
single screw (P b 0.05). Although total sacrectomy + dual screw exhib-
ited no signicant difference from under-S1 sacrectomy+ single screw,
under-S1 sacrectomy+ single screw, and the intact state, it displayed
markedly higher compressive stiffness than one-side sacroiliac joint
resection + single screw and total sacrectomy+ single screw (P b 0.05)
(Table 1).
Table 1
Construct stiffness (% of intact state): mean (standard deviation).
Compressive
stiffness (%)
Intact condition 100(0)
Under-S1 sacrectomy + single screw 106.4(9.0)
Under-S1 sacrectomy + single screw 102.7(8.0)
One-side sacroiliac joint resection + single screw 92.2(10.1)
Total sacrectomy + single screw 72.7b(8.0)
Total sacrectomy + dual screw 107.7c(10.7)
a
Signicantly different from the under-S1 sacrectomy + single screw, total sacrect-
omy + dual screw, and intact condition.
b
Signicantly different from all the other groups.
c
Signicantly different from the one-side sacroiliac joint resection + single screw.
d
Signicantly different from the under-S1 sacrectomy + single screw.
3.2. Torsional stiffness of the L3-iliac xation construct
Signicant differences on construct torsion stiffness were found
between the six groups (F = 18.2; P = 0.001). No signicant differences
were detected among under-S1 sacrectomy+ single screw, under-S1
sacrectomy + single screw, and the intact condition (P N 0.05), also
between the two groups of one-side sacroiliac joint resection and total
sacrectomy with single iliac screw technique (P N 0.05). However, the
one-side sacroiliac joint resection + single screw and total sacrectomy
+ single screw exhibited signicantly lower torsion stiffness than the
under-S1 sacrectomy+ single screw and the intact condition. Impor-
tantly, total sacrectomy + dual screw offered remarkably higher torsion
stiffness than the under-S1 sacrectmy + single screw, one-side
sacroiliac joint resection + single screw, and total sacrectomy + single
screw (P b 0.05); moreover, its stiffness was comparable with those of
the under-S1 sacrectomy + single screw and intact state (P N 0.05)
(Table 1).
4. Discussion
We evaluated the single and dual iliac screw techniques in
different instability types caused by various sacrectomies and found
that: single iliac screw technique could effectively restore local
stability for under-S1 sacrectomy and it could not provide adequate
stability for further resection of one-side sacroiliac joint in tortion and
total sacrectomy in compression; however, dual iliac screw technique
could restore the stability after total sacrectomy to the intact
condition in both compression and torsion.
Our biomechanical ndings from this study are in agreement with
some of the previous clinical studies. Xiao et al. (2003) performed L3-
iliac screw-rod xation with single iliac screw technique and
posterolateral fusion followed by under-S1 or S1 partial sacrectomy
for 12 patients with the primary distal sacral tumor. They observed a
solid fusion within the early postoperative period in all patients and
no implant failure until 25 months of postoperative follow-up in any
Torsional
of the patients. For 3 patients of total sacrectomy, Doita et al. (2003)
stiffness (%) conducted L3-iliac screw-rod xation using single iliac screw
technique and bone fusion between the iliac wings (with bular
100(0)
105.4(8.8) bone) and the posterolateral sides of the spine (with iliac bone).
89.8(12.3) Unfortunately, nonunion in 1 case and iliac screw loosening in all the
75.9a(10.6) cases were observed in Doita et al.'s study within the rst year of
71.2a(10.2)
postoperative follow-up. In contrast, Acharya et al. (2008) and Dickey
109.1d(16.9)
et al. (2005) executed the L4-iliac fusion and screw-rod instrumen-
tation using the bilateral dual iliac screw technique in sacral tumor
patients with total sacrectomy. In both these studies, good clinical and
functional results were obtained with no failure of the instrumenta-
tion even at 2 years of postoperative follow-up.
 B.-S. Yu et al. / Clinical Biomechanics 25 (2010) 867872
Biomechanical studies by Gunterberg et al. (1976) and Hugate et
al. (2006) conrmed that partial sacrectomy cephalad to 1/2 of S1
notably damaged the vertical loading function of sacrum. Yu et al.
(2009) further conducted a detailed analysis of 5 sequential partial
sacrectomies on sacroiliac joints by applying rotational and compres-
sive loads. They concluded that, the under-S1 partial sacrectomy
procedure could result in rotational instability; the distal resection
beyond S1 lead to vertical instability; and the under-S1 partial
sacrectomy with one-side sacroiliac joint resection produced a more
unstable condition with about 50% loss in the stiffness from the intact
sacroiliac joints. Based on these studies, we have simulated 4
sequential destabilized patterns which need local instrumentation
in this study.
In this study, for the unstable condition of under-S1 partial
sacrectomy with one-side sacroiliac joint resection, the single iliac
screw technique could still maintain 92.2% compressive stiffness of
the intact condition. Furthermore, such instability instrumented with
the single iliac screw offered obviously higher construct compression
stiffness than the total sacrectomy + single screw. These ndings
reveal that partially preserving the S1 body may be benecial to
support the lumbo-iliac implantation. However, in torsional loading,
the one-side sacroiliac joint resection + single screw restored only
75.9% construct stiffness of the intact condition and further exhibited
signicantly lower stiffness than the intact state. The purpose of
reconstruction is to improve stability in all three-dimensions of the
spine to eliminate local pain and improve the quality of life (Jackson
and Gokaslan, 2000). Consequently, single iliac screw xation should
not be applied for the treatment of the instability resulting from high
level sacrectomy combined with one-side sacroiliac joint resection,
because it fails to restore the rotational stability as observed in this
study.
Although anatomically ideal for biomechanical evaluation, human
cadavers provide large specimen-to-specimen differences in BMD,
size, and gender. This study has been carefully designed to minimize
such variation. Firstly, sequential sacral resections and testing were
performed on the same specimen; secondly, all destabilizations and
reconstructions were carried out under the displacement control
mode of MTS machine to maintain the consistency of the experimen-
tal setting; thirdly, the construct stiffness of each reconstruction was
normalized to its intact condition for each specimen in the statistical
analysis. Because of the hysteresis characteristics of viscoelastic
materials, the testing from one step to next step would potentially
lead to an unstable condition. However, this study only performed 5
cycles of compression and torsion testing within the physiological
loading range. Therefore, we believe that the sequential testing, as
validated by previous studies, could hardly inuence the nal results
(Oda et al., 2003; Yu et al., 2009).
This study approximates the displacement and angular ROM
acquired by the inbuilt transducer of the MTS as the ROM of the
xation segments, and the same measurement method has been
applied by several previous biomechanical studies (Yu et al., 2009;
Zheng et al., 2009). In fact, the ROM recorded by the MTS machine
may also include the micro-motion from the iliac bones. To minimize
such motion from the ilia, the cadaveric specimens with normal BMD
value have been used in this biomechanical study. Another advantage
in using specimens with normal BMD value is to prevent loosening of
the instrumentation during the biomechanical testing. As a result,
screw loosening was not observed in any of the 9 specimens after the
nal testing. Although the experimental design is well controlled,
some biomechanical parameters such as fatigue, effects of extrinsic
stabilizers of the neuromuscular system were not evaluated. There-
fore, some limitations still exist when using the data from this study to
guide the clinical practice.
The ultimate goal of the reconstructive surgery is fusion of the
instrumented segments. Without biological fusion, mechanical failure
of any iliac screw technique is predictable. The instrumentation aims
871
to create an environment of mechanical stability in which fusion can
occur. If a lumbo-iliac xation cannot provide adequate initial
stability, the long-term stability through fusion, which might take
many months, would likely not occur. Thus, the evaluation of the
initial stability is perhaps more important than that of the delayed
stability through fusion.
The results of this study suggest that total sacrectomy or under-
S1 partial sacrectomy with one-side sacroiliac joint resection is an
indication to choose the dual iliac screw technique in the lumbo-iliac
reconstruction procedures. To the best of our knowledge, this is the
rst study to investigate how to select the single and dual iliac screw
technique for treating the instabilities generated from various
sacrectomy. However, iliac screw technique is also commonly applied
in long segment xation during the treatment of spinal deformities
and severe lumbo-sacral injuries (Moshirfar et al., 2005; Schildhauer
et al., 2002). Therefore, in those spinal elds, how to choose the single
or dual iliac screw technique needs to be investigated further by
clinical and biomechanical studies.
5. Conclusions
For under-S1 partial sacrectomies, L3-iliac xation using single
iliac screw could effectively restore local stability; however, it could
not provide adequate stability for further resection of one-side sacro-
iliac joint or total sacrectomy. In such situations, the use of dual iliac
screw technique could obtain sufcient construct stability. Therefore,
lumbo-iliac reconstruction with the dual iliac screw technique should
be considered for treating unstable conditions following total
sacrectomy or under-S1 sacrectomy with one-side sacroiliac joint
resection.
Acknowledgement
The authors would like to thank Medtronic-WeiGao Inc. for
providing the posterior spinal xation system. This study was supported
by Guangdong Provincial Medical Research Foundation of China
(A2008161 and 2008B050100012).
References
Acharya, N.K., Bijukachhe, B., Kumar, R.J., Menon, V.K., 2008. Ilio-lumbar xationthe
Amrita technique. J. Spinal Disord. Tech. 21 (7), 493499.
Dickey, I.D., Hugate Jr., R.R., Fuchs, B., Yaszemski, M.J., Sim, F.H., 2005. Reconstruction
after total sacrectomy: early experience with a new surgical technique. Clin.
Orthop. Relat. Res. 438, 4250.
Doita, M., Harada, T., Iguchi, T., Sumi, M., Sha, H., Yoshiya, S., et al., 2003. Total
sacrectomy and reconstruction for sacral tumors. Spine 28 (15), E296E301.
Fujibayashi, S., Neo, M., Nakamura, T., 2007. Palliative dual iliac screw xation for
lumbosacral metastasis. Technical note. J. Neurosurg. Spine 7 (1), 99102.
Gunterberg, B., Romanus, B., Stener, B., 1976. Pelvic strength after major amputation of
the sacrum. An experimental study. Acta Orthop. Scand. 47 (6), 635642.
Hugate Jr., R.R., Dickey, I.D., Phimolsarnti, R., Yaszemski, M.J., Sim, F.H., 2006.
Mechanical effects of partial sacrectomy: when is reconstruction necessary? Clin.
Orthop. Relat. Res. 450, 8288.
Jackson, R.J., Gokaslan, Z.L., 2000. Spinalpelvic xation in patients with lumbosacral
neoplasms. J. Neurosurg. 92 (Suppl. 1), 6170.
Mindea, S.A., Salehi, S.A., Ganju, A., Rosner, M.K., O'Shaughnessy, B.A., Jorge, A., et al.,
2003. Lumbosacropelvic junction reconstruction resulting in early ambulation for
patients with lumbosacral neoplasms or osteomyelitis. Neurosurg. Focus 15 (2), E6.
Moshirfar, A., Rand, F.F., Sponseller, P.D., Parazin, S.J., Khanna, A.J., Kebaish, K.M., et al.,
2005. Pelvic xation in spine surgery. Historical overview, indications, biome-
chanical relevance, and current techniques. J. Bone Joint Surg. Am. 87 (Suppl. 2),
89106.
Oda, I., Abumi, K., Yu, B.S., Sudo, H., Minami, A., 2003. Types of spinal instability that
require interbody support in posterior lumbar reconstruction: an in vitro
biomechanical investigation. Spine 28 (14), 15731580.
Schildhauer, T.A., McCulloch, P., Chapman, J.R., Mann, F.A., 2002. Anatomic and
radiographic considerations for placement of transiliac screws in lumbopelvic
xations. J. Spinal Disord. Tech. 15 (3), 199205.
Schwend, R.M., Sluyters, R., Najdzionek, J., 2003. The pylon concept of pelvic anchorage
for spinal instrumentation in the human cadaver. Spine 28 (6), 542547.
Xiao, J.R., Jia, L.S., Chen, H.J., Wei, H.F., Yang, X.H., Chen, D.Y., et al., 2003. Investigation of
resection and reconstruction procedure of high-sacrum tumors. Zhonghua Wai Ke
Za Zhi 41 (8), 575577 (in Chinese).
872 B.-S. Yu et al. / Clinical Biomechanics 25 (2010) 867872

Yu, B., Zheng, Z., Zhuang, X., Chen, H., Xie, D., Luk, K.D., et al., 2009. Biomechanical effects
of transverse partial sacrectomy on the sacroiliac joints: an in vitro human
cadaveric investigation of the borderline of sacroiliac joint instability. Spine 34
(13), 13701375.
Yu, B.S., Zhuang, X.M., Zheng, Z.M., Li, Z.M., Wang, T.P., Lu, W.W., 2010. Biomechanical
advantages of dual over single iliac screws in lumbo-iliac xation construct. Eur.
Spine J. (Electronic publication ahead of print).
Zhang, H.Y., Thongtrangan, I., Balabhadra, R.S., Murovic, J.A., Kim, D.H., 2003. Surgical
techniques for total sacrectomy and spinopelvic reconstruction. Neurosurg. Focus
15 (2), E5.
Zheng, Z.M., Yu, B.S., Chen, H., Aladin, D.M., Zhang, K.B., Zhang, J.F., et al., 2009. Effect of
iliac screw insertion depth on the stability and strength of lumbo-iliac xation
constructs: an anatomical and biomechanical study. Spine 34 (16), E565E572.