The Harvard Orthopaedic Journal

57 Pages·2013·2.01 MB·English

Volume 14 · December 2012
The Harvard
Orthopaedic
Journal
http://www.orthojournalhms.org
The Harvard Orthopaedic Journal http://www.orthojournalhms.org
Volume 14 · December 2012
Editorial Board
Editors-in-Chief Advisors
Sang Do Kim, MD Dempsey S. Spring(cid:976)ield, MD
Peter S. Vezeridis, MD Peter M. Waters, MD
Harry E. Rubash, MD
Associate Editors
Thomas S. Thornhill, MD
Terrill P. Julien, MD Mark C. Gebhardt, MD
Bryce T. Wolf, MD
Eric C. Fu, MD
Shawn G. Anthony, MD
Carl M. Harper, MD
Beverlie L. Ting, MD
2
The Harvard Orthopaedic Journal http://www.orthojournalhms.org
Volume 14 · December 2012
Original Articles
4 | Hindfoot Alignment in Surgical 34 | The “Almost Open” Calcaneus
Planning for Total Knee Fracture: Tips for Soft Tissue
Arthroplasty Management
Naven Duggal, Gabrielle Paci, Leandro Grimaldi John Y. Kwon, M.D.
Bournissaint, Abhinav Narain, Ara Nazarian
40 | An Update on Assessing the Validity
of the Lauge Hansen Classification
7 | Distal Interphalangeal and Thumb
Interphalangeal Joint Arthrodesis System for In-vivo Ankle
Fractures Using YouTube videos
with New Generation Small Headless,
of Accidentally Sustained Ankle
Variable Pitch Fixation Devices
Fractures as a Tool for the Dynamic
Christopher V. Cox, M.D., Brandon E. Earp, M.D.,
Philip E. Blazar, M.D. Assessment of Injury
Edward K Rodriguez M.D.,Ph.D, John Y. Kwon, M.D., Aron T.
12 | Time to Union as a Measure of Chacko, B.S., John J. Kadzielski, M.D., Lindsay Herder, B.S.,
Paul T. Appleton, M.D.
Effectiveness
Johan A.P.A.C. van Kollenburg, M.D., David Ring, M.D., Ph.D.
44 | Reengineering Operating Room
Workflows: Surgeon to Central
22 | Advances in Single-cell Tracking
Processing Staff Interventions
of Mesenchymal Stem Cells
Improve the Mechanics of Safe
(MSCs) During Musculoskeletal
Surgical Care Delivery and Decrease
Regeneration
Infection Rates
Joseph A. Phillips, Luke J. Mortensen, Juan P. Ruiz,
Rukmani Sridharan, Sriram Kumar, Marie Torres, Parul John Kadzielski, M.D., Angela Kelly, R.N., Brett MacTavish,
Sharma, Charles P. Lin, Ph.D., Jeffrey M. Karp,Ph.D., Peter S.T., Jaehon Kim, M.D., Paul Appleton, M.D., Ken
V. Hauschka, Ph.D. Rodriguez, M.D., Ph.D.
29 | Osteochondral Interpositional 50 | Inside the Value Revolution at
Allograft for Revision of a Children’s Hospital Boston:
Malunited Olecranon Fracture, Time-Driven Activity-Based Costing
Olecranon Malunion Revision: A in Orthopaedic Surgery
Case Report William P. Hennrikus, B.A., Peter M. Waters, M.D., Donald S.
Bae, M.D., Sohrab S. Virk, B.S., Apurva S. Shah, M.D., M.B.A
Sang Do Kim, M.D., Jesse B. Jupiter, M.D.
3
The Harvard Orthopaedic Journal http://www.orthojournalhms.org
Volume 14 · December 2012
Hindfoot Alignment in Surgical Planning
for Total Knee Arthroplasty
Naven Duggal2, Gabrielle Paci1*, Leandro Grimaldi Bournissaint1*, Abhinav Narain1, Ara Nazarian1

1Center for Advanced Orthopaedic Studies, Department of Orthopaedic Surgery, Beth Israel Deaconess Medical
Center, Harvard Medical School, Boston, MA
2Department of Orthopaedic Surgery, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA
*These authors have contributed equally.
N
umerous biomechanical factors, includ- of malalignment using MADC and MADG. We
ing malalignment of the lower limbs, are hypothesized that there would be a signi(cid:976)icant
associated with increased force across the difference between estimates using MADC and
joints leading to higher incidence and progres- those where MADG was used. We will now bring
sion of OA of the knee and ankle.1 Traditionally, our focus to the biomechanics laboratory where
this mechanical axis deviation has been measured we aim to compare load transmission applied
from the center of the femoral head to the center of at the femoral head using MADC and MADG on
the ankle and is called the conventional mechan- a cadaveric model. We hypothesized that MADG
ical axis deviation (MADC). However, numerous would prove a superior method for predicting
studies have indicated that a more accurate mea- actual weightbearing axis of the lower extremity.
surement of the actual weightbearing axis would
also account for hindfoot malalignment distal to
the ankle, including the subtalar joint.2-6 This axis Methods
can be measured from the center of the femoral Sample lower extremity x-rays combining
head to the ground reaction point and is called standard AP radiographs with hindfoot align-
the ground mechanical axis deviation (MADG). ment views were compared for estimated MADC
The relationship between knee OA and hind- and MADG. (Figure 1) Computer simulation free-
foot deformity has been examined in the litera- body diagrams of single leg stance, double leg
ture. Though no predictable relationship between stance, toe off and heel strike were drawn and
knee and hindfoot malalignment has been found, geometrically compared using MADC and MADG.
it is known that a signi(cid:976)icant number of patients Length of tibia (286.5 mm), femur (353.5 mm)
with knee OA will also have some degree of and the height of foot (70.6 mm) were derived
hindfoot deformity.6-7 The importance of pre- from anthropological data of an average adult
cise alignment for knee implant success cannot male.9 Guichet et al.’s predicted trigonometry
be underestimated, as even a minor deviation equation was coded in MATLAB and used to
can lead to increased edge loading, polyethylene derive MADC and MADG for each stance over
implant wear, early failure and subluxation.8 As a range of foot-tibial angles and genu valgum
such, a more accurate measurement for operative angles at the knee.3 By convention, valgus devi-
planning that accounts for alignment distal to the ations were considered positive and varus devi-
ankle, such as MADG, is desirable. We designed a ations were considered negative. A 3D graph
dynamic computer model to compare measures was plotted to illustrate the differences between
4
The Harvard Orthopaedic Journal http://www.orthojournalhms.org
Volume 14 · December 2012
FIGURE 1.
MADC and MADG for a given angle of deformity malalignments. Load transmissions at the knee,
in each stance. ankle and hinfoot will be measured to determine
In the laboratory, baseline measurements of the weightbearing axis in the setting of hindfoot
anatomic axis, MADC and MADG will be recorded deformity.
for 14 cadaveric lower extremity specimens. We
have designed and are building a biomechanical Results
testing jig to apply load to the specimens. The Using the computer model, the two evaluative
jig will be constructed using a hydraulic jack to methods, MADC and MADG, produced greatly
apply load to the femoral head at an angle cal- varying results. MADG signi(cid:976)icantly exceeded
culated to simulate physiologic loading. We will MADC values, which stresses the severity of
measure the load transmitted through the medi- the malalignment. Higher angles of hindfoot
al and lateral knee joint, the center of the ankle deformity were associated with greater MADG
joint and the ground reaction point of the calca- from anatomical axis. In future studies using
neus. Loads measured at the center of the ankle the cadaveric model, we expect to (cid:976)ind that load
and the ground reaction point will be compared measured at the ground reaction point is closer
to determine which more accurately estimate to the actual force applied when compared with
actual load transmission. Deformities at the level load measured at the center of the ankle. This
of the hindfoot will then be created by an ortho- difference between measured load transmission
paedic surgeon to simulate typical physiologic at the center of the ankle and the ground
5
The Harvard Orthopaedic Journal http://www.orthojournalhms.org
Volume 14 · December 2012
reaction point should be greater for each speci- of tibia, length of femur and genu-valgum angle,
men after hindfoot deformity has been simulat- MADG also considers height of foot, valgus angle
ed. Based on our computer model (cid:976)indings, we of foot and theta (angle between the line joining the
also expect that knees implanted using MADC to sole of the foot to the knee and the femur) in addi-
plan realignment will demonstrate more uneven tion to these conventional parameters.
loading at the medial and lateral knee joint when In conclusion, it is essential to accurately
compared to those implanted using MADG for evaluate limb mechanics. Planning for knee and
surgical planning. ankle surgery and arthroplasty requires the eval-
uation of the conventional mechanical axis align-
Conclusions ment as well as hindfoot malalignment. Precise
Th e results of this analysis so far illustrate that the and comprehensive evaluation will reduce the
incorporation of hindfoot deformity into calculation risks of postoperative malalignment, early failure
of mechanical axis deviation creates a more dynam- of osteotomies and increased wear of polyeth-
ic model. MADG results for each stance appeared ylene components in knee arthroplasty. Future
much wider than MADC, which suggests possible work will include application of our (cid:976)indings to a
errors due to the limited number of parameters used cadaveric model.
for MADC. Where MADC accounts for only length
References
1. Guilak F. Biomechanical factors in osteo- surgery. J Am Podiatr Med Assoc. 2005;95(1):2-12.
arthritis. Best Pract Res Clin Rheumatol. 6. Mullaji A, Shetty GM. Persistent hindfoot
2011;25(6):815-823. valgus causes lateral deviation of weightbearing
2. Chandler JT, Moskal JT. Evaluation of knee and axis after total knee arthroplasty. Clin Orthop
hindfoot alignment before and after total knee Relat Res.
arthroplasty. A prospective analysis. J Arthro- 7. Gross KD, Felson DT, Niu J, et al. Association of
plasty. 2004;19(2):211-216. (cid:976)lat feet with knee pain and cartilage damage in
3. Guichet J, Javed A, Russell J, Saleh M. Effect of older adults. Arthritis Care Res. 2011;63(7):937-
the foot on the mechanical alignment of the low- 944.
er limbs. Clin Orthop Relat Res. 2003;415:193- 8. Frigg A, Nigg B, Hinz L, Valderrabano V, Rus-
201. sell Iain. Clinical relevance of hindfoot alignment
4. Meding JB, Keating EM, Ritter MA, Faris PM, view in total ankle replacement. Foot Ankle Int.
Berend ME, Malinzak RA. The planovalgus foot: a 2010;31(10):871-879.
harbinger of failure of posterior cruciate-retain- 9. Ali, M. Human Bones: Longest Or Largest
ing total knee replacement. J Bone Joint Surg Am. Human Body Bones. Information Of The World.
2005;87:59-62. Web Site: <http://www.einfopedia.com/human-
5. Mendicino RW, Catanzariti AR, Reeves CL, bones-longest-or-largest-human-body-bones.
King GL. A systematic approach to evaluation of php>. January 26, 2010. Accessed May 2, 2011.
the rearfoot, ankle and leg in reconstructive
6
The Harvard Orthopaedic Journal http://www.orthojournalhms.org
Volume 14 · December 2012
Distal Interphalangeal and Thumb
Interphalangeal Joint Arthrodesis
with New Generation Small Headless,
Variable Pitch Fixation Devices
Christopher V. Cox, M.D., Brandon E. Earp, M.D., Philip E. Blazar, M.D.

Department of Hand and Upper Extremity Surgery, Brigham and Women’s Hospital, Boston, MA
D
istal interphalangeal (DIP) joint and nail injury and distal phalanx fracture, be more
thumb interphalangeal joint (IP) arthrod- technically forgiving, and permit a greater bone-
eses are well-accepted procedures for the to-bone contact area at the fusion site. We pres-
treatment of painful or unstable joints. Numer- ent a retrospective case series summarizing our
ous techniques for accomplishing fusion have experience with smaller, headless, variable pitch
been described in the literature, using methods implants for DIP and IP joint arthrodeses along
of (cid:976)ixation including Kirschner Wires (K-wires), with our technique and observed complications.
interosseous wiring,1 standard bone screws,2-4
bioabsorbable implants,5 plates,6 external (cid:976)ixa-
tors,7 and headless variable pitch screws such as Materials And Methods
Herbert (Zimmer, Warsaw, Indiana)8-12 or Acu- Patients were located by querying our bill-
trak (Acumed USA, Hillsboro, OR)13-16 screws. ing database for CPT codes 29860 or 29862.
An arthroscopic-assisted technique has been Between July 2007 and January 2012 there were
described as well.17 57 fusions in 36 consecutive patients treated
Implant size plays an important role in (cid:976)ixa- with arthrodesis of the DIP or thumb IP joint
tion of DIP joint arthrodeses, in light of the small with either the Acutrak Micro or Fusion (9 dig-
size of the distal phalanx, especially in the small its in 9 patients) or AcuTwist (48 digits in 28
(cid:976)inger. Wyrsch et al18 noted that the average dor- patients)*. Revision arthrodeses were excluded.
sopalmar diameter of the distal phalangeal neck Radiographic healing of the arthrodesis site
(3.55mm) was smaller than the diameter of the was de(cid:976)ined as bridging callus on two or more
lagging threads of the Herbert screw (3.90mm). cortices on plain radiographs. All procedures
In 10 of 15 male cadaveric specimens and 15 of were performed by one of two attending hand
15 female specimens, these threads penetrated surgeons within a tertiary referral academ-
either the volar or dorsal cortex. In those pene- ic practice in a metropolitan setting. Hospital
trating dorsally, this lead to apparent nail matrix charts were reviewed for clinical data and radio-
injury. graphs were evaluated for alignment and healing.
There are commercially available headless
variable pitch devices now available in smaller *Note: One patient had IF, MF, RF, SF arthrodeses
sizes than previous implants (Table 1). These with Acutwist devices and a T arthrodesis with
devices should theoretically decrease the risks of an Acutrak Fusion device.
7
The Harvard Orthopaedic Journal http://www.orthojournalhms.org
Volume 14 · December 2012
TABLE 1. Selected Commercially Available Cannulated Headless Screws
Leading Thread Trailing Thread
Implant
Diameter (mm) Diameter (mm)
AcuTwist 1.5 2.0
Acutrak Micro 2.5 2.8
SBI AutoFix 2.0 3.0
Synthes 2.4mm Cannulated Headless Screw 2.4 3.1
Zimmer Herbert Mini 2.5 3.2
Acutrak Mini 2.8 3.2
Synthes 3.0mm Cannulated Headless Screw 3.0 3.5
Zimmer Herbert Screw 3.0 3.9
Surgical Technique lateral mini-C arm fl uoroscopy. Th e skin was incised
Th e technique for Acutrak micro screws is simi- at the tip to a 2mm opening. Th e length was then
lar to the technique described by Brutus et al.13 Th e measured, either with a supplied depth gauge or with
technique for the Acutwist is described below. a second guide wire and ruler. Next, while holding
Patients were positioned supine utilizing a hand the reduction, the wire was removed and the tract
table. Pre-operative antibiotics were administered. A tapped (when necessary); in our series tapping was
transverse incision was made at the level of the DIP used only when the surgeon felt the bone was partic-
joint. Th is was carried down sharply through skin ularly dense. Th e appropriate length Acutwist device
and extensor tendon to the bone. Th e fl ap was not was inserted inserted taking care to maintain the
undermined distally so as to protect the germinal reduction of the arthrodesis site to allow the screw
matrix. Aft er exposing the DIP joint any remain- to follow the proper wire tract. Once seated to the
ing cartilage was curetted out and osteophytes were desired depth, the implant placement and clinical
removed with a rongeur. Th e bone was contoured alignment were again confi rmed. Th e device was
at this point, if necessary, to correct any coronal or then toggled in the anteroposterior and mediolateral
sagittal plane deformities, but the overall shape of planes while securing the arthrodesis site. Th e shaft
the two opposing surfaces was maintained except for of the device then would break off from the screw at
correcting angular deformity and exposing deep to the machined “snap-off groove”. Final fl uoroscopic
the subchondral bone. A small K-wire was used to images were then taken. Bone graft ing was used at
penetrate the subchondral surfaces of the surface of either at this point or prior to the fi nal placement
the distal phalanx in areas of dense sclerotic bone. of the implant depending on surgeon preferences.
Th en, a 0.045-inch diameter double tipped wire Wounds were then irrigated and typically closed
was advanced in an antegrade fashion through the with 5-0 or 6-0 nylon sutures. Soft bandages and a
fl exed distal phalanx, exiting through the tip of the fi nger cap splint were placed, leaving the PIP joint
fi nger in the midline, just volar to the nail plate. Th is completely free.
was then advanced until the tip was just proximal Sutures were removed at 10-14 days post-op-
to the surface of the distal phalanx. Th e fi nger was eratively. Hand therapy was not typically deemed
then reduced to a position of neutral coronal plane necessary, unless required for any concomitantly
alignment and 0-10 degrees of fl exion, and the wire performed procedures. Patients were followed with
was advanced proximally into the middle phalanx. interval clinical visits and radiographs until bony
Positioning was confi rmed on anteroposterior and and clinical union occurred.
8
The Harvard Orthopaedic Journal http://www.orthojournalhms.org
Volume 14 · December 2012
Results Th ere was one case of a deep infection occurring
Th ere were 7 males and 29 females. Average prior to bony union. Th is required implant removal.
age was 58.3 years (range 33-84) at the time of sur- Th e patient was left with a fl ail joint, but was pain
gery. Average duration of follow up was 321 days. 2 free in an orthosis and declined further operative
patients were lost to follow up at a time period before intervention. Th e remaining non-union occurred in
radiographic union would have been expected (0 a patient with lupus who underwent arthrodesis of
days and 35 days). Th e primary diagnosis was osteo- the thumb, index, and long fi ngers. Th e thumb and
arthritis in 23, trauma in 4, Lupus in 3, Mallet/Bou- long fi gers healed uneventfully; the index did not. No
tonniere deformity in 3, and there was one case each further operative intervention has been performed,
of Dupuytren’s, post infectious arthritis, and neuro- although she does report discomfort at this site.
muscular disorder. Th ere were 7 thumbs, 17 index Our major complication (nonunion, deep infec-
fi ngers, 12 long fi ngers, 9 ring fi ngers, and 12 small tion) rate was 10.5% and our minor complication
fi ngers included. 21 patients (58%) had other associ- (intraoperative fracture, symptomatic hardware)
ated procedures performed concomitantly. rate was 3.5%.
Th ere were no cases of nail deformity, signifi cant
skin sloughing, or clinically signifi cant malalign- Conclusions
ment. Th ere were no cases of implant breakage Arthrodesis of the DIP/IP joints is proven and
intra-operatively at another site than the planned eff ective for dealing with a myriad of painful and
site. Th ere was one case of prominent hardware at deforming ailments of the DIP and IP joints. In this
the volar pulp requiring hardware removal following setting, headless variable pitch screws have many the-
union. Th is patient was asymptomatic at the most oretical benefi ts compared to other potential fi xation
recent follow up. Th ere was one intraoperative distal methods. Unlike K-wires they are buried deeply and
phalangeal fracture that occurred in the small fi nger avoid having a potential conduit for deep infection.
of a patient with lupus. Th is was noted on fi nal fl u- Th is may explain the low instance of either deep or
oscopic imaging; however, the arthrodesis site was superfi cial infections seen in this series. Unlike stan-
noted to be stable. Th e fracture healed uneventfully dard bone screws, they are completely intraosseous
and the arthrodesis site went on to union. and avoid having a prominent screw head situated in
Radiographic union was noted in 50 of 55 fi n- the sensitive volar pulp region. Th is may account for
gers (91%). [2 fi ngers in 2 patients were lost to follow the lack of complaints of tip sensitivity and the limit-
up]. Local autograft (typically from the dorsal osteo- ed need for hardware removal in our series.
phytes) was used in 27 of 57 digits. In 2 cases, bone Our results compare favorably to prior reported
graft from a distant site (e.g. distal radius) was used. series. In 1992, Stern and Fulton12 published a series
Th ere were fi ve non-unions. One was in an of 181 arthrodeses of DIP and IP joints. Th eir major
osteoarthritic patient who underwent 3 simultane- complication rate was 20% (infections, non-unions,
ous DIP/IP arthrodeses, all with AcuTwist devices, etc) and their minor complication rate was an addi-
which resulted in loss of fi xation of the thumb IP tional 16% (skin necrosis, prominent hardware,
arthrodesis site around 6 months post-operatively. paresthesias, etc). Th ey reported non-unions in 21
Th is was treated with revision to an Acutrak Fusion (12%), however, 13 of these were pain free. A variety
screw with distal radius autograft and a supple- of techniques were employed.
mentary 26-gauge interosseous wire, progressing to Several case series have documented usage of
union at 4 months aft er the revision surgery. Anoth- headless variable pitch screws, which have the the-
er patient underwent ring fi nger DIP joint arthrod- oretical benefi t of being completely intraosseous
esis for post-traumatic arthritis did not demonstrate to avoid hardware prominence while providing
radiographic union at a 7 month follow up visit, but inter-fragmentary compression. Faithfull and Her-
he was asymptomatic at that time. bert9 noted 100% union and no complications in 11
9
The Harvard Orthopaedic Journal http://www.orthojournalhms.org
Volume 14 · December 2012
DIP joints in their early series using Herbert screws. worth noting. Its retrospective nature makes it diffi -
In Stern’s12 subgroup of Herbert screws, a major cult to make direct comparisons with other studies.
complication rate of 19% and minor complication Our radiographs were obtained at non-standardized
rate of 44% was documented in 27 cases. More intervals, thus making a determination of time to
recent case series have documented variable results. healing unreliable. Our patients had a broad array
El-Hadidi and Al-Kdah8 documented fusion in 14 of of diagnoses, which limits the ability to elucidate dif-
15 digits. Th ey had one case of poor screw placement ferent subgroup characteristics. We also were unable,
causing pain. Lamas Gomez et al11 had fusion in 19 given the low complication rate, to determine the
of 20 digits with one case of amputation related to relative complication rates for DIP/IP arthrodeses
dorsal skin necrosis. Th ey recommended using the for these diff ering diagnoses. One of the nonunion
mini-Herbert screw to facilitate placement. Brutus cases was in a thumb IP joint and this patient went
et al13 utilized mini-Acutrak screws and noted non- on to heal with a larger diameter implant. As the
unions in 3/22 (14%), infection in 4/22 (18%), and distal phalanx of the thumb is typically signifi cantly
nail bed injury in 3/22 (14%). Th ey noted the diffi - larger then the other digits, larger implants may be
culty of using the mini-Acutrak screws, especially in preferable. Th e authors have switched to using larger
the small fi nger. diameter implants for arthrodesis of the thumb IP
Th e smaller diameter of these devices is more joint. We were unable to determine the role or eff ect
appropriate for the tight confi nes of the distal pha- of autogenous bone graft ing.
langeal medullary canal. Perhaps due to this sizing, Reliable fusion rates were achieved with a mod-
we had no instances of nail plate deformities due to est complication rate. Insertion of these implants is
penetration of the dorsal cortex of the distal phalanx perhaps more technically forgiving than with prior
as seen in the biomechanical study by Wyrsch et al.18 generations of larger implants. Th ese devices seem to
While our case series is larger than any other be an improvement over prior generations of head-
series utilizing headless variable pitch screws for less variable pitch screws.
DIP/IP arthrodeses, there are several limitations
References
1. Zavitsanos, G., et al., Distal Interphalangeal phalangeal joint using a bioabsorbable rod as an
Joint Arthrodesis Using Intramedullary and intramedullary nail. Scand J Plast Reconstr Surg
Interosseous Fixation. Hand Surg, 1999. 4(1): p. Hand Surg, 2003. 37(4): p. 228-31.
51-55. 6. Mantovani, G., et al., Alternative to the distal
2. Leibovic, S.J., Internal (cid:976)ixation for small joint interphalangeal joint arthrodesis: lateral ap-
arthrodeses in the hand. The interphalangeal proach and plate (cid:976)ixation. J Hand Surg Am, 2008.
joints. Hand Clin, 1997. 13(4): p. 601-13. 33(1): p. 31-4.
3. Olivier, L.C., et al., Arthrodesis of the distal 7. Seitz, W.H., Jr., et al., Compression arthrodesis
interphalangeal joint: description of a new of the small joints of the hand. Clin Orthop Relat
technique and clinical follow-up at 2 years. Arch Res, 1994(304): p. 116-21.
Orthop Trauma Surg, 2008. 128(3): p. 307-11. 8. El-Hadidi, S. and H. Al-Kdah, Distal interpha-
4. Teoh, L.C., S.J. Yeo, and I. Singh, Interphalan- langeal joint arthrodesis with Herbert screw.
geal joint arthrodesis with oblique placement of Hand Surg, 2003. 8(1): p. 21-4.
an AO lag screw. J Hand Surg Br, 1994. 19(2): p. 9. Faithfull, D.K. and T.J. Herbert, Small joint fu-
208-11. sions of the hand using the Herbert Bone Screw.
5. Arata, J., et al., Arthrodesis of the distal inter- J Hand Surg Br, 1984. 9(2): p. 167-8.
10