SCIENTIFIC ARTICLE

The Durability of the Intrascaphoid Compression of

Headless Compression Screws: In Vitro StudyD. S. Gruszka, MD, K. J. Burkhart, MD, T. E. Nowak, MD, T. Achenbach, MD, P. M. Rommens, MD, PhD,

L. P. Müller, MD, PhD

Purpose To test a new generation of compression screws: the Acumed Acutrak 2 Mini (AA;Acumed, Hillsboro, OR), the Stryker TwinFix (ST; Stryker, Kalamazoo, MI), and theSynthes 3.0 headless compression screw (SH; Synthes, Solothurn, Switzerland).

Methods We used 40 fresh-frozen human scaphoids for this study. Bone density was measured.A K-wire was inserted centrally. A perpendicular osteotomy was created in the middle third(Herbert B2 fracture). A custom-made load sensor was placed between the bone fragments. Allscrews were implanted according to the manufacturers’ instructions. The Synthes 2.0 corticalscrew (SC), implanted as a lag screw, was used as a reference. The compression force during eachexperiment was digitally monitored for 12 hours while the data were acquired. The data wereanalyzed using analysis of variance with the Bonferroni correction.

Results Immediately after screw insertion, ST reached 226 N, followed by AA with 191 N,SH with 137 N, and SC with 72 N. After 12 hours, ST displayed the highest residualcompression force, 141 N, followed by AA with 121 N, SH with 78 N, and SC with 32 N.The differences were significant for ST and AA compared to SC. The loss of compressionforce over 12 hours was 39% for ST, 42% for AA, 49% for SH, and 55% for SC.

Conclusions The new generation of headless compression screws, especially ST and AA,provided significantly higher compression forces after 12 hours, as well as the least loss ofcompression force over time, in comparison to a classic cortical lag screw.

Clinical relevance A new generation of headless compression screws, by producing highercompression forces, increase stability at the fracture site and might thereby promote bonehealing. (J Hand Surg 2012;37A:1142–1150. Copyright © 2012 by the American Society forSurgery of the Hand. All rights reserved.)

Key words Biomechanics, compression force, fracture, headless compression screws, scaphoid.

RIGID INTERNAL FIXATION IS preferred in all acuteunstable or displaced scaphoid fractures, frac-tures of the proximal pole, fractures with pre-

existing cystic bone formation, transscaphoid perilunatedislocations, untreated fractures more than 4 weeks old,or in cases in which a patient prefers to avoid long-term

From the Department of Trauma Surgery, Centre for Musculoskeletal Surgery, The University MedicalCenter Mainz, Mainz, Germany; Department of Orthopedic and Trauma Surgery, The University HospitalCologne, Cologne, Germany; Department of Diagnostic and Interventional Radiology, The UniversityMedical Center Mainz, Mainz, Germany.

Received for publication July 7, 2011; accepted in revised form March 8, 2012.

No benefits in any form have been received or will be received related directly or indirectly to the

subject of this article. h

1142 � © ASSH � Published by Elsevier, Inc. All rights reserved.

immobilization.1–3 The surgical methods available forthe treatment of acute scaphoid fractures and nonunionsare now more diverse than ever.

The Herbert screw, the first headless compressionscrew, has become a common implant for internal fix-ation. In biomechanical studies, the interfragmentary

The biomechanical laboratory of the Department of Trauma Surgery, University Medical CenterMainz, is supported by a yearly grant from Synthes, Switzerland. The 3.0-mm headless compressionscrews and 2.0-mm cortical screws were donated by Synthes.

Corresponding author: Dominik S. Gruszka, MD, The University Medical Center Mainz, Centre forMusculoskeletal Surgery, Department of Trauma Surgery, Langenbeckstraße 1, D-55131 Mainz, Ger-many; e-mail: dominik.gruszka@unimedizin-mainz.de.

363-5023/12/37A06-0005$36.00/0

0 ttp://dx.doi.org/10.1016/j.jhsa.2012.03.018

DURABILITY OF SCAPHOID SCREW FIXATION 1143

compression achieved with its pitch difference betweenthe proximal and distal threads is not as good as that ofconventional screw designs or new-generation headlesscompression screws from various manufacturers.4–7

The mean time to union after compression screwfixation of acute, nondisplaced, scaphoid waist fracturesis within 6 to 8 weeks, which is consistently less thanafter nonsurgical treatment (within 12–15 weeks).1,8,9

Moreover, a long period of immobilization in young,active men is not well tolerated and leads to socioeco-nomic loss due to missed work.10–12 Therefore, scaph-oid internal fixation is increasingly recommended incases of nondisplaced, stable fractures. Despite the in-troduction of newly developed fixation techniques, thenonunion rate of scaphoid fractures remains as high as10% after surgical treatment13 and from 5% to 23%after conservative treatment.14–15 Some of the mainproblems of internal fixation are the interfragmentarystability and durability of this stabilization. Secondaryloss of reduction and nonunion due to unstable fixationremain challenging problems, although additional fac-tors such as placement of the implant, the type andlocation of the fracture, and accuracy of the reductionare also involved.16,17

The purpose of this study was to evaluate the inter-fragmentary compression force generated across a sim-ulated fracture in unpreserved cadaveric scaphoids bymodern headless compression screws in comparisonwith the standard Synthes 2.0 cortical screw (Synthes,Solothurn, Switzerland) for 12 hours after insertion.

MATERIALS AND METHODS

Specimens

We obtained 20 pairs of fresh-frozen, cadaveric scaph-oids after receiving approval from the local ethics com-mittee. The average age of our specimens was 75 years(range, 60–90 y). After explantation, the scaphoidswere stripped of soft tissue and frozen at �24°C untilquantitative computed tomography measurement, in-strumentation, and testing.

Quantitative computed tomography scans were per-formed (Siemens Somatom, Munich, Germany). Cus-tom-designed software was used to draw 3-dimensionalbone models. The mean voxel value (Hounsfield units)of each bone was calculated without differentiation ofcortical and cancellous bone. The air entrapmentswithin the bone were not included in the calculation, asthey do not provide additional anchoring possibilitiesfor screws. The average voxel density in Hounsfieldunits was noted to have a high correlation to bonemineral density units, so that these values were ac-

cepted as sufficient for the purpose of our study.18 We

JHS �Vol A,

then assigned 40 bones to 10 bone density groups, eachconsisting of 4 bones of similar density (Fig. 1). After-ward, the sequence of groups was randomized. Eachbone from a group with similar bone density was alsorandomly assigned to treatment with one of the 4screws.

Implants

We chose 3 headless screws designed to increase com-pression of the fixation and a reference screw, as fol-lows:

● The Acumed Acutrak 2 Mini (AA; Acumed, Hills-boro, OR) has a conical screw design and contin-uous, variable thread. The wider thread pitch at thetip of the screw penetrates the bone faster than thefiner trailing thread, compressing the 2 fragmentsgradually as the screw is advanced.

● The Stryker TwinFix (ST; Stryker, Kalamazoo,MI) has proximal and distal screw threads workingindependently, which allows for in situ dynamicadjustable interfragmentary compression. Edges atthe screw body enable reaming at the core hole forthe trailing screw thread. This screw requires aspecial screwdriver with 2 concentric hexagonalheads and a clutch.

● The Synthes 3.0 headless compression screw (SH;Synthes, Solothurn, Switzerland) has a sleeve thatallows for compression and placement of the screwunder the cartilage surface with precision. Thisscrew offers no pitch difference between leadingand trailing ends. No additional compression isgenerated during countersinking. We used screwswith a long leading thread.

● The Synthes 2.0 cortical screw (SC; Synthes, So-lothurn, Switzerland), which has a conventionaldesign, is inserted as a lag screw (Table 1, Fig. 2).It was used as a reference screw.

Test set-up

Approximately 40 minutes before each experiment, onescaphoid was thawed and warmed in 0.9% NaCl to35°C. The manufacturer’s recommended K-wire (Table1) was introduced in a retrograde fashion. Centralplacement was performed according to Menapace et al,using a scaphoid aiming instrument (OGR 160; Heinz-Waldrich, Mülheim, Germany). The position of a K-wire was checked radiographically in 2 planes becausethe placement would influence the stability of the fix-ation19,20 (Fig. 3). We then drilled over the K-wire withthe manufacturer’s drill bit.

The osteotomy was done with the K-wire implanted

to improve orientation and to ensure that the saw cut

June

grou

1144 DURABILITY OF SCAPHOID SCREW FIXATION

was perpendicular to the long axis of the scaphoid.After the saw reached the K-wire, we removed theK-wire and completed the cut. The oscillating sawcreated a 1-mm kerf. In each case, a cut simulating themost common Herbert B2 scaphoid fracture was pro-

FIGURE 1: Scatter plot showing the assignment of bones intobones with similar bone density (Hounsfield units, HU) in each

TABLE 1. Biomechanical Properties of Tested Screw

Acumed Acutrak 2 Mini(AA) Stryker Twi

Cannulation/K-wire Yes/ø1.1 mm Yes/ø1.0 mm

Insertion technique Self-tapping; self-drilling,optionally

Self-tapping

Screwdriver 2.0-mm hex Double hex wclutch

Threads ø3.5–3.6 mm; conicalshape, continuous

ø3.2 mm; indproximal athreads

All screws except SC were cannulated. Each examined screw was self-diameter of the thread ranged from 2.0 mm to 3.6 mm. ø, diameter.

duced at the waist.21

JHS �Vol A,

A custom-manufactured load sensor was interposedbetween the 2 bone fragments. The load cell (PWFC-2MPA; Tokyo Sokki Kenkyujo, Japan), with a sensi-tivity of 3%, was placed on 1 side and a hole was drilledin the middle of the 2 metal profiles to allow for passage

roups according to the increasing bone density. There were 4p.

ST)Synthes 3.0 Headless

Compression Screw (SH)Synthes 2.0 Cortical

Screw (SC)

Yes/ø1.1 mm with a threaded tip No

Self-tapping; self-drilling,optionally; compression sleeve

Self-tapping

StarDrive T8 Cruciform

dentstal

ø3.0 mm; identical pitch ofthreads; short or long leadingthread options available

ø2.0 mm; continuous

g (the self-drilling option available for AA and SH was not used). The

10 g

s

nfix (

ith a

epennd di

tappin

of the screw. This construction used the principle of a

June

cortic

pos

DURABILITY OF SCAPHOID SCREW FIXATION 1145

long lever arm with a compression force put preciselyin the middle of this arm, and the result was measuredon the side opposite the pivot (Fig. 4). The recordedforce was double the measured force.

To prevent the rotation of bone fragments dur-ing screw introduction, we covered the polishedmetal contact surfaces of the load cell with sand-paper (Fig. 5).

Bones were temporarily held together on the loadsensor with a small clamp, which produced 8 to 10 N ofcompression force in each case (Fig. 5). The load cell

FIGURE 2: Screws used in the study. From left: Acumed Acompression screw (SH), and the reference screw, Synthes 2.0

FIGURE 3: Radiographic verification of a guide-wire

was 4 mm thick, so we needed to use longer screws

JHS �Vol A,

than those that would be used clinically. We subtracted4 mm from the measured system length to allow forsubchondral placement of the screw ends and to simu-late a clinical setting.

Next, we inserted the screw using the manufacturer’sscrewdriver and instructions. A single trauma surgeonperformed the experiments. The screws were advancedover the guide wire, through the distal scaphoid half andload cell, and into the proximal scaphoid. At this point,the holding clamp was removed. Screws were inserteduntil they were buried under the cartilage and no further

k 2 Mini (AA), Stryker TwinFix (ST), Synthes 3.0 headlessal screw (SC).

ition in 2 projections. A Volar view. B Lateral view.

cutra

gain in compression was noted. In the last phase, each

June

e mid

1146 DURABILITY OF SCAPHOID SCREW FIXATION

screw was carefully advanced incrementally. A 5-sec-ond pause after each partial rotation allowed for boneadaptation to the screw and for elimination of axialforces applied by the surgeon. The highest level ofstabilized force was interpreted as the maximum com-pressive force sustainable with a screw. After amplifi-cation (PICAS 4; Peekel Instruments, Rotterdam, TheNetherlands) load data were collected every second for12 hours and stored using DASYLab software (Mea-surement Computing, Norton, MA) on a personal com-puter. To prevent drying out, each bone was continu-ously immersed in physiological saline.

Statistical methods

Compression force data were assessed using the anal-ysis of variance and subsequent pairwise comparisons.We adjusted the compression force achieved to the

FIGURE 4: The custom-made load sensor from different viewbuilt in at 1 end and a hole for screw introduction exactly in th

FIGURE 5: Proximal and distal bone fragments fixed temporarilywith a clamp on the load sensor. Black arrows show the clamp.Yellow arrow shows the position of sandpaper. The sandpaperwas applied to both metal–bone interfaces.

bone density of each group. The Bonferroni correction

JHS �Vol A,

was applied to reduce the problems imposed by multi-ple tests. P � .05 was set as statistically significant.

RESULTSDuring the statistical testing, the influence of thebone density groups to the compression generatedby the screws was noted to be important (P � .07)but did not reach statistical significance. In con-trast, a statistically significant influence of differ-ent screws on the compression force generated wasobserved (P � .004).

Differences among screws could be noted directlyafter screw insertion. Screw ST reached 226 N, fol-lowed by AA with 191 N, SH with 137 N, and SC with72 N. Screws ST (P � .005) and AA (P � .04)achieved significantly higher compression forces com-pared with SC. Screw SH (P � .73) showed no differ-ence in comparison to the reference lag screw, AA (P �1.00), or ST (P � .22). There was no difference be-tween ST and AA (P � .99).

The mean compression force after 12 hours washighest for ST at 141 N, followed by AA at 121 N, SHwith 78 N, and SC with 32 N. (Table 2, Fig. 6). The Pvalue for the compression force after 12 hours in theanalysis of variance was .004, demonstrating significantdifferences among screws.

The compression forces after 12 hours for ST(P � .006) and AA (P � .03) were significantlydifferent from those observed for SC. Screw SH(P � .78) showed no significant difference fromthe reference SC, AA (P � .90), or ST (P � .24).The differences between ST and AA were notsignificant (P � .99).

The mean decay of compression force over 12 hourswas the lowest for ST with 39%, followed by AA with42%, SH with 49%, and SC with 55% (Fig. 7). Con-cerning the compression decay over time, there were nodifferences in comparison to the reference SC in be-

metal profiles connected loosely on both ends with a load celldle. A Side view. B Upper view. C Bottom view.

s: 2

tween-subjects effects (P � .06) or in pairwise com-

June

essio

DURABILITY OF SCAPHOID SCREW FIXATION 1147

parisons: ST (P � .1), AA (P � .25), SH (P � .99).There was no difference between ST and AA (P � .99).

No significant influence of the side (P � .31) or thesex (P � .15) of specimens on the compression forcesafter 12 hours could be observed.

DISCUSSIONScaphoid fractures fail to heal because of micro-motion

TABLE 2. Descriptive Statistics of Compression ForInsertion

Screw N_12 Min_12 Max_12

ST 141.02 30.57 297.69

AA 121.10 9.78 286.10

SH 77.82 8.04 203.80

SC 32.31 9.59 74.43

N_12, mean compression force remaining after 12 h; Min_12, minimforce remaining after 12 h; SD_12, standard deviation for compressionscrew introduction; Min_0, minimal compression force immediately ascrew insertion; SD_0, standard deviation for compression force imm

FIGURE 6: Distribution of the compr

at the fracture site because shearing results in fibrous

JHS �Vol A,

tissue formation.22 Stability determines the type of heal-ing that occurs. Strain is defined as the relative changein a fracture gap divided by the width of the fracturegap. When strain is kept to less than 2%, primary bonehealing occurs (endosteal healing). If the strain is keptin a range between 2% and 10%, secondary bone heal-ing occurs (endochondral ossification), in which newlyproduced vessels are torn and fibrous tissue formation is

N) After 12 Hours and Immediately After Screw

12 N_0 Min_0 Max_0 SD_0

3 226.09 47.95 399.79 112.91

0 190.79 14.15 358.55 116.59

6 136.71 25.78 294.23 98.16

5 71.73 17.47 143.91 40.46

pression force remaining after 12 h; Max_12, maximal compressionremaining after 12 h; N_0, mean compression force immediately after

crew insertion; Max_0, maximal compression force immediately afterly after insertion.

n force after 12 hours for each screw.

ce (

SD_

86.7

93.3

69.9

21.3

al comforcefter sediate

promoted. If strain is greater than 10%, bone healing

June

ell as

1148 DURABILITY OF SCAPHOID SCREW FIXATION

does not take place.17 Cadaveric biomechanical studiesdemonstrate that waist fractures of the scaphoid aresubjected to bending, rotational, and translational forcesthat together result in an enlarged strain at the fracturesite and the development of shearing forces that tearnew vessels.23–25 As a result of those complex stressesacting on the fracture, scaphoid internal fixation mustbe as rigid as possible to be successful.16 The optimalfixation device for scaphoid internal fixation should beable to withstand these complex forces during dailyfunctional loading. Because the scaphoid is mainly cov-ered with cartilage, fracture stabilization through callusformation is not possible. Rigid fixation should be pro-vided during this early phase of healing. Factors thatimprove bone healing, such as bone quality and fracturegeometry, cannot be controlled; it is possible to influ-ence only other factors such as the quality of fracturereduction, implant characteristics, and implant place-ment.

Among a number of studies investigating differentcompression screws in a simulated scaphoid fracture,the Herbert compression screw revealed lower com-pression forces and stability of internal fixation com-pared to lag screws. Newer compression screws arecapable of delivering similar or greater amounts ofinterfragmentary compression and stability than mini-fragment cortical screws.4–7,28–33

Lo et al performed 2 analyses of intrascaphoid com-pression that were similar to the experimental set-upreported here. They examined the intrascaphoid com-pression obtained using the Herbert screw compared tothe 3.0-mm Synthes cannulated screw with a threadedwasher. The Synthes screw proved to be a satisfactory

FIGURE 7: Diagram showing the relationships between thecompression observed at the beginning of the experiment, as w

alternative to the Herbert screw. Mean compression of

JHS �Vol A,

the 3.0 Synthes with threaded washer was 108 N,whereas the Herbert screw provided 20 N after therelease of the jig guide. In addition, the 3.0 Synthesscrew allowed the surgeon to build up compression in amore controlled manner. The old 3.0 Synthes screwused by Lo et al is functionally similar to the ST screwused in this study. The ST, with its free rotating headand use of a double hex screwdriver, also allows con-trolled increases in compression. This feature might beof great value when treating comminuted scaphoid frac-tures.31,32

The testing performed previously was not limited tocompression force. Several studies examined the stabil-ity of the internal fixation of scaphoids during cyclicalloading. These types of testing should simulate a clin-ical situation without cast treatment following internalfixation. Burkhart et al tested the stability of osteosyn-thesis, not the compressive force, of the 3.0-mm Syn-thes headless compression screw compared to the Syn-thes 2.0-mm cortical screw used as a lag screw inhuman cadaver radial heads. No statistical differencesbetween the 2 fixation methods could be detected.27

Wheeler et al noted greater fracture fragment stabil-ity and compression with the Acutrak Standard screwthan with the AO and Herbert screws. The Acutraksystem also provided greater resistance to torque thanthe AO and Herbert screws. This might increase rota-tional stability and the ability to maintain interfragmen-tary contact.33

The results of our 12-hour observations showed thatST and AA produced significantly more compressioninitially and after 12 hours than either SH or the refer-ence lag screw, SC. The ST built up compression force

pressive forces of the screws over 12 hours. The highestat the end, was achieved by the ST screw.

com

consistently, as reflected by the lower standard devia-

June

DURABILITY OF SCAPHOID SCREW FIXATION 1149

tion and the notably lower compression decay over 12hours. With such compression forces, the surgeonshould keep in mind that over-compression of a fracturewith comminution can also lead to an iatrogenic hump-back deformity.

The greater interfragmentary compression generatedby ST compared with AA might be due to differencesin the thread pitch, which determines the magnitude ofcompression generated by the screw. Screw SH, with-out a differential pitch but with a special sleeve forprecompression and precise subchondral placement ofthe head, showed the highest compressive force withthe sleeve connected. During disconnection from thesleeve, a great amount of the compressive force waslost. During countersinking, the force did not changesubstantially. The latter findings correlate with the ex-periences of Pensy et al, who tested SH in comparisonwith the Acutrak Standard screw in a similar setup. Themeasurement time was 5 minutes, and no significantdifference between tested screws was shown in meancompression immediately after screw insertion or com-pression 5 minutes after insertion.26 For statistical eval-uation, the value of the compression force of SH ob-tained with a disconnected sleeve was used in our study,as it is the most clinically relevant.

Our study has limitations. The average age of ourdonors was 75 years, much older than a typical patientwith a scaphoid fracture. As a result, the bone qualityand biomechanical properties of our specimens wereworse and could influence the results.

The measured bone density from a computed tomog-raphy scan did not differentiate cortical and trabecularbone volume. This could influence anchoring possibil-ities for a screw, although this knowledge is also notoffered to a surgeon before surgery.

Given the concept that greater interfragmentary com-pression is important for fracture stability and fracturehealing, the ST and AA screws, which have the highestinitial compression and the least reduction of compres-sion over time, would be expected to most reliablypromote fracture healing. Ease of insertion, implantsize, and surgeon experience with a specific screwmight override biomechanical performance and mightbe of great importance when finally selecting an im-plant.

REFERENCES1. Adolfsson L, Lindau T, Arner M. Acutrak screw fixation versus cast

immobilization for undisplaced scaphoid waist fractures. J HandSurg 2001;26B:192–195.

2. Slade JF III, Dodds SD. Minimally invasive management of scaph-oid nonunions. Clin Orthop Relat Res 2006;445:108–119.

3. McQueen MM, Gelbke MK, Wakefield A, Will EM, Gaebler C.

Percutaneous screw fixation versus conservative treatment for frac-

JHS �Vol A,

tures of the waist of the scaphoid: a prospective randomized study.J Bone Joint Surg 2008;90B:66–71.

4. Rankin G, Kuschner SH, Orlando C, McKellop H, Brien WW,Sherman R. A biomechanical evaluation of a cannulated compres-sive screw for use in fractures of the scaphoid. J Hand Surg 1991;16A:1002–1010.

5. Shaw JA. Biomechanical comparison of cannulated small bonescrews: a brief follow-up study. J Hand Surg 1991;16A:998–1001.

6. Bailey CA, Kuiper JH, Kelly CP. Biomechanical evaluation of a newcomposite bioresorbable screw. J Hand Surg 2006;31B:208–212.

7. Beadel GP, Ferreira L, Johnson JA, King GJ. Interfragmentarycompression across a simulated scaphoid fracture—analysis of 3screws. J Hand Surg 2004;29A:273–278.

8. Bond CD, Shin AY, McBride MT, Dao KD. Percutaneous screwfixation or cast immobilization for nondisplaced scaphoid fractures.J Bone Joint Surg 2001;83A:483–488.

9. Schädel-Höpfner M, Marent-Huber M, Gazyakan E, Tanzer K, WerberKD, Siebert HR. Acute non-displaced fractures of the scaphoid: earlierreturn to activities after operative treatment. A controlled multicentercohort study. Arch Orthop Trauma Surg 2010;130:1117–1127.

10. Arora R, Gschwentner M, Krappinger D, Lutz M, Blauth M, Gabl M.Fixation of nondisplaced scaphoid fractures: making treatment costeffective. Prospective controlled trial. Arch Orthop Trauma Surg2007;127:39–46.

11. Haisman JM, Rohde RS, Weiland AJ; American Academy of Or-thopaedic Surgeons. Acute fractures of the scaphoid. J Bone JointSurg 2006;88A:2750–2758.

12. Papp S. Carpal bone fractures. Orthop Clin North Am 2007;38:251–260. Review.

13. Kawamura K, Chung KC. Treatment of scaphoid fractures andnonunions. J Hand Surg 2008;33A:988–997.

14. Leslie IJ, Dickson RA. The fractured carpal scaphoid. Natural history andfactors influencing outcome. J Bone Joint Surg 1981;63B:225–230.

15. Dias JJ, Wildin CJ, Bhowal B, Thompson JR. Should acute scaphoidfractures be fixed? A randomized controlled trial. J Bone Joint Surg2005;87A:2160–2168.

16. Slade JF III, Gillon T. Retrospective review of 234 scaphoid frac-tures and nonunions treated with arthroscopy for union and compli-cations. Scand J Surg 2008;97:280–289.

17. Egol KA, Kubiak EN, Fulkerson E, Kummer FJ, Koval KJ. Biome-chanics of locked plates and screws. J Orthop Trauma 2004;18:488–493.

18. Todisco M, Trisi P. Bone mineral density and bone histomorphometry arestatistically related. Int J Oral Maxillofac Implants 2005;20:898–904.

19. Chan KW, McAdams TR. Central screw placement in percutaneousscrew scaphoid fixation: a cadaveric comparison of proximal anddistal techniques. J Hand Surg 2004;29A:74–79.

20. McCallister WV, Knight J, Kaliappan R, Trumble TE. Central place-ment of the screw in simulated fractures of the scaphoid waist: abiomechanical study. J Bone Joint Surg 2003;85A:72–77.

21. Herbert TJ, Fisher WE. Management of the fractured scaphoid usinga new bone screw. J Bone Joint Surg 1984;66B:114–123.

22. Augat P, Burger J, Schorlemmer S, Henke T, Peraus M, Claes L.Shear movement at the fracture site delays healing in a diaphysealfracture model. J Orthop Res 2003;21:1011–1017.

23. Kobayashi M, Garcia-Elias M, Nagy L, Ritt MJ, An KN, CooneyWP, Linscheid RL. Axial loading induces rotation of the proximalcarpal row bones around unique screw-displacement axes. J Bio-mech 1997;30:1165–1167.

24. Garcia-Elias M. Kinetic analysis of carpal stability during grip. HandClin 1997;13:151–158.

25. Smith DK, Cooney WP III, An KN, Linscheid RL, Chao EY. Theeffects of simulated unstable scaphoid fractures on carpal motion.J Hand Surg 1989;14A:283–291.

26. Pensy RA, Richards AM, Belkoff SM, Mentzer K, Andrew Eg-lseder W. Biomechanical comparison of two headless compres-sion screws for scaphoid fixation. J Surg Orthop Adv 2009;18:

182–188.

June

1150 DURABILITY OF SCAPHOID SCREW FIXATION

27. Burkhart KJ, Nowak TE, Appelmann P, Sternstein W, Rommens PM,Mueller LP. Screw fixation of radial head fractures: compression screwversus lag screw-a biomechanical comparison. Injury 2010;41:1015–1019.

28. Christodoulou LS, Kitsis CK, Chamberlain ST. Internal fixation of scaphoidnon-union: a comparative study of three methods. Injury 2001;32:625–630.

29. Faran KJ, Ichioka N, Trzeciak MA, Han S, Medige J, Moy OJ. Effectof bone quality on the forces generated by compression screws.J Biomech 1999;32:861–864.

30. Hausmann JT, Mayr W, Unger E, Benesch T, Vécsei V, Gäbler C.Interfragmentary compression forces of scaphoid screws in a saw-

bone cylinder model. Injury 2007;38:763–768.

JHS �Vol A,

31. Lo IK, King GJ, Milne AD, Johnson JA, Chess DC. A biomechanicalanalysis of intrascaphoid compression using the Herbert scaphoidscrew system. An in vitro cadaveric study. J Hand Surg 1998;23B:209–213.

32. Lo IK, King GJ, Patterson SD, Johnson JA, Chess DG. A biome-chanical analysis of intrascaphoid compression using the 3.00 mmSynthes cannulated screw and threaded washer: an in vitro cadavericstudy. J Hand Surg 2001;26B:22–24.

33. Wheeler DL, McLoughlin SW. Biomechanical assessment ofcompression screws. Clin Orthop Relat Res 1998;350:237–

245.

June