Injury, Int. J.

Care Injured 45 (2014) 510514

Contents lists available at ScienceDirect

Injury
journal homepage: www.elsevier.com/locate/injury

A comprehensive update on current xation options for two-part

proximal humerus fractures
A biomechanical investigation
Richard S. Yoon a, Daniel Dziadosz b, David A. Porter b, Matthew A. Frank b, Wade R. Smith c,
Frank A. Liporace a,*
a
Division of Orthopaedic Trauma, Department of Orthopaedic Surgery, NYU Hospital for Joint Diseases, New York, NY 10003, USA
b
Division of Orthopaedic Trauma, Department of Orthopaedic Surgery, INOVA Fairfax Hospital, Faiefax, VA 22003, USA
c
Division of Orthopaedic Trauma, Department of Orthopaedic Surgery, Swedish Medical Center, Denver, CO 80204, USA

A R T I C L E I N F O A B S T R A C T

Article history: Background: Recent advancements in implant technology offer updated options for surgical manage-
Accepted 29 August 2013 ment that have been rapidly adopted into clinical practice. The objective of this study is to
biomechanically test and compare the current xation options available for surgical xation of two-part
Keywords: proximal humerus fractures and establish load to failure and stiffness values.
Proximal humerus fracture Methods: Sixteen match-paired (32 total) fresh-frozen, cadaveric specimens were randomized to receive
Intramedullary nail 1 of 4 xation constructs following creation of an AO/OTA Type 11A3 (two-part) proximal humerus
Fixed angle plate
fractures. Fixation constructs tested consisted of 3.5 mm xed angle plate (3.5-FAP), 4.5 mm xed angle
Locked plate
Fixed angle blade
plate (4.5-FAP), humeral intramedullary nail (IMN), and a humeral intramedullary nail with a xed angle
blade (IMN-FAB). Specimen bone density was measured to ensure no adequate, non-osteoporotic bone.
Constructs were tested for stiffness and ultimate load to failure and compared via one-way ANOVA
analysis with subsequent post hoc Tukey HSD multiple group comparison statistical analysis.
Results: The IMN-FAB construct was signicantly stiffer than the 3.5-FAP construct (123.8 vs. 23.9,
p < 0.0001), the 4.5-FAP construct (123.8 vs. 33.3, p < 0.0001) and the IMN construct (123.8 vs. 60.1,
p = 0.005). The IMN-FAB construct reported a signicantly higher load to failure than the 3.5-FAB
construct (4667.3 N vs. 1756.9 N, p < 0.0001), and the 4.5-FAP construct (4667.3 N vs. 2829.4 N,
p = 0.019, Table 2). The IMN construct had a signicantly higher load to failure than the 3.5-FAP
construct (3946.8 vs. 1756.9, p = 0.001, Table 2).
Conclusion: Biomechanical testing of modern xation options for two-part proximal humerus fracture
exhibited that the stiffest and highest load to failure construct was the IMN-FAB followed by the IMN,
3.5-FAP and then the 4.5-FAP constructs. However, prospective clinical trials with longer-term follow-up
are required for denitive assessment of the ideal xation construct for surgical management of two-
part proximal humerus fractures.
2013 Elsevier Ltd. All rights reserved.

Introduction population of North America and Western Europe, these fractures

Proximal humerus fractures account for nearly 5% of all
fractures, with increasing incidence during the later decades of
life.1,2 Ideal treatments to maximize outcomes remain controver-
sial; in regards to surgical intervention, despite the evolution of
implants, absolute surgical indications leading reproducible out-
comes are still lacking concensus.1,3,4 However, given the ageing
* Corresponding author at: Department of Orthopaedic Surgery, NYU Hospital for
Joint Diseases, 301 E 17th Street, Suite 1402, New York, NY 10003, USA.
Tel.: +1 201 309 2426; fax: +1 212 598 7654.
E-mail address: liporace33@gmail.com (F.A. Liporace).
are increasing in incidence. Dening optimal treatment
approaches is an important public health care issue as well as a
specic concern for the quality of life of affected patients and
treating surgeons.
Nearly two decades ago, Koval et al. performed a biomechanical
study comparing the ten most common xation methods available
for proximal humerus fractures at that time.5 Classically, this
biomechanical study was a true test of the xation constructs as it
focused on a shear load-to-failure and stiffness comparison in non-
osteoporotic cadaver models. This protocol has rarely been
repeated in more recent studies.612
Since Kovals original article, several new xation methods for
proximal humerus fractures have been introduced.612 With a

00201383/$ see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.injury.2013.08.024
 R.S. Yoon et al. / Injury, Int. J. Care Injured 45 (2014) 510514
primary focus on xed angle devices, locking technology and
intramedullary nailing, many of the more recent biomechanical
studies have tested cyclic loading parameters, torsional resistance
and bending, but few have reported on axial load to failure or
stiffness.79,1326 Furthermore, while many have tested these
constructs in a head-to-head fashion, few studies have looked at
comparing these modern constructs in the same study.9,15
Traditional implant testing in the axial load-to-failure type or
catastrophic failure provides valuable information regarding
implant safety. Specically, concerning the more current xation
methods, incremental load to failure in a 208 abduction angle to
simulate maximal axial and shear load has not been performed.
This is a critical test that closely parallels the physiologic loads
placed upon implants during routine post operative rehabilitation.
The objective of our study was to apply the standardized
techniques as described by Koval to current xation techniques.
Our study goal was to compare xed angle plates and intrame-
dullary nail constructs to determine which construct has the
highest stiffness and ultimate load to failure when placed in the
maximum shearing load position, simulating a catastrophic load to
failure scenario in an angle seen during early active range of
motion.5 These ndings can provide the biomechanical equipoise
to initiate randomized clinical studies to determine best practice
for these common yet vexing injuries.
Materials and methods
In a biomechanical study, one of four methods of fracture
xation: 3.5 mm locked xed angle plates (3.5-FAP), 4.5 mm
locked xed angle plates (4.5-FAP), proximal humerus intrame-
dullary nails with locked and screws (IMN), and intramedullary
nails with xed angle blades (IMN-FAB), was applied to 16
matched pairs of cadaveric proximal humerii. Each specimen was
then incrementally loaded to failure at a rate of 10 cm/min. Failure
was dened as a marked decrease or discontinuity in the load
displacement curve.
Specimens
Sixteen matched paired (32 total) cadaveric proximal humerii
were prepared for use with all soft tissue stripped from the
specimens. Once prepared, a simulated fracture was made at the
Fig. 1. (AD) Radiographic images of (A) 3.5 mm locking plate construct, (B) 4.5 mm locking plate construct, (C) intramedullary nail with locking screws and (D)
intramedullary nail with a xed angle blade.
511
surgical neck of the proximal humerus using a reciprocating saw.
The fracture line was made at a 10-degree oblique angle directed in
a medial and inferior direction based on the model proposed by
Koval.5 In addition, a 1 cm medially based wedge of cortex at the
fracture site was removed to simulate a comminuted fracture
pattern as per Chudnik, ultimately creating an unstable, AO/OTA
Type 11A3 proximal humerus fracture.7 Dual-energy X-ray
absorption scans of the proximal humerus were obtained to
quantify bone quality in terms of T-scores. Additionally, all
specimens were radiographed to identify any lesions that might
affect osseous integrity, of which there were none.
Fixation constructs
All specimens were xed according to industry/implant
protocols. Four different constructs were created with the
aforementioned implants for a total of eight specimens per
implant. The matched pairs were randomized using a Monte Carlo
ordered randomization method with an equal side-to-side
distribution among all implants.27 Once nal instrumentation
and osteotomies were created, all specimens orthogonal radio-
graphic evaluation to ensure appropriate hardware position
(Fig. 1AD).
Test setup
The distal one-third of each specimen was removed and then
each construct was mounted in a polymethyl-methacrylate
(PMMA) at a 20-degree abduction angle. This angle simulates
maximal axial and shear load to the humerus at a motion
analogous to the load seen in early active abduction.28,29 These
were then mounted in an Instron materials testing simulator
(Instron Inc., Norwood, Massachusetts, USA, Fig. 2). The load was
focused one cm lateral to the articular margin, as per the previously
dened protocol.7 A load rate of 10 cm/min was then applied to the
construct until failure (dened by a marked decrease or
discontinuity in the load displacement curve).5,7 The mode of
each constructs failure was determined by inspection. Average
values for displacement and maximum load to failure were
measured during testing and average stiffness for each of the four
constructs was calculated. One-way analysis of variance (ANOVA)
tests were performed comparing both stiffness and ultimate load
512 R.S. Yoon et al. / Injury, Int. J. Care Injured 45 (2014) 510514

Fig. 3. Mean stiffness ranges for the four tested biomechanical constructs. One-way
ANOVA indicated statistically signicant differences across all four groups
(p < 0.0001).

Similarly, in descending order, the construct that required the

Fig. 2. Mounted specimen in the MTS machine, at a 208 abduction angle to maximize
shear axial load forces.
to failure between all four groups; Tukey HSD post hoc analysis for
multiple mean comparisons across all four groups was also
performed. Signicant was set at p < 0.05.
Results
Bone quality did not differ signicantly among the specimens
(T-score = 1.3  1.8) and none of the T-scores were greater than one
standard deviation of the mean; therefore none of the specimens
were osteoporotic, but osteopenic, mimicking a hybrid between the
representative bimodal distribution of patients suffering from such
an injury that undergoes surgical xation. Achieving this goal of bone
density allowed to test what surgeons can truly control, implant
selection.
In descending order, the stiffest construct was the IMN-FAB
group (123.8  50.0 N/mm) followed by the IMN group
(60.1  51.0 N/mm), the 4.5-FAP group (33.3  23.1 N/mm), and
the 3.5-FAP group (23.9  15.1); one-way ANOVA analysis yielded
statistically signicant differences across the four groups (Table 1 and
Fig. 3, p < 0.0001).
most force to fail was the IMN-FAB group (4667.3  1157.3 N)
followed by the IMN group (3946.8  1539.9 N), the 4.5-FAP group
(2829.4  1452.3 N), and the 3.5-FAP group (1756.9  609.0 N); one-
way ANOVA analysis also noted statistically signicant differences
amongst the four group means (Table 1 and Fig. 4, p < 0.0001).
More specic construct comparisons, statistically analyzed via
Tukey HSD post hoc analysis following the initial one-way ANOVA,
reported several signicant differences between the constructs in
both stiffness and load to failure comparisons (Table 2). The IMN-
FAB construct was signicantly stiffer than the 3.5-FAP construct
(123.8 vs. 23.9, p < 0.0001), the 4.5-FAP construct (123.8 vs. 33.3,
p < 0.0001) and the IMN construct (123.8 vs. 60.1, p = 0.005,
Table 2). No signicant differences were found when comparing
mean stiffness between the 3.5-FAP and 4.5 FAP constructs (23.9
vs. 33.3, p = 0.95), the 3.5-FAP and IMN constructs (23.9 vs. 60.1,
p = 0.15) or the 4.5-FAP and IMN constructs (33.3 vs. 60.1, p = 0.46,
Table 2).
Head-to-head load to failure comparisons also yielded signi-
cant differences between the constructs (Table 2). The IMN-FAB
construct reported a signicantly higher load to failure than the
3.5-FAB construct (4667.3 N vs. 1756.9 N, p < 0.0001), and the 4.5-
FAP construct (4667.3 N vs. 2829.4 N, p = 0.019, Table 2). The IMN
construct had a signicantly higher load to failure than the 3.5-FAP
construct (3946.8 vs. 1756.9, p = 0.001, Table 2). No signicant
differences between ultimate load to failure was found between

Table 1
Mean stiffness and load to failure measurements with 95% condence interval
ranges.

Mean (SD) 95% CI range (lowerupper)

Stiffness*
3.5-FAP 23.9 (15.1) 13.834.1
4.5-FAP 33.3 (23.1) 13.952.6
IMN 60.1 (51.0) 23.696.6
IMN-FAB 123.8 (50.0) 85.4162.2

Load to failure*
3.5-FAP 1756.9 (609.0) 1347.72166.0
4.5-FAP 2829.4 (1452.3) 1615.24043.5
IMN 3946.8 (1539.9) 2845.35048.4
IMN-FAB 4667.3 (1157.3) 3777.85556.9
* Fig. 4. Mean load to failure measurements for the four biomechanical constructs.
One-way ANOVA analysis yielded statistically signicant differences across all
four groups, p < 0.0001. Please see Table 2 for more detailed group comparisons via One-way ANOVA analysis yielded statistically signicant difference across all four
Tukey HSD post hoc analysis. groups (p < 0.0001).
 R.S. Yoon et al. / Injury, Int. J. Care Injured 45 (2014) 510514 513

Table 2
Head-to-head stiffness and load to failure comparisons between constructs. Statistical analysis via Tukey HSD was performed post hoc, after establishing signicance via one-
way ANOVA.

Construct 1 Mean (N/mm) Construct 2 Mean (N/mm) Absolute mean difference p Value

Stiffness comparisons
3.5-FAP 23.9 4.5-FAP 33.3 9.4 0.95
3.5-FAP 23.9 IMN 60.1 36.2 0.15
3.5-FAP 23.9 IMN-FAB 123.8 99.9 <0.0001
4.5-FAP 33.3 IMN 60.1 26.8 0.46
4.5-FAP 33.3 IMN-FAB 123.8 90.5 <0.0001
IMN 60.1 IMN-FAB 123.8 63.7 0.005

Load to failure comparisons

3.5-FAP 1756.9 4.5-FAP 2829.4 1072.5 0.25
3.5-FAP 1756.9 IMN 3946.8 2190 0.001
3.5-FAP 1756.9 IMN-FAB 4667.3 2910.5 <0.0001
4.5-FAP 2829.4 IMN 3946.8 1117.5 0.234
4.5-FAP 2829.4 IMN-FAB 4667.3 1838 0.019
IMN 3946.8 IMN-FAB 4667.3 720.5 0.58

the 3.5-FAP and 4.5-FAP constructs (1756.9 vs. 2829.4, p = 0.25),
the 4.5-FAP and IMN groups (2829.4 vs. 1117.5, p = 0.234), or the
IMN and IMN-FAB constructs (3946.8 vs. 4667.3, p = 0.58, Table 2).
Observation of the construct failures revealed that each failed in
a similar fashion. The intramedullary devices failed with the
proximal head xation remaining in place and failure was noted in
the shaft, with all ten specimens sustaining shaft fractures.
Similarly, the plate failures occurred with shaft cut out and not
head cut out.
Discussion
Treatment of unstable proximal humerus fractures can pose a
challenge to the treating physician. With an inability to maintain a
satisfactory reduction by closed means, operative intervention
provides the best option to maintain an adequate reduction.2 The
proximal portion of the humerus undergoes multiple forces
through any given range of motion and xation of these fractures
needs to be able to withstand these forces. The implants tested can
all withstand the rigours of motion and can maintain anatomic
length and alignment, but in this study it was determined that the
proximal humeral nail with xed angle blade is biomechanically
more stable to varus collapse than the other implants tested
exhibiting the highest stiffness and axial load to failure in a
position of maximum shear.
All implants have advantages and disadvantages. Fixation with
an intramedullary device prevents the soft tissue stripping that is
required for xation with a proximal humeral plate. Additonally,
precise plate/screw placement is required to achieve and maintain
reduction, which can increase operative time and blood loss.30
However, intramedullary nail insertion does require dissection
through the rotator cuff tendons and has been noted to have an
increased incidence of shoulder pain.8 Both techniques are useful
with documented positive outcomes. When failure does occur, the
mode is most often varus collapse.31 Therefore, determining which
implants have the highest resistance to varus collapse is an
important factor in implant selection.
Implant stiffness however is not always the most valued
parameter in regards to implant xation. Observational documen-
tation over the years has shown that excessively stiff implants can
decrease micromotion, subsequent callus formation and lead to
delayed or non-unions in many fracture types.32,33 The optimal
degree or parameter of stiffness for fracture healing is not know.
We do know that proximal humerus fractures have short,
osteoporotic metaphyseal fracture segments, which permit a
limited amount of xation that must be stiff enough to permit bone
healing while preventing failure. With this reason, we chose to test
the implants within osteopenic bone, achieving a hybrid between
the bimodal distribution of patients representative of receiving
surgical xation. Given the anatomy of these fractures and the
common failure modes, having implants that are too stiff is an
unlikely and possibly unachievable situation. Therefore, stiffer
implants that can be placed with minimal intrusion and damage to
the local blood supply are likely to have lower failure rates.
Previous biomechanical studies have focused primarily on
cyclic loading, bending, torsion and other testing modalities, but
have left out the baseline stiffness and axial load data reported in
our study.79,15,17,18,2022,25,34,35 Edwards et al. in an effort to
compare a proximal humeral nail to a 3.5 mm locking plate
construct similarly found biomechanical superiority with the nail.
However, their study was performed with a cyclical loading
construct and did not report on pure axial load to failure.8 We did
test to axial load to failure as this is a critical component of failure
in the in vivo patient given the deforming forces of the rotator cuff
and deltoid muscles. Similarly, other authors have tested various
nail and plate constructs, but did not report baseline axial load and
stiffness values despite measuring more complex, multiplanar
shoulder forces.9,17,18 We feel, given the practical and predictable
nature of failure in these patients, that stiffness and axial load to
failure are critical characteristics that need to be described and
compared between implants, particularly given the relatively high
clinical failure rates reported in the literature.31,36
The primary limitation of our study is that there is no
standardized testing protocol for the proximal humerus. Multiple
models have been proposed to mimic the complex range of motion
of the shoulder joint. Consequently, basic biomechanical tests are
often not performed. Without a consensus about how best to
examine in vitro, the forces affecting a shoulder injury researchers
must make basic assumptions about what testing studies are most
relevant to a particular injury to the shoulder girdle. Given the
most common modes of failure following xation of proximal
humerus fractures, we chose biomechanically basic parameters:
axial load to failure and stiffness. More complex motions such as
torsion, shear and circumduction may be part of the stresses on
these fractures and therefore our model of testing may not
accurately mimic all the forces affecting outcome in the in vivo
situation. We would point out however, that few patients in the
true clinical situation move their shoulder in a normal manner and
that failure usually occurs relatively early before complex
movement has been initiated. Our study results therefore provide
important baseline information on modern implants that is
previously lacking in the literature. The clinical relevance of our
ndings is that despite the current trend towards anatomically
specic locking plates for proximal humerus fracture xation, xed
514 R.S. Yoon et al. / Injury, Int. J. Care Injured 45 (2014) 510514
angle intramedullary nails may be a better alternative. Further-
more, an important note is allowing for an increased working
length for the xed angle plates may have had a marginal increase
in load to failure, albeit a decrease in stiffness. This is a parameter
that should have been studied in depth.
In conclusion, intramedullary nail devices provided more
stability and stiffness, compared to locking plates in an axial load
with shear model. Given the relatively high failure rate of these
difcult fractures, despite anatomically designed locking plate
constructs, understanding the failure modes of cutout is valuable
information for treating surgeons. Reproducing the landmark
study of Koval et al. with modern implants has lled a necessary
gap in the current literature. The ndings of our study, coupled
with previous reported clinical series provide the equipoise for a
randomized, prospective comparison of plate vs. nail xation of
these vexing fractures.
Conict of interest
The authors have no direct conict of interest to report.
References
1. Jensen GF, Christiansen C, Boesen J, Hegedus V, Transbol I. Relationship between
bone mineral content and frequency of postmenopausal fractures. Acta Med
Scand 1983;213(1):613.
2. Leyshon RL. Closed treatment of fractures of the proximal humerus. Acta Orthop
Scand 1984;55(February (1)):4851.
3. Stromsoe K. Fracture xation problems in osteoporosis. Injury 2004;35(Febru-
ary (2)):10713.
4. Korsic M, Grazio S. Non-hip peripheral osteoporotic fractures: epidemiology
and signicance. Arh Hig Rada Toksikol 2008;59(March (1)):538.
5. Koval KJ, Blair B, Takei R, Kummer FJ, Zuckerman JD. Surgical neck fractures of
the proximal humerus: a laboratory evaluation of ten xation techniques. J
Trauma 1996;40(May (5)):77883.
6. Brunner A, Weller K, Thormann S, Jockel JA, Babst R. Closed reduction and
minimally invasive percutaneous xation of proximal humerus fractures using
the Humerusblock. J Orthop Trauma 2010;24(July (7)):40713.
7. Chudik SC, Weinhold P, Dahners LE. Fixed-angle plate xation in simulated
fractures of the proximal humerus: a biomechanical study of a new device. J
Shoulder Elbow Surg 2003;12(NovemberDecember (6)):57888.
8. Edwards SL, Wilson NA, Zhang LQ, Flores S, Merk BR. Two-part surgical neck
fractures of the proximal part of the humerus. A biomechanical evaluation of
two xation techniques. J Bone Joint Surg Am 2006;88(October (10)):225864.
9. Hessmann MH, Hansen WS, Krummenauer F, Pol TF, Rommens PM. Locked plate
xation and intramedullary nailing for proximal humerus fractures: a bio-
mechanical evaluation. J Trauma 2005;58(June (6)):1194201.
10. Matziolis D, Kaeaeb M, Zandi SS, Perka C, Greiner S. Surgical treatment of two-
part fractures of the proximal humerus: comparison of xed-angle plate
osteosynthesis and Zifko nails. Injury 2010;41(10):10416.
11. Williams Jr GR, Wong KL. Two-part and three-part fractures: open reduction
and internal xation versus closed reduction and percutaneous pinning. Orthop
Clin North Am 2000;31(January (1)):121.
12. Zhu Y, Lu Y, Wang M, Jiang C. Treatment of proximal humeral fracture with a
proximal humeral nail. J Shoulder Elbow Surg 2010;19(March (2)):297302.
13. Bae JH, Oh JK, Chon CS, Oh CW, Hwang JH, Yoon YC. The biomechanical
performance of locking plate xation with intramedullary bular strut graft
augmentation in the treatment of unstable fractures of the proximal humerus. J
Bone Joint Surg Br 2011;93(July (7)):93741.
14. Brianza S, Plecko M, Gueorguiev B, Windolf M, Schwieger K. Biomechanical
evaluation of a new xation technique for internal xation of three-part
proximal humerus fractures in a novel cadaveric model. Clin Biomech (Bristol
Avon) 2010;25(9):88692.
15. Fuchtmeier B, May R, Hente R, et al. Proximal humerus fractures: a comparative
biomechanical analysis of intra and extramedullary implants. Arch Orthop
Trauma Surg 2007;127(6):4417.
16. Gillespie RJ, Ramachandran V, Lea ES, Vallier HA. Biomechanical evaluation of 3-
part proximal humerus fractures: a cadaveric study. Orthopedics 2009;32(11):
816.
17. Horn J, Gueorguiev B, Brianza S, Steen H, Schwieger K. Biomechanical evaluation
of two-part surgical neck fractures of the humerus xed by an angular stable
locked intramedullary nail. J Orthop Trauma 2011;25(7):40613.
18. Huff LR, Taylor PA, Jani J, Owen JR, Wayne JS, Boardman ND. 3rd. Proximal
humeral fracture xation: a biomechanical comparison of two constructs. J
Shoulder Elbow Surg 2013;22(Jan (1)):12936.
19. Kitson J, Booth G, Day R. A biomechanical comparison of locking plate and
locking nail implants used for fractures of the proximal humerus. J Shoulder
Elbow Surg 2007;16(3):3626.
20. Mathison C, Chaudhary R, Beaupre L, Reynolds M, Adeeb S, Bouliane M.
Biomechanical analysis of proximal humeral xation using locking plate xa-
tion with an intramedullary bular allograft. Clin Biomech (Bristol Avon)
2010;25(7):6426.
21. Roderer G, Gebhard F, Krischak G, Wilke HJ, Claes L. Biomechanical in vitro
assessment of xed angle plating using a new concept of locking for the
treatment of osteoporotic proximal humerus fractures. Int Orthop 2011;35(April
(4)):53541.
22. Rothstock S, Plecko M, Kloub M, Schiuma D, Windolf M, Gueorguiev B. Bio-
mechanical evaluation of two intramedullary nailing techniques with different
locking options in a three-part fracture proximal humerus model. Clin Biomech
(Bristol Avon) 2012;27(7):68691.
23. Siffri PC, Peindl RD, Coley ER, Norton J, Connor PM, Kellam JF. Biomechanical
analysis of blade plate versus locking plate xation for a proximal humerus
fracture: comparison using cadaveric and synthetic humeri. J Orthop Trauma
2006;20(September (8)):54754.
24. Wallace MJ, Bledsoe G, Moed BR, Israel HA, Kaar SG. Relationship of cortical
thickness of the proximal humerus and pullout strength of a locked plate and
screw construct. J Orthop Trauma 2012;26(April (4)):2225.
25. Yamamoto N, Hongo M, Berglund LJ, et al. Biomechanical analysis of a novel
locking plate with smooth pegs versus a conventional locking plate with
threaded screws for proximal humerus fractures. J Shoulder Elbow Surg
2013;22(April(4)):44560.
26. Zettl R, Muller T, Topp T, et al. Monoaxial versus polyaxial locking systems: a
biomechanical analysis of different locking systems for the xation of proximal
humeral fractures. Int Orthop 2011;35(8):124550.
27. Chen LH, Lee WC. Two-way minimization: a novel treatment allocation method
for small trials. PLoS One 2011;6(12):e42860.
28. Inman VT, Saunders M, Abbot LC. Observations on the function of the shoulder
joint. J Bone Joint Surg Am 1944;26:130.
29. Poppen NK, Walker PS. Forces at the glenohumeral joint in abduction. Clin
Orthop 1978;135:16570.
30. Gardner MJ, Weil Y, Barker JU, Kelly BT, Helfet DL, Lorich DG. The importance of
medial support in locked plating of proximal humerus fractures. J Orthop
Trauma 2007;21(March (3)):18591.
31. Agudelo J, Schurmann M, Stahel P, et al. Analysis of efcacy and failure in
proximal humerus fractures treated with locking plates. J Orthop Trauma
2007;21(NovemberDecember (10)):67681.
32. Bottlang M, Doornink J, Byrd GD, Fitzpatrick DC, Madey SM. A nonlocking end
screw can decrease fracture risk caused by locked plating in the osteoporotic
diaphysis. J Bone Joint Surg Am 2009;91(March (3)):6207.
33. Perren SM. Evolution of the internal xation of long bone fractures. The
scientic basis of biological internal xation: choosing a new balance between
stability and biology. J Bone Joint Surg Br 2002;84(November (8)):1093110.
34. Schumer RA, Muckley KL, Markert RJ, et al. Biomechanical comparison of a
proximal humeral locking plate using two methods of head xation. J Shoulder
Elbow Surg 2010;19(June (4)):495501.
35. Walsh S, Reindl R, Harvey E, Berry G, Beckman L, Steffen T. Biomechanical
comparison of a unique locking plate versus a standard plate for internal
xation of proximal humerus fractures in a cadaveric model. Clin Biomech
(Bristol Avon) 2006;21(December (10)):102731.
36. Erhardt JB, Stoffel K, Kampshoff J, Badur N, Yates P, Kuster MS. The position and
number of screws inuence screw perforation of the humeral head in modern
locking plates: a cadaver study. J Orthop Trauma 2012;26(October (10)):
e18892.