- Email: [email protected]
Improving socket design to prevent difficult removal of locking screws
Accepted Manuscript Title: Improving socket design to prevent difficult removal of locking screws Authors: Chen-Huei Lin, Ching-Kong Chao, Yi-Hsuan Tang, Jinn Lin PII: DOI: Reference:
S0020-1383(18)30056-1 https://doi.org/10.1016/j.injury.2018.01.036 JINJ 7577
To appear in:
Injury, Int. J. Care Injured
Accepted date:
31-1-2018
Please cite this article as: Lin Chen-Huei, Chao Ching-Kong, Tang Yi-Hsuan, Lin Jinn.Improving socket design to prevent difficult removal of locking screws.Injury https://doi.org/10.1016/j.injury.2018.01.036 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Improving socket design to prevent difficult removal of locking screws
SC RI PT
Chen-Huei Lin, MDa, Ching-Kong Chao, PhDa, Yi-Hsuan Tang, MDa,Jinn Lin, PhDb,* a
Department of Mechanical Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan b Department of Orthopedics, National Taiwan University Hospital, Taipei, Taiwan
A
CC
EP
TE
D
M
A
N
U
* Corresponding author. E-mail address: [email protected] (J. Lin).
Abstract Introduction: Reports of driver slippage leading to difficult locking screw removals have increased since the adoption of titanium for screw fabrication; the use of
SC RI PT
titanium is known to cause cross-threading and cold welding. Such problems occur most frequently in screws with hex sockets, and may cause serious surgical
complications. This study aimed to improve screw socket design to prevent slippage and difficult screw removal.
Materials and methods: Three types of small sockets (hex, Torx, and cruciate) and
U
six types of large sockets (hex, Torx, Octatorx, Torx+ I, Torx+ II, and Torx+ III) with screw head diameters of 5.5 mm were manufactured from titanium, and
N
corresponding screwdrivers were manufactured from stainless steel. The screw heads
A
and drivers were mounted on a material testing machine, and torsional tests were
M
conducted to simulate screw usage in clinical settings at two insertion depths: 1 and 2
D
mm. Ten specimens were tested from each design, and the maximum torque and
TE
failure patterns were recorded and compared. Results: For small sockets in 2 mm conditions, the hex with the largest driver core
EP
had the highest torque, followed by Torx and cruciate. In these tests, the drivers were twisted off in all specimens. However, under the 1 mm condition, the hex slipped and
CC
the torque decreased markedly. Overall, torque was higher for large sockets than for
A
small sockets. The Octatorx, with a large core and simultaneous deformation of the driver and socket lobes, had the highest torque at almost twice that of the small hex. The hex had the lowest torque, a result of slippage in both the 1 and 2 mm conditions. Torx plus designs, with more designed degrees of freedom, were able to maintain a higher driving angle and larger core for higher torque. Conclusions: The hex design showed slipping tendencies with a marked decrease in
torque, especially under conditions with inadequate driver engagement. Large sockets allowed for substantial increases in torque. The Torx, Octatorx, and Torx plus designs displayed better performance than the hexes. Improvements to the socket design could
SC RI PT
effectively prevent slippage and solve difficult screw removal problems.
A
CC
EP
TE
D
M
A
N
U
Keywords: Locking plate; Titanium; Screw head socket; Torsion test; Maximal torque.
Introduction The use of locking plates can provide high fixation stability for reliable healing in various types of bodily fractures, aiding in positive functional recovery.1-5 Minimally invasive surgical techniques can further decrease tissue damage and hasten the
SC RI PT
postoperative recovery process.5-7 These plates are commonly manufactured from stainless steel and titanium alloys, the latter of which has been particularly advocated for recently because of its advantages: higher biocompatibility, fewer resultant artifacts
on computer tomographic scans and magnetic resonance images, higher fatigue strength, lower infection risk, and isoelasticity to bone as compared with stainless steel.3,6,8,9
U
However, as a malleable metal, titanium is prone to cross-threading or cold welding
N
under strong metal-to-metal contact; therefore, its use more easily results in socket
A
damage.6,10-12 Reports of difficulty removing jammed titanium screws have become
M
more frequent.13,14 Difficult removals in a clinical setting may result in serious complications such as prolonged operations, bleeding, infection, neurovascular injuries,
D
and even iatrogenic fractures.3,13,14 Typically, difficulties in removal begin when the
TE
screw head sockets are stripped, rendering drivers useless. 10,15,16 Multiple removal methods have been invented, including foil insertion, screw head drilling, and reverse
EP
thread extraction, but none yet guarantee a successful removal.3,11,16-18 Screw jamming
CC
prevention has also been emphasized by methods such as the use of an aiming sleeve, proper driver engagement, use of torque-limiting screwdrivers, and using screwdriver
A
tips in good condition.2,3,8,19,20 Mechanically, improving screw head socket designs to prevent stripping and to increase torque to facilitate unscrewing may prove more important than these preventative techniques; however, few studies have focused on the mechanical behaviors of different socket designs, especially for titanium screws.21-23 In the present study, different screw head socket designs were specially
manufactured from titanium, including the most conventional hexagonal sockets. The corresponding screwdrivers were made from stainless steel. The screw heads and drivers were then mounted on a material testing machine, and torsional tests simulating clinical applications were conducted. For results, the maximum torque values and
SC RI PT
failure patterns were recorded. The testing results from each of the different socket
designs were compared with one another, and factors that could increase the torque and prevent stripping were analyzed.
Materials and Methods
U
Structures of screw heads and drivers
N
The screw heads and drivers were specially manufactured from a titanium alloy
A
(Ti6Al4V, F136-13) and stainless steel (AISI 420, F138) (Carpenter Technology,
M
Reading, PA, USA), respectively. The yield strength was 627 MPa for the stainless steel and 798 MPa for the titanium. Two types of socket (recess) were used: small and
D
large (Fig. 1). Three design types were used for the small sockets: hex, Torx, and
TE
cruciate (Table 1). The small hex is currently the most commonly used design for 3.5 mm screws (ISO 4762, M3). To ensure a fair comparison, the areas of the cruciate and
EP
Torx sockets were kept equal to that of the hex. Six large socket designs were also
CC
tested: hex, Torx, Octatorx, Torx+ I, Torx+ II, and Torx+ III. Each screw head had an outer diameter of 5.5 mm, a height of 3 mm, and a conical angle of 19° (Fig. 2). As
A
seen in Fig. 1, Torx+ III had wider and deeper sockets than Torx+ II. The large sockets were the modified designs of the small ones to improve the performance. All sockets had a depth of 2 mm and completely straight sidewalls to prevent cam-out. A 0.02 mm manufacturing tolerance was set for both sockets and drivers. The screw heads were attached to a solid titanium cylinder that could be firmly gripped by the fixture,
thereby avoiding any confounding effects the screw shafts may have caused. Corresponding drivers, which were also attached to a solid stainless steel cylinder for firm grip by the fixture, were fabricated with a length of 3 mm.
SC RI PT
Biomechanical test and Failure analysis
Single load torsional tests using a servohydraulic material testing machine (8874,
Instron Industrial Products, Grove City, PA, USA) were conducted to evaluate the
mechanical performance of the screws and drivers (Fig. 3). The screw head and driver cylinders were first mounted on the machine co-linearly, and checked repeatedly to
U
ensure exact alignment between the rotational centers of the screw heads and drivers.
N
The drivers were inserted into sockets at two depths: 1 and 2 mm. The 2 mm insertion
A
depth represented an appropriate screwing technique, whereas the 1 mm depth
M
represented inappropriate insertion. Ten pairs of new screws and drivers were tested for each socket design. During the tests, the drivers were rotated approximately 240°
D
in the clockwise direction until the torque dropped to approximately zero with a
TE
loading rate of 15 °/min. The torque and angle were recorded continuously throughout
EP
this procedure, and the highest recorded torque was defined as the maximum torque. Once testing was complete, the failure patterns were observed and correlated to
CC
the maximum torque and load-deformation curve.
A
Statistical analysis One-way ANOVA tests (SPSS v22.0, SPSS Inc., Chicago, IL, USA) were used during biomechanical testing to compare the torsional strengths of small and large sockets. Bonferroni tests were used for post-hoc pairwise comparisons. The significance level was defined as p <0.05.
Results Small sockets Under the 2 mm condition, the hex design had highest torque, followed by Torx, and
SC RI PT
then cruciate (Table 2). All socket designs demonstrated driver warping and failure by
the completion of testing. All drivers were twisted off at the top of the sockets (Fig. 4). Because each screw experienced very small standard deviations (4.9%, 1.42%, and
5.15% for hex, Torx, and cruciate, respectively), the differences in each comparison were statistically significant. Conversely, under the 1 mm condition the hex
U
experienced the highest torque decrease in all cases (23%) because the drivers slipped
N
in the sockets. The decrease was lower in the cruciate (3.5%) and Torx (-0.04%),
M
A
while the failure mode was the same as that observed in 2 mm conditions.
Large sockets
D
There were two types of load deformation curves: those with one peak and those with
TE
multiple peaks. Single peak curves reflected conditions without driver slippage.
EP
Multiple peaks with progressively decreasing peak torques reflected driver slippage in the sockets (Fig. 5). The maximum torque observed in the large sockets was
CC
substantially higher than that observed in the small sockets. For the hex designs, all sockets displayed both slipping and stripping under 1 and 2 mm conditions, with an
A
obvious torque decrease of 38% under the 1 mm condition. Octatorx allowed the highest torque for the 2 mm condition, followed by Torx and Torx+ I (Fig. 5) (Table 2). All the drivers were eventually twisted off in these three designs (Fig. 6). There was also a marked deformation of the driver lobes in the Octatorx design, which remained tightly incarcerated in the socket lobes. Driver lobes remained relatively
well preserved in the Torx and Torx + I designs. Under 1 mm conditions, these three designs displayed similar failure patterns with breakage of the driver lobes. The torque decrease was lower in Torx and Torx+ I (14.8% and 7.6%, respectively) with small driver cores, but higher in Octatorx (24.4%) with larger cores. Under the 2 mm
SC RI PT
condition, the socket lobes were reamed off in the Torx+ II design, although the
drivers remained relatively intact. In 1 mm conditions, the failure pattern was the
same as that in 2 mm conditions and the torque was only approximately half (54.2%). For Torx+ III, the torque was relatively low in the 2 mm condition with screw head
failure occurring because of the thin walls of the screw head (0.19 mm). The standard
U
deviations were also small in the large socket screws, ranging from 1% to 11.3%, and
A
M
designs except hex and Torx+ III.
N
the differences between each pair reached a level of statistical significance for all
Discussion
D
Many different screw head socket types have been designed for orthopaedic screws. The
TE
earliest single slot and cruciate sockets proved to be disadvantaged by driver slippage
EP
and low torque transfer.6,20 Later hexagonal and square sockets with vertical side walls were able to prevent cam-out and allow for higher torque transfer, and their wide
CC
contact points also ensured a tight grip between the driver and screw head for both reliable screw transfer between the operating staffs and effective screw turning with a
A
small end load. The hex socket, possessing smaller wrench-swing arcs, is better than a square design in situations with surrounding obstacles. However, hex screws frequently experience slipping, which may easily strip the sockets. 3,9,21 The Torx design, developed in 1967 with a six-point star-shaped configuration, allows for higher torque transfer than a similarly sized hex socket.4,24 Its large driving angles prevent damage to the screw
head and tool and increase torque transfer. Driving angles represent the angles between the screwing direction and the side of the lobes (Fig. 2). Low driving angles may cause higher radial force, which is perpendicular to the turning force and not effective in screwing, whereas high driving angles can generate higher screwing torque and reduce
SC RI PT
the damaging radial force.23,24 When the driving angle is too low, it may not only burst the socket and crush the driver, and excessive radial force will also round off the corners
of both socket and driver, causing slippage. High driving angles allow the Torx design to have a smaller head under the same required torque. This can be an advantage when space is limited, such as in locking plates where a smaller screw head allows for smaller
U
screw holes and thus increases the mechanical strength of the plates. However, the
N
disadvantage of Torx sockets is their smaller internal "splines;" these can wear easily,
A
resulting in slippage and shorted tool life. A Torx successor was introduced around 1990
M
when the original Torx patent was expiring.24 This design, the Torx Plus, features more
transfer and lower wear.
D
square lobes for 90° driving angles, allowing for even greater increases in torque
TE
As locking plates are increasingly made from titanium, with its cold welding and bone integration properties,8,15,25 implant removal has become a challenging task,
EP
especially for screws with hex sockets.10,23 Although the cold welding between the
CC
locking plates and screws has never been demonstrated micro- or macroscopically in the literature and its reality has been questioned,6,8,11,14 different degrees of cold
A
welding have been observed by the present authors (Fig. 7). High ductility and flowing material is considered a general prerequisite for excessive cold welding.26 It has been reported that some factors contributing to increased risk of difficult removals include screw size (3.5 mm in particular, though larger screws are not exempt), screws near the elbow joints, screws at the diaphysis, delayed removal time, good
bone quality, screw length, and young age.6,11,12,17 Recently, Torx sockets were introduced as a “cutting-edge” advancement in locking plates, aiming to increase torque and resist stripping.4 However, there have been no mechanical studies comparing the performance of this screw socket to other designs, especially those
SC RI PT
fabricated from titanium. Various parameters have been used in the literature to assess the mechanical properties of screw heads, including maximum torque, 9,20,23 absorbed energy,9,23 maximum stress,9,23 and polar moment of inertia,9,23 but inappropriate
parameters may lead to incorrect conclusions. In reality, the most reasonable indicator is the torque (defined in this study as maximum torque) required to disengage the
U
screws by overcoming the tight bonding between the screws and the plates or bone.
N
As for the other parameters, high energy without sufficiently high torque results in
A
deformation without screw turning, stress can only be used to evaluate tool life, and
M
the polar moment of inertia cannot be used to analyze shafts with non-circular cross-sections. In one previous study that tested 3.5 mm stainless steel screw heads
D
with hex sockets, all the screw sockets slipped with multiple peaks on the load
TE
deformation curves.20 The second peak reached only half the first maximum torque,
EP
indicating that once the sockets slipped, the drivers lost their capability to drive the screws. These authors recommended a fluted multi-edged design akin to Torx to
CC
reduce slippage.
In the present study, drivers warped and eventually failed in all small socket
A
screws (3.5 mm) under the 2 mm condition because of the relatively weak driver core. The hex had a slightly higher torque than the Torx design because of its larger core. The cruciate design had the lowest torque because it has the smallest core. These results showed that, in screw jamming situations, small sockets might not provide enough power for further screw driving. Consequently, enlarging the sockets to obtain a higher
torque was warranted, which was proven in the present study. However, although the torque was higher for large hex screws than for small ones under the 2 mm condition (40%), all screws slipped and stripped because of their small driving angles (30°). Contrastingly, large Torx heads showed driver failures similar to those of small Torx
SC RI PT
heads, but had a much greater increase in maximum torque (81.9%). The Octatorx
design achieved the highest torque overall, over twice that of the small hex. Its failure pattern included yielding of driver and socket lobes with tight incarceration in each
other, followed by driver failure. The Torx+ I also experienced driver failure because of its smaller driver core. The Torx+ II had higher torque than the Torx and Torx+ I
U
designs because of its larger core, even with a failure pattern of complete reaming of the
N
socket lobes. It was postulated that Torx+ II torque could be further improved by
A
increasing the area ratio of the socket and driver lobes. The torque of the Torx+ III
M
design was much lower, with screw head twisting failures resulting from the overly thin screw head wall (0.19 mm). With the 1 mm insertion depth, the torques of all designs
D
were lower than the torques observed under the 2 mm condition. The 1 mm insertion
TE
depth represented inappropriate driver insertion during clinical screwing. Increasing the
EP
torque in 1 mm conditions could prevent the sequelae of inappropriate driver insertion. In the present study, the Octatorx exhibited the greatest decrease in torque from the 2
CC
mm condition to the 1 mm condition; however, it displayed the highest torque in the 1 mm condition because its driver lobes were located more peripherally than those of
A
Torx and Torx+ I, owing to the larger driver core. Although Torx+ II has a larger driver core than Octatorx, the torque in the 1 mm condition remained lower than that of Octatorx because of its small socket lobes. This also supported the above-mentioned strategy of increasing the area ratio between the socket and driver lobes. Based on the above findings, increasing the driver core could effectively increase the torque in both
the 1 and 2 mm conditions, and should be of primary consideration during socket design. Based on the failure patterns and the maximum torques, driver-screw constructs could be divided into three regions if no slippage occurred: the driver core, lobe (area
SC RI PT
between the screw head and driver core, including driver lobes and socket lobes), and
the screw head. Mechanically, when a cylinder is subjected to a torsional load, torque will be highest in the center and progressively decrease toward the periphery because of the decreased lever arm. Therefore, the driver core region sustained the highest torque during screwing, followed by the lobe region and the screw head region. This
U
explained why drivers required much more material than screw heads to resist torsion,
N
as well as why smaller driver lobes located more peripherally were able to resist
A
higher torques than larger driver lobes located close to the center. Basically, with the
M
same total area, increasing the core area of drivers would increase their core strength, but conversely decrease lobe area and strength, and vice versa. Therefore, the
D
torsional strength of the driver is maximized when the strength of the core is equal to
TE
that of the lobes. This explains why the maximum torque was the highest in Octatorx,
EP
with simultaneous failure of both the core and lobes. Similarly, the torque of the driver-screw constructs should be the highest when the screw head, lobes, and driver
CC
can fail simultaneously. Accordingly, the Octatorx and Torx plus designs could be further improved through socket enlargement. Furthermore, in Torx plus, the driving
A
angle could be maintained closer to 90° if eight rather than six lobes are used. It was additionally found that in situations with driver failure, including small hex, small Torx, cruciate, large Torx, Octatorx, and Torx+ I, the maximum torque was proportional to the polar moment of inertia (r=0.95), which displayed a linear relationship with the fourth order of the diameter of the driver core. This complies
with the mathematical equation of torsion T/J=Gθ/L, in which T is the torque, J is the polar moment of inertia (equal to π/32(D4), where D is the diameter), G is the material modulus, θ is the angle of rotation, and L is the length of the cylinder. Using this equation, it is possible to calculate the driver core strength. The strength of the lobe
SC RI PT
region and screw head can also be calculated in a similar manner using equations for hollow cylinders, in which J is equal to π/32(D4- d4), where R is the outer diameter and r is the inner diameter. The strength of the three regions in the driver-screw
constructs can be calculated and compared, and the maximum torque of the design
can be predicted by taking the lowest one among them. This mathematical method
U
may be useful in designing new sockets, reducing cost and effort in research and
N
development. It remains important to note that the strength of the lobe region is also
A
affected by the structural ratio between the driver and socket lobes.
M
The present study met several practical limitations. First, only screw heads with a diameter of 5.5 mm and conical angle of 19° were tested. The methods and
D
principles of this study could, however, be applied to all types of screw heads with
TE
different structures as well as screw extractors, which may break during application.
EP
Second, the present study may not be entirely accurate in its simulation of clinical conditions, in which repeated driver usage may be associated with wear. However, it
CC
is reasonable to assume that socket designs with higher torques might still be more capable of driving screws when worn. Further, the ultimate failure torque might be
A
somewhat higher than the highest achievable torque during manual use for ordinary screwdrivers. However, the turning torque of these screwdrivers could be improved by using a T-handle, increasing the handle size, and so on. In the present study, smaller tolerance between the driver and the socket might explain the higher maximum torque and higher incidence of driver failure, which is not common in
clinical practice.20 Third, the material used for screwdriver fabrication in this study was quenched stainless steel. Results may vary if other materials are used. Theoretically, torque could be increased by using materials with a higher stiffness and strength. Fourth, if the degree of freedom for the socket design can be increased (e.g.,
SC RI PT
no lobe arcs), the maximum torque might be further increased. Lastly, under conditions with severe cold welding (seen in Fig. 7B), unscrewing was impossible regardless of socket design. Further studies to prevent cold welding are necessary.
Conclusions
U
Conventional hex sockets showed the disadvantages of slipping and low
N
maximum torque, and were vulnerable to inappropriate driver insertion. Enlarging the
A
socket could effectively increase the torque. Torx designs could prevent such slippage
M
and had significantly higher torque than hex; however, Octatorx and Torx plus designs could achieve even better performance. Furthermore, it was possible to predict the
D
maximum torque mathematically, and the equations can be used for future design
TE
modifications. Further studies are required to explore more optimal designs.
EP
Acknowledgement
We would like to thank the Ministry of Science and Technology, Taiwan for their
CC
support under the Grant MOST 104-2221-E-002 -102 -MY2.
A
References 1. Cole PA, Zlowodzki M, Kregor PJ. Treatment of proximal tibia fractures using the less invasive stabilization system: surgical experience and early clinical results in 77 fractures. J Orthop Trauma 2004;18:528-35. 2. Gyuricza C, Carlson MG, Weiland AJ, Wolfe SW, Hotchkiss RN, Daluiski A. Removal of locked volar plates after distal radius fractures. J Hand Surg Am
2011;36:982-5. 3. Musters GD, Boele van Hensbroek P, Ponsen KJ, Luitse JS, Goslings JC. Locking Compression Plates are more difficult to remove than conventional non-locking plates. Eur J Trauma Emerg S 2013;39:159-62.
SC RI PT
4. Haug RH, Jenkins WS, Brandt MT. Advances in plate and screw technology: thoughts on design and clinical applications. Semin Plast Surg 2002;16:219-28.
5. Miller DL, Goswami T. A review of locking compression plate biomechanics and their
advantages
as
internal fixators
in fracture
2007;22:1049-62.
healing.
Clin
Biomech
U
6. Cronier P, Pietu G, Dujardin C, Bigorre N, Ducellier F, Gerard R. The concept of
N
locking plates. Orthop Traumatol Surg Res 2010;96S:S17-S36.
A
7. Dhakar A, Annappa R, Gupta M, Harshwardhan H, Kotian P, Suresh PK. Minimally
Diagn Res 2016;10:RC01-4.
M
Invasive Plate Osteosynthesis with Locking Plates for Distal Tibia Fractures. J Clin
D
8. Van Nortwick SS, Yao J, Ladd AL. Titanium integration with bone, welding, and
TE
screw head destruction complicating hardware removal of the distal radius: report of 2
EP
cases. J Hand Surg Am 2012;37:1388-92. 9. Jun DS, Kou H, Kim YC, Jun C-S, Lee SC. Evaluation of a cloverleaf screw head
CC
to minimize the slippage of medical screws. Int J Precis Eng Man 2011;12:703-9. 10. Lehmen JA, Della Rocca GJ, Murtha YM, Crist BD. Removal technique for
A
cold-welded titanium locking screws. Injury 2011;42:1377-9. 11. Ehlinger M, Adam P, Simon P, Bonnomet F. Technical difficulties in hardware removal in titanium compression plates with locking screws. Orthop Traumatol Surg Res 2009;95:373-6. 12. Schwarz N, Euler S, Schlittler M, Ulbing T, Wilhelm P, Fronhofer G, Irnstorfer M.
Technical complications during removal of locking screws from locking compression plates: a prospective multicenter study. Eur J Trauma Emerg S 2013;39:339-44. 13. Kumar G, Dunlop C. Case report: A technique to remove a jammed locking screw from a locking plate. Clin Orthop Relat R 2011;469:613-6.
SC RI PT
14. Suzuki T, Smith WR, Stahel PF, Morgan SJ, Baron AJ, Hak DJ. Technical problems and complications in the removal of the less invasive stabilization system. J Orthop Trauma 2010;24:369-73.
15. Maehara T, Moritani S, Ikuma H, Shinohara K, Yokoyama Y. Difficulties in removal of the titanium locking plate in Japan. Injury 2013;44:1122-6.
U
16. Kim SJ, Kim MU. A simple technique for removing a locking compression plate
N
with a stripped locking screw. J Orthop Trauma 2012;26:e51-e3.
A
17. Fujita K, Yasutake H, Horii T, Hashimoto N, Kabata T, Tsuchiya H. Difficulty in
M
locking head screw removal. J Orthop Sci 2014;19:304-7. 18. Mitsukawa N. Clinical experience with the screw extraction set for broken screw.
D
J Craniofac Surg 2011;22:226-9.
TE
19. Gopinathan NR, Dhillon MS, Kumar R. Surgical technique: Simple technique for
EP
removing a locking recon plate with damaged screw heads. Clin Orthop Relat R 2013;471:1572-5.
CC
20. Behring JK, Gjerdet NR, Mølster A. Slippage between screwdriver and bone screw. Clin Orthop Relat R 2002;404:368-72.
A
21. Spencer KR, Ferguson JW, Smith AC, Palamara JEA. Screw head design: an experimental study to assess the influence of design on performance. J Oral Maxil Surg 2004;62:473-8. 22. Klein SA, Kenney NA, Nyland JA, Seligson D. Evaluation of cruciate and slot auxiliary screw head design modifications for extracting stripped screw heads. Acta
Orthop Belg 2007;73:772. 23. Chien CY, Chuang WH, Tsai WC, Lin SC, Luh YP, Chen YJ. A finite element study about cam-out failure of the recess-screwdriver interfaces for the cold-welded periarticular fixation. J Mech 2012;29:N1-N7. Wikipedia
contributors.
Torx.
15
August
2017
23:35
UTC,
from
SC RI PT
24.
https://en.wikipedia.org/w/index.php?title=Torx&oldid=795704660
25. Hayes J, Richards R. The use of titanium and stainless steel in fracture fixation. Expert Rev Med Devic 2010;7:843-53. 26.
Wikipedia
contributors.
Galling.
21
April
2017
06:47
Figure Legends
A
CC
EP
TE
D
M
Fig. 1 Geometry of socket designs.
A
N
U
https://en.wikipedia.org/w/index.php?title=Galling&oldid=776477518
UTC,
from
SC RI PT U N A M
A
CC
EP
TE
D
Fig. 2 Screw head structure terminology.
A
N
U
SC RI PT
Fig. 3 Biomechanical testing set-up.
M
Fig. 4 Small socket load-deformation curves and failure patterns under 2 mm
A
CC
EP
TE
D
conditions.
A
N
U
SC RI PT
Fig. 5 Large socket load-deformation curves under 2 mm conditions.
A
CC
EP
TE
D
M
Fig. 6 Large socket failure patterns under 2 mm conditions.
Fig. 7 Different degrees of cold welding (red arrows) between plates and screws (50X) in clinical setting: approximately complete in femoral plate (A) and partial in humeral
A
CC
EP
TE
D
M
A
N
U
SC RI PT
plate (B).
Table 1. Geometric design values Diameter of Diameter of Radius of Driving Screw head Depth of socket Size Type driver core lobe arc angle wall thickness socket (mm) (mm) (mm) (mm) (°) (mm) 2.89 2.5 30 0.96 2 Hex Small 3.23 2.32 1.16 63.5 0.76 2 Torx
Cruciate
3.85
0.57
-
81.5
0.48
2
Hex
3.93
3.4
-
30
0.44
2
Torx
3.95
2.84
1.51
62.8
0.43
2
Octatorx
3.85
3.07
1.00
59.4
0.48
2
Torx+ I
4.25
2.72
0.98
90.0
0.28
2
Torx+ II
3.95
3.25
0.39
89.9
0.43
2
Torx+ III
4.29
3.63
0.27
106.4
0.19
2.41
SC RI PT
Large
Table 2. The maximum torque, rotation angle and failure pattern. Maximum torque (Nm)
Maximum angle (°)
5.76±0.28 5.07±0.07
56.2±13.6 51.8±3.71
*Driver failure Driver failure
Hex Torx Cruciate
4.59±0.24 4.42±0.66 5.09±0.12 4.43±0.31
74.2±9.20 40.1±5.78 72.4±6.28 99.2±8.50
Driver failure Driver slip Driver failure Driver failure
Hex Torx Octatorx Torx+ I
8.06±0.46 9.22±0.09 10.97±0.13 8.93±0.10
25.8±3.65 20.9±2.06 22.3±2.11 17.1±0.56
Driver slip Driver failure Driver failure with lobe incarceration Driver failure
Torx+ II Torx+ III
9.87±0.34 8.07±0.91
9.44±0.90 10.9±1.55
Socket lobe failure Screw head failure
Hex Torx Octatorx Torx+ I
5.03±0.46 7.85±0.39 8.31±0.38 8.25±0.25
21.6±3.16 21.1±1.69 16.7±0.71 23.8±5.17
Driver slip Driver lobe failure Driver lobe failure Driver failure (2), driver lobe failure (8)
Torx+ II
4.53±0.65
7.28±0.43
Socket lobe failure
Small 1
CC
Large
EP
2
1
A
Torx Cruciate
TE
2
U
Failure pattern (specimen number)
N
Hex
A
Socket
M
Depth (mm)
D
Size
*Driver failure indicated failure of both core and lobes.
S0020-1383(18)30056-1 https://doi.org/10.1016/j.injury.2018.01.036 JINJ 7577
To appear in:
Injury, Int. J. Care Injured
Accepted date:
31-1-2018
Please cite this article as: Lin Chen-Huei, Chao Ching-Kong, Tang Yi-Hsuan, Lin Jinn.Improving socket design to prevent difficult removal of locking screws.Injury https://doi.org/10.1016/j.injury.2018.01.036 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Improving socket design to prevent difficult removal of locking screws
SC RI PT
Chen-Huei Lin, MDa, Ching-Kong Chao, PhDa, Yi-Hsuan Tang, MDa,Jinn Lin, PhDb,* a
Department of Mechanical Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan b Department of Orthopedics, National Taiwan University Hospital, Taipei, Taiwan
A
CC
EP
TE
D
M
A
N
U
* Corresponding author. E-mail address: [email protected] (J. Lin).
Abstract Introduction: Reports of driver slippage leading to difficult locking screw removals have increased since the adoption of titanium for screw fabrication; the use of
SC RI PT
titanium is known to cause cross-threading and cold welding. Such problems occur most frequently in screws with hex sockets, and may cause serious surgical
complications. This study aimed to improve screw socket design to prevent slippage and difficult screw removal.
Materials and methods: Three types of small sockets (hex, Torx, and cruciate) and
U
six types of large sockets (hex, Torx, Octatorx, Torx+ I, Torx+ II, and Torx+ III) with screw head diameters of 5.5 mm were manufactured from titanium, and
N
corresponding screwdrivers were manufactured from stainless steel. The screw heads
A
and drivers were mounted on a material testing machine, and torsional tests were
M
conducted to simulate screw usage in clinical settings at two insertion depths: 1 and 2
D
mm. Ten specimens were tested from each design, and the maximum torque and
TE
failure patterns were recorded and compared. Results: For small sockets in 2 mm conditions, the hex with the largest driver core
EP
had the highest torque, followed by Torx and cruciate. In these tests, the drivers were twisted off in all specimens. However, under the 1 mm condition, the hex slipped and
CC
the torque decreased markedly. Overall, torque was higher for large sockets than for
A
small sockets. The Octatorx, with a large core and simultaneous deformation of the driver and socket lobes, had the highest torque at almost twice that of the small hex. The hex had the lowest torque, a result of slippage in both the 1 and 2 mm conditions. Torx plus designs, with more designed degrees of freedom, were able to maintain a higher driving angle and larger core for higher torque. Conclusions: The hex design showed slipping tendencies with a marked decrease in
torque, especially under conditions with inadequate driver engagement. Large sockets allowed for substantial increases in torque. The Torx, Octatorx, and Torx plus designs displayed better performance than the hexes. Improvements to the socket design could
SC RI PT
effectively prevent slippage and solve difficult screw removal problems.
A
CC
EP
TE
D
M
A
N
U
Keywords: Locking plate; Titanium; Screw head socket; Torsion test; Maximal torque.
Introduction The use of locking plates can provide high fixation stability for reliable healing in various types of bodily fractures, aiding in positive functional recovery.1-5 Minimally invasive surgical techniques can further decrease tissue damage and hasten the
SC RI PT
postoperative recovery process.5-7 These plates are commonly manufactured from stainless steel and titanium alloys, the latter of which has been particularly advocated for recently because of its advantages: higher biocompatibility, fewer resultant artifacts
on computer tomographic scans and magnetic resonance images, higher fatigue strength, lower infection risk, and isoelasticity to bone as compared with stainless steel.3,6,8,9
U
However, as a malleable metal, titanium is prone to cross-threading or cold welding
N
under strong metal-to-metal contact; therefore, its use more easily results in socket
A
damage.6,10-12 Reports of difficulty removing jammed titanium screws have become
M
more frequent.13,14 Difficult removals in a clinical setting may result in serious complications such as prolonged operations, bleeding, infection, neurovascular injuries,
D
and even iatrogenic fractures.3,13,14 Typically, difficulties in removal begin when the
TE
screw head sockets are stripped, rendering drivers useless. 10,15,16 Multiple removal methods have been invented, including foil insertion, screw head drilling, and reverse
EP
thread extraction, but none yet guarantee a successful removal.3,11,16-18 Screw jamming
CC
prevention has also been emphasized by methods such as the use of an aiming sleeve, proper driver engagement, use of torque-limiting screwdrivers, and using screwdriver
A
tips in good condition.2,3,8,19,20 Mechanically, improving screw head socket designs to prevent stripping and to increase torque to facilitate unscrewing may prove more important than these preventative techniques; however, few studies have focused on the mechanical behaviors of different socket designs, especially for titanium screws.21-23 In the present study, different screw head socket designs were specially
manufactured from titanium, including the most conventional hexagonal sockets. The corresponding screwdrivers were made from stainless steel. The screw heads and drivers were then mounted on a material testing machine, and torsional tests simulating clinical applications were conducted. For results, the maximum torque values and
SC RI PT
failure patterns were recorded. The testing results from each of the different socket
designs were compared with one another, and factors that could increase the torque and prevent stripping were analyzed.
Materials and Methods
U
Structures of screw heads and drivers
N
The screw heads and drivers were specially manufactured from a titanium alloy
A
(Ti6Al4V, F136-13) and stainless steel (AISI 420, F138) (Carpenter Technology,
M
Reading, PA, USA), respectively. The yield strength was 627 MPa for the stainless steel and 798 MPa for the titanium. Two types of socket (recess) were used: small and
D
large (Fig. 1). Three design types were used for the small sockets: hex, Torx, and
TE
cruciate (Table 1). The small hex is currently the most commonly used design for 3.5 mm screws (ISO 4762, M3). To ensure a fair comparison, the areas of the cruciate and
EP
Torx sockets were kept equal to that of the hex. Six large socket designs were also
CC
tested: hex, Torx, Octatorx, Torx+ I, Torx+ II, and Torx+ III. Each screw head had an outer diameter of 5.5 mm, a height of 3 mm, and a conical angle of 19° (Fig. 2). As
A
seen in Fig. 1, Torx+ III had wider and deeper sockets than Torx+ II. The large sockets were the modified designs of the small ones to improve the performance. All sockets had a depth of 2 mm and completely straight sidewalls to prevent cam-out. A 0.02 mm manufacturing tolerance was set for both sockets and drivers. The screw heads were attached to a solid titanium cylinder that could be firmly gripped by the fixture,
thereby avoiding any confounding effects the screw shafts may have caused. Corresponding drivers, which were also attached to a solid stainless steel cylinder for firm grip by the fixture, were fabricated with a length of 3 mm.
SC RI PT
Biomechanical test and Failure analysis
Single load torsional tests using a servohydraulic material testing machine (8874,
Instron Industrial Products, Grove City, PA, USA) were conducted to evaluate the
mechanical performance of the screws and drivers (Fig. 3). The screw head and driver cylinders were first mounted on the machine co-linearly, and checked repeatedly to
U
ensure exact alignment between the rotational centers of the screw heads and drivers.
N
The drivers were inserted into sockets at two depths: 1 and 2 mm. The 2 mm insertion
A
depth represented an appropriate screwing technique, whereas the 1 mm depth
M
represented inappropriate insertion. Ten pairs of new screws and drivers were tested for each socket design. During the tests, the drivers were rotated approximately 240°
D
in the clockwise direction until the torque dropped to approximately zero with a
TE
loading rate of 15 °/min. The torque and angle were recorded continuously throughout
EP
this procedure, and the highest recorded torque was defined as the maximum torque. Once testing was complete, the failure patterns were observed and correlated to
CC
the maximum torque and load-deformation curve.
A
Statistical analysis One-way ANOVA tests (SPSS v22.0, SPSS Inc., Chicago, IL, USA) were used during biomechanical testing to compare the torsional strengths of small and large sockets. Bonferroni tests were used for post-hoc pairwise comparisons. The significance level was defined as p <0.05.
Results Small sockets Under the 2 mm condition, the hex design had highest torque, followed by Torx, and
SC RI PT
then cruciate (Table 2). All socket designs demonstrated driver warping and failure by
the completion of testing. All drivers were twisted off at the top of the sockets (Fig. 4). Because each screw experienced very small standard deviations (4.9%, 1.42%, and
5.15% for hex, Torx, and cruciate, respectively), the differences in each comparison were statistically significant. Conversely, under the 1 mm condition the hex
U
experienced the highest torque decrease in all cases (23%) because the drivers slipped
N
in the sockets. The decrease was lower in the cruciate (3.5%) and Torx (-0.04%),
M
A
while the failure mode was the same as that observed in 2 mm conditions.
Large sockets
D
There were two types of load deformation curves: those with one peak and those with
TE
multiple peaks. Single peak curves reflected conditions without driver slippage.
EP
Multiple peaks with progressively decreasing peak torques reflected driver slippage in the sockets (Fig. 5). The maximum torque observed in the large sockets was
CC
substantially higher than that observed in the small sockets. For the hex designs, all sockets displayed both slipping and stripping under 1 and 2 mm conditions, with an
A
obvious torque decrease of 38% under the 1 mm condition. Octatorx allowed the highest torque for the 2 mm condition, followed by Torx and Torx+ I (Fig. 5) (Table 2). All the drivers were eventually twisted off in these three designs (Fig. 6). There was also a marked deformation of the driver lobes in the Octatorx design, which remained tightly incarcerated in the socket lobes. Driver lobes remained relatively
well preserved in the Torx and Torx + I designs. Under 1 mm conditions, these three designs displayed similar failure patterns with breakage of the driver lobes. The torque decrease was lower in Torx and Torx+ I (14.8% and 7.6%, respectively) with small driver cores, but higher in Octatorx (24.4%) with larger cores. Under the 2 mm
SC RI PT
condition, the socket lobes were reamed off in the Torx+ II design, although the
drivers remained relatively intact. In 1 mm conditions, the failure pattern was the
same as that in 2 mm conditions and the torque was only approximately half (54.2%). For Torx+ III, the torque was relatively low in the 2 mm condition with screw head
failure occurring because of the thin walls of the screw head (0.19 mm). The standard
U
deviations were also small in the large socket screws, ranging from 1% to 11.3%, and
A
M
designs except hex and Torx+ III.
N
the differences between each pair reached a level of statistical significance for all
Discussion
D
Many different screw head socket types have been designed for orthopaedic screws. The
TE
earliest single slot and cruciate sockets proved to be disadvantaged by driver slippage
EP
and low torque transfer.6,20 Later hexagonal and square sockets with vertical side walls were able to prevent cam-out and allow for higher torque transfer, and their wide
CC
contact points also ensured a tight grip between the driver and screw head for both reliable screw transfer between the operating staffs and effective screw turning with a
A
small end load. The hex socket, possessing smaller wrench-swing arcs, is better than a square design in situations with surrounding obstacles. However, hex screws frequently experience slipping, which may easily strip the sockets. 3,9,21 The Torx design, developed in 1967 with a six-point star-shaped configuration, allows for higher torque transfer than a similarly sized hex socket.4,24 Its large driving angles prevent damage to the screw
head and tool and increase torque transfer. Driving angles represent the angles between the screwing direction and the side of the lobes (Fig. 2). Low driving angles may cause higher radial force, which is perpendicular to the turning force and not effective in screwing, whereas high driving angles can generate higher screwing torque and reduce
SC RI PT
the damaging radial force.23,24 When the driving angle is too low, it may not only burst the socket and crush the driver, and excessive radial force will also round off the corners
of both socket and driver, causing slippage. High driving angles allow the Torx design to have a smaller head under the same required torque. This can be an advantage when space is limited, such as in locking plates where a smaller screw head allows for smaller
U
screw holes and thus increases the mechanical strength of the plates. However, the
N
disadvantage of Torx sockets is their smaller internal "splines;" these can wear easily,
A
resulting in slippage and shorted tool life. A Torx successor was introduced around 1990
M
when the original Torx patent was expiring.24 This design, the Torx Plus, features more
transfer and lower wear.
D
square lobes for 90° driving angles, allowing for even greater increases in torque
TE
As locking plates are increasingly made from titanium, with its cold welding and bone integration properties,8,15,25 implant removal has become a challenging task,
EP
especially for screws with hex sockets.10,23 Although the cold welding between the
CC
locking plates and screws has never been demonstrated micro- or macroscopically in the literature and its reality has been questioned,6,8,11,14 different degrees of cold
A
welding have been observed by the present authors (Fig. 7). High ductility and flowing material is considered a general prerequisite for excessive cold welding.26 It has been reported that some factors contributing to increased risk of difficult removals include screw size (3.5 mm in particular, though larger screws are not exempt), screws near the elbow joints, screws at the diaphysis, delayed removal time, good
bone quality, screw length, and young age.6,11,12,17 Recently, Torx sockets were introduced as a “cutting-edge” advancement in locking plates, aiming to increase torque and resist stripping.4 However, there have been no mechanical studies comparing the performance of this screw socket to other designs, especially those
SC RI PT
fabricated from titanium. Various parameters have been used in the literature to assess the mechanical properties of screw heads, including maximum torque, 9,20,23 absorbed energy,9,23 maximum stress,9,23 and polar moment of inertia,9,23 but inappropriate
parameters may lead to incorrect conclusions. In reality, the most reasonable indicator is the torque (defined in this study as maximum torque) required to disengage the
U
screws by overcoming the tight bonding between the screws and the plates or bone.
N
As for the other parameters, high energy without sufficiently high torque results in
A
deformation without screw turning, stress can only be used to evaluate tool life, and
M
the polar moment of inertia cannot be used to analyze shafts with non-circular cross-sections. In one previous study that tested 3.5 mm stainless steel screw heads
D
with hex sockets, all the screw sockets slipped with multiple peaks on the load
TE
deformation curves.20 The second peak reached only half the first maximum torque,
EP
indicating that once the sockets slipped, the drivers lost their capability to drive the screws. These authors recommended a fluted multi-edged design akin to Torx to
CC
reduce slippage.
In the present study, drivers warped and eventually failed in all small socket
A
screws (3.5 mm) under the 2 mm condition because of the relatively weak driver core. The hex had a slightly higher torque than the Torx design because of its larger core. The cruciate design had the lowest torque because it has the smallest core. These results showed that, in screw jamming situations, small sockets might not provide enough power for further screw driving. Consequently, enlarging the sockets to obtain a higher
torque was warranted, which was proven in the present study. However, although the torque was higher for large hex screws than for small ones under the 2 mm condition (40%), all screws slipped and stripped because of their small driving angles (30°). Contrastingly, large Torx heads showed driver failures similar to those of small Torx
SC RI PT
heads, but had a much greater increase in maximum torque (81.9%). The Octatorx
design achieved the highest torque overall, over twice that of the small hex. Its failure pattern included yielding of driver and socket lobes with tight incarceration in each
other, followed by driver failure. The Torx+ I also experienced driver failure because of its smaller driver core. The Torx+ II had higher torque than the Torx and Torx+ I
U
designs because of its larger core, even with a failure pattern of complete reaming of the
N
socket lobes. It was postulated that Torx+ II torque could be further improved by
A
increasing the area ratio of the socket and driver lobes. The torque of the Torx+ III
M
design was much lower, with screw head twisting failures resulting from the overly thin screw head wall (0.19 mm). With the 1 mm insertion depth, the torques of all designs
D
were lower than the torques observed under the 2 mm condition. The 1 mm insertion
TE
depth represented inappropriate driver insertion during clinical screwing. Increasing the
EP
torque in 1 mm conditions could prevent the sequelae of inappropriate driver insertion. In the present study, the Octatorx exhibited the greatest decrease in torque from the 2
CC
mm condition to the 1 mm condition; however, it displayed the highest torque in the 1 mm condition because its driver lobes were located more peripherally than those of
A
Torx and Torx+ I, owing to the larger driver core. Although Torx+ II has a larger driver core than Octatorx, the torque in the 1 mm condition remained lower than that of Octatorx because of its small socket lobes. This also supported the above-mentioned strategy of increasing the area ratio between the socket and driver lobes. Based on the above findings, increasing the driver core could effectively increase the torque in both
the 1 and 2 mm conditions, and should be of primary consideration during socket design. Based on the failure patterns and the maximum torques, driver-screw constructs could be divided into three regions if no slippage occurred: the driver core, lobe (area
SC RI PT
between the screw head and driver core, including driver lobes and socket lobes), and
the screw head. Mechanically, when a cylinder is subjected to a torsional load, torque will be highest in the center and progressively decrease toward the periphery because of the decreased lever arm. Therefore, the driver core region sustained the highest torque during screwing, followed by the lobe region and the screw head region. This
U
explained why drivers required much more material than screw heads to resist torsion,
N
as well as why smaller driver lobes located more peripherally were able to resist
A
higher torques than larger driver lobes located close to the center. Basically, with the
M
same total area, increasing the core area of drivers would increase their core strength, but conversely decrease lobe area and strength, and vice versa. Therefore, the
D
torsional strength of the driver is maximized when the strength of the core is equal to
TE
that of the lobes. This explains why the maximum torque was the highest in Octatorx,
EP
with simultaneous failure of both the core and lobes. Similarly, the torque of the driver-screw constructs should be the highest when the screw head, lobes, and driver
CC
can fail simultaneously. Accordingly, the Octatorx and Torx plus designs could be further improved through socket enlargement. Furthermore, in Torx plus, the driving
A
angle could be maintained closer to 90° if eight rather than six lobes are used. It was additionally found that in situations with driver failure, including small hex, small Torx, cruciate, large Torx, Octatorx, and Torx+ I, the maximum torque was proportional to the polar moment of inertia (r=0.95), which displayed a linear relationship with the fourth order of the diameter of the driver core. This complies
with the mathematical equation of torsion T/J=Gθ/L, in which T is the torque, J is the polar moment of inertia (equal to π/32(D4), where D is the diameter), G is the material modulus, θ is the angle of rotation, and L is the length of the cylinder. Using this equation, it is possible to calculate the driver core strength. The strength of the lobe
SC RI PT
region and screw head can also be calculated in a similar manner using equations for hollow cylinders, in which J is equal to π/32(D4- d4), where R is the outer diameter and r is the inner diameter. The strength of the three regions in the driver-screw
constructs can be calculated and compared, and the maximum torque of the design
can be predicted by taking the lowest one among them. This mathematical method
U
may be useful in designing new sockets, reducing cost and effort in research and
N
development. It remains important to note that the strength of the lobe region is also
A
affected by the structural ratio between the driver and socket lobes.
M
The present study met several practical limitations. First, only screw heads with a diameter of 5.5 mm and conical angle of 19° were tested. The methods and
D
principles of this study could, however, be applied to all types of screw heads with
TE
different structures as well as screw extractors, which may break during application.
EP
Second, the present study may not be entirely accurate in its simulation of clinical conditions, in which repeated driver usage may be associated with wear. However, it
CC
is reasonable to assume that socket designs with higher torques might still be more capable of driving screws when worn. Further, the ultimate failure torque might be
A
somewhat higher than the highest achievable torque during manual use for ordinary screwdrivers. However, the turning torque of these screwdrivers could be improved by using a T-handle, increasing the handle size, and so on. In the present study, smaller tolerance between the driver and the socket might explain the higher maximum torque and higher incidence of driver failure, which is not common in
clinical practice.20 Third, the material used for screwdriver fabrication in this study was quenched stainless steel. Results may vary if other materials are used. Theoretically, torque could be increased by using materials with a higher stiffness and strength. Fourth, if the degree of freedom for the socket design can be increased (e.g.,
SC RI PT
no lobe arcs), the maximum torque might be further increased. Lastly, under conditions with severe cold welding (seen in Fig. 7B), unscrewing was impossible regardless of socket design. Further studies to prevent cold welding are necessary.
Conclusions
U
Conventional hex sockets showed the disadvantages of slipping and low
N
maximum torque, and were vulnerable to inappropriate driver insertion. Enlarging the
A
socket could effectively increase the torque. Torx designs could prevent such slippage
M
and had significantly higher torque than hex; however, Octatorx and Torx plus designs could achieve even better performance. Furthermore, it was possible to predict the
D
maximum torque mathematically, and the equations can be used for future design
TE
modifications. Further studies are required to explore more optimal designs.
EP
Acknowledgement
We would like to thank the Ministry of Science and Technology, Taiwan for their
CC
support under the Grant MOST 104-2221-E-002 -102 -MY2.
A
References 1. Cole PA, Zlowodzki M, Kregor PJ. Treatment of proximal tibia fractures using the less invasive stabilization system: surgical experience and early clinical results in 77 fractures. J Orthop Trauma 2004;18:528-35. 2. Gyuricza C, Carlson MG, Weiland AJ, Wolfe SW, Hotchkiss RN, Daluiski A. Removal of locked volar plates after distal radius fractures. J Hand Surg Am
2011;36:982-5. 3. Musters GD, Boele van Hensbroek P, Ponsen KJ, Luitse JS, Goslings JC. Locking Compression Plates are more difficult to remove than conventional non-locking plates. Eur J Trauma Emerg S 2013;39:159-62.
SC RI PT
4. Haug RH, Jenkins WS, Brandt MT. Advances in plate and screw technology: thoughts on design and clinical applications. Semin Plast Surg 2002;16:219-28.
5. Miller DL, Goswami T. A review of locking compression plate biomechanics and their
advantages
as
internal fixators
in fracture
2007;22:1049-62.
healing.
Clin
Biomech
U
6. Cronier P, Pietu G, Dujardin C, Bigorre N, Ducellier F, Gerard R. The concept of
N
locking plates. Orthop Traumatol Surg Res 2010;96S:S17-S36.
A
7. Dhakar A, Annappa R, Gupta M, Harshwardhan H, Kotian P, Suresh PK. Minimally
Diagn Res 2016;10:RC01-4.
M
Invasive Plate Osteosynthesis with Locking Plates for Distal Tibia Fractures. J Clin
D
8. Van Nortwick SS, Yao J, Ladd AL. Titanium integration with bone, welding, and
TE
screw head destruction complicating hardware removal of the distal radius: report of 2
EP
cases. J Hand Surg Am 2012;37:1388-92. 9. Jun DS, Kou H, Kim YC, Jun C-S, Lee SC. Evaluation of a cloverleaf screw head
CC
to minimize the slippage of medical screws. Int J Precis Eng Man 2011;12:703-9. 10. Lehmen JA, Della Rocca GJ, Murtha YM, Crist BD. Removal technique for
A
cold-welded titanium locking screws. Injury 2011;42:1377-9. 11. Ehlinger M, Adam P, Simon P, Bonnomet F. Technical difficulties in hardware removal in titanium compression plates with locking screws. Orthop Traumatol Surg Res 2009;95:373-6. 12. Schwarz N, Euler S, Schlittler M, Ulbing T, Wilhelm P, Fronhofer G, Irnstorfer M.
Technical complications during removal of locking screws from locking compression plates: a prospective multicenter study. Eur J Trauma Emerg S 2013;39:339-44. 13. Kumar G, Dunlop C. Case report: A technique to remove a jammed locking screw from a locking plate. Clin Orthop Relat R 2011;469:613-6.
SC RI PT
14. Suzuki T, Smith WR, Stahel PF, Morgan SJ, Baron AJ, Hak DJ. Technical problems and complications in the removal of the less invasive stabilization system. J Orthop Trauma 2010;24:369-73.
15. Maehara T, Moritani S, Ikuma H, Shinohara K, Yokoyama Y. Difficulties in removal of the titanium locking plate in Japan. Injury 2013;44:1122-6.
U
16. Kim SJ, Kim MU. A simple technique for removing a locking compression plate
N
with a stripped locking screw. J Orthop Trauma 2012;26:e51-e3.
A
17. Fujita K, Yasutake H, Horii T, Hashimoto N, Kabata T, Tsuchiya H. Difficulty in
M
locking head screw removal. J Orthop Sci 2014;19:304-7. 18. Mitsukawa N. Clinical experience with the screw extraction set for broken screw.
D
J Craniofac Surg 2011;22:226-9.
TE
19. Gopinathan NR, Dhillon MS, Kumar R. Surgical technique: Simple technique for
EP
removing a locking recon plate with damaged screw heads. Clin Orthop Relat R 2013;471:1572-5.
CC
20. Behring JK, Gjerdet NR, Mølster A. Slippage between screwdriver and bone screw. Clin Orthop Relat R 2002;404:368-72.
A
21. Spencer KR, Ferguson JW, Smith AC, Palamara JEA. Screw head design: an experimental study to assess the influence of design on performance. J Oral Maxil Surg 2004;62:473-8. 22. Klein SA, Kenney NA, Nyland JA, Seligson D. Evaluation of cruciate and slot auxiliary screw head design modifications for extracting stripped screw heads. Acta
Orthop Belg 2007;73:772. 23. Chien CY, Chuang WH, Tsai WC, Lin SC, Luh YP, Chen YJ. A finite element study about cam-out failure of the recess-screwdriver interfaces for the cold-welded periarticular fixation. J Mech 2012;29:N1-N7. Wikipedia
contributors.
Torx.
15
August
2017
23:35
UTC,
from
SC RI PT
24.
https://en.wikipedia.org/w/index.php?title=Torx&oldid=795704660
25. Hayes J, Richards R. The use of titanium and stainless steel in fracture fixation. Expert Rev Med Devic 2010;7:843-53. 26.
Wikipedia
contributors.
Galling.
21
April
2017
06:47
Figure Legends
A
CC
EP
TE
D
M
Fig. 1 Geometry of socket designs.
A
N
U
https://en.wikipedia.org/w/index.php?title=Galling&oldid=776477518
UTC,
from
SC RI PT U N A M
A
CC
EP
TE
D
Fig. 2 Screw head structure terminology.
A
N
U
SC RI PT
Fig. 3 Biomechanical testing set-up.
M
Fig. 4 Small socket load-deformation curves and failure patterns under 2 mm
A
CC
EP
TE
D
conditions.
A
N
U
SC RI PT
Fig. 5 Large socket load-deformation curves under 2 mm conditions.
A
CC
EP
TE
D
M
Fig. 6 Large socket failure patterns under 2 mm conditions.
Fig. 7 Different degrees of cold welding (red arrows) between plates and screws (50X) in clinical setting: approximately complete in femoral plate (A) and partial in humeral
A
CC
EP
TE
D
M
A
N
U
SC RI PT
plate (B).
Table 1. Geometric design values Diameter of Diameter of Radius of Driving Screw head Depth of socket Size Type driver core lobe arc angle wall thickness socket (mm) (mm) (mm) (mm) (°) (mm) 2.89 2.5 30 0.96 2 Hex Small 3.23 2.32 1.16 63.5 0.76 2 Torx
Cruciate
3.85
0.57
-
81.5
0.48
2
Hex
3.93
3.4
-
30
0.44
2
Torx
3.95
2.84
1.51
62.8
0.43
2
Octatorx
3.85
3.07
1.00
59.4
0.48
2
Torx+ I
4.25
2.72
0.98
90.0
0.28
2
Torx+ II
3.95
3.25
0.39
89.9
0.43
2
Torx+ III
4.29
3.63
0.27
106.4
0.19
2.41
SC RI PT
Large
Table 2. The maximum torque, rotation angle and failure pattern. Maximum torque (Nm)
Maximum angle (°)
5.76±0.28 5.07±0.07
56.2±13.6 51.8±3.71
*Driver failure Driver failure
Hex Torx Cruciate
4.59±0.24 4.42±0.66 5.09±0.12 4.43±0.31
74.2±9.20 40.1±5.78 72.4±6.28 99.2±8.50
Driver failure Driver slip Driver failure Driver failure
Hex Torx Octatorx Torx+ I
8.06±0.46 9.22±0.09 10.97±0.13 8.93±0.10
25.8±3.65 20.9±2.06 22.3±2.11 17.1±0.56
Driver slip Driver failure Driver failure with lobe incarceration Driver failure
Torx+ II Torx+ III
9.87±0.34 8.07±0.91
9.44±0.90 10.9±1.55
Socket lobe failure Screw head failure
Hex Torx Octatorx Torx+ I
5.03±0.46 7.85±0.39 8.31±0.38 8.25±0.25
21.6±3.16 21.1±1.69 16.7±0.71 23.8±5.17
Driver slip Driver lobe failure Driver lobe failure Driver failure (2), driver lobe failure (8)
Torx+ II
4.53±0.65
7.28±0.43
Socket lobe failure
Small 1
CC
Large
EP
2
1
A
Torx Cruciate
TE
2
U
Failure pattern (specimen number)
N
Hex
A
Socket
M
Depth (mm)
D
Size
*Driver failure indicated failure of both core and lobes.