The Journal of

White HAND SURGERY

8. Ogden J: In Skeletal injury in the child. Philadelphia,
1982, Lea & Febiger, p 365
9. Parikh M, Nahigian S, Froimson A: Gamekeeper's
thumb. Plast Reconstr Surg 58:24-31, 1976
10. Smith MA: The mechanism of acute ulnar instability of
the metacarpophalangeal joint of the thumb. Hand
12:225-30, 1980
II. Smith RJ: Post-traumatic instability of the metacarpo-
phalangeal joint of the thumb. J Bone Joint Surg [Am]
59:14-21, 1977
12. Stener B: Displacement of the ruptured ulnar collateral
ligament of the metacarpophalangeal joint at the thumb:
A clinical and anatomical study. J Bone Joint Surg [Br]
44:869-79, 1962

The stability of internal fixation in the proximal

phalanx
A biomechanical study was performed to assess quantitative differences in the stability obtained
with five commonly used types of internal fixation used in proximal phalangeal fractures. The
techniques included dorsal plating, dorsal plating combined with an interfragmentary lag screw,
two interfragmentary lag screws, tension band technique, and crossed Kirschner wires. Rigidity
and strength in apex palmar bending were determined after oblique osteotomy and fixation of
the proximal phalanx. The failure modes for each fixation technique were also observed and
described. The results showed that both of the techniques that used interfragmentary lag screws
across the oblique osteotomies provided significantly more rigidity than did dorsal plating alone
or the wired configurations but that measurements of strength were similar between all techniques
tested. Dorsal plates were at a mechanical disadvantage on the compression surface in our apex
palmar bend test and consequently provided no more rigidity and strength than did the wired
techniques. The tension band technique represented a combination of stiff and flexible intraos
seous wires without strict application of tension band principles and provided intermediate
rigidity and strength. Rigidity and strength in intact proximal phalanges in the controls were
significantly greater than in all techniques tested. (J HAND SURG llA:6727, 1986.)

David M. Black, M.D., Ronald J. Mann, M.D., Ronald M. Constine, M.D., and
A. U. Daniels, Ph.D. Salt Lake City, Utah

Fractures of the proximal phalanx are
common and usually amenable to closed treatment. The
basic goals of treatment are union in satisfactory align-
ment, with special emphasis on avoiding loss of func-
tion. Unstable fracture patterns may lend themselves to
rigid internal fixation, especially when early joint mo-
From the Division of Orthopaedics, University of Utah School of
Medicine, Salt Lake City, Utah.
Supported in part by the American Society for Surgery of the Hand
seed grant No. 84-3.
Received for publication Feb. 22, 1985; accepted in revised form
Nov. 5, 1985.
Reprint requests: Ronald J. Mann, M.D., Division of Orthopedics,
University of Utah School of Medicine, 50 N. Medical Dr., Salt
Lake City, UT 84132.
tion is desired after surgery.I.S Although various internal
fixation methods have been advocated, only a few stud-
ies have dealt with quantitative measurement of the
relative stability achieved. Previous investigators have
used transverse osteotomies, and no data exist on
oblique osteotomies in the hand, which may model
certain types of fractures better than transverse osteoto-
mies. 6 l4 Our interest was in studying fixation in the
proximal phalanx after oblique osteotomy, which al-
lows for the application of interfragmentary lag screws.
In fractures of the proximal phalanx, the displacement
pattern is typically apex palmar angulation, combined
with a variable degree of shortening and rotation (Fig.
1). The proximal fragment is flexed by the bony in-
sertion of the interossei muscles into the base of the

672 THE JOURNAL OF HAND SURGERY

Vol. llA, No.5
September 1986
proximal fragment while the intrinsic muscles pull on
the distal fragment through the extensor expansion.
In 1979 Fyfe and Mason8 described the results of
rigidity and strength tests in a single mode, apex palmar
cantilever bending, after transverse osteotomy and fix-
ation of embalmed human proximal phalanges. These
authors concluded that both crossed Kirschner wires
and an intraosseous wire combined with a Kirschner
wire are rigid enough fixation in proximal phalanges to
withstand the anticipated' 'flexor forces." Plate fixation
was considered inferior because it seemed to provide
less rigidity and required more extensive surgical ex-
posure. The lateral plate performed somewhat better
than did the dorsal plate. A single intramedullary
Kirschner wire provided the least stability of all meth-
ods compared. 8. 10
Massengill et al. II published a study on proximal
phalangeal fixation in 1982. They used a fresh pig meta-
carpal model and tested in a single mode, apex palmar
four-point bending, after transverse osteotomy and in-
ternal fixation. Palmar and lateral plates were found to
afford the greatest rigidity and strength and were com-
parable with intact controls. Dorsal plates were not
included in the study. Crossed Kirschner wires, an in-
traosseous wire combined with a Kirschner wire, and
an intraosseous wire alone provided markedly less sta-
bility than did plates and were statistically equivalent
to each other. Kirschner wires were found to lose fix-
ation by slipping and twisting within their bony chan-
nels. Intraosseous wiring was found to fail either by
cutting the bone or untwisting of the wire. II
Gould et al. 9 published their results on proximal pha-
langeal fixation in 1984. They used a fresh pig meta-
carpal model and transverse osteotomy similar to that
of Massengill et al. However, testing was done in the
opposite mode of apex dorsal four-point bend, which
would seem more applicable to metacarpals. Tension
band fixation was at a mechanical advantage and pro-
vided superior rigidity and strength comparable with
techniques that used multiple Kirschner or intraosseous
wires while allowing the least deformation and fracture
separation. Crossed Kirschner wires provided the poor-
est rigidity and strength, but techniques with additional
Kirschner or intraosseous wires augmented the stability
significantly.9 A subsequent study by Rayhack et al. 13
in 1984 used strain-gauge analysis of a transverse os-
teotomy in a chicken femur model in an apex dorsal
cantilever bend test. Wire loops placed at an advantage
on the dorsal tension surface provided more rigid fix-
ation than did wires in the neutral axis and the addition
of a stiff Kirschner wire supplemented the rigidity. 13
Stability of internal fixation in proximal phalanx 673
Apex Dorsal Angulation
PROX. PHALANX FRACTURE:
Apex Palmar Angulation
Fig. 1. The typical displacement patterns of metacarpals and
proximal phalanges after fracture.
Black et al. 7 published results on the stability of in-
ternal fixation techniques in metacarpals. Extrapolating
the results obtained in apex palmar bending to proximal
phalanges suggests that there is no difference statisti-
cally in the rigidity provided by dorsal plating and the
wired techniques tested. However, the addition of an
interfragmentary lag screw added significant rigidity to
the dorsal plate. 7 Yanik et al. 14 have also published
results on biomechanical testing of internal fixation in
metacarpals.
Material and method
The magnitude and direction of forces borne by the
proximal phalanx in normal activities are unknown; thus
it was considered important to study intact bones to
establish their normal strengths and rigidities as a stan-
dard of comparison. Preserved human proximal pha-
langes from the index, long, and ring fingers were cho-
sen for testing. Embalming has been found to alter
mechanical characteristics of bone minimally as long
as specimens are kept moist. 15 The bones were tested
mechanically after 45 oblique osteotomies were per-
formed and internal fixation was applied to nine prox-
imal phalanges in each of five fixation groups and the
intact controls. Five commonly used types of internal
fixation were chosen for analysis: dorsal miniplating,
dorsal miniplates combined with an interfragmentary
lag screw, two interfragmentary lag screws, tension
band technique, and crossed Kirschner wires (Fig. 2).
For dorsal plate fixation, an ASIF five-hole miniplate
was selected and mounted on the dorsal aspect of the
proximal phalanx with four 2 mm cortical screws and
located the central empty hole over the osteotomy site.
When an interfragmentary lag screw was combined with
a dorsal miniplate, the lag screw was passed from side

674 Black et al.
Fig. 2. The five fixation techniques and controls studied,
including intact proximal phalanges, dorsal plate, dorsal plate
combined with an interfragmentary lag screw, two innter-
fragmentary lag screws, tension band, and crossed Kirschner
wires .
to side across the osteotomy and outside of the plate.
Fixation with two interfragmentary lag screws used 2
mm cortial screws placed in parallel. Their orientation
was from side to side in the midlateral plane at an angle
halfway between a perpendicular to the shaft and the
osteotomy site. The modified tension band technique
involved two parallel anteroposterior 1 mm Kirschner
wires and a figure of eight 0.8 mm flexible wire on the
dorsal surface secured on the protruding pins on one
side of the osteotomy and through a transverse bony
channel on the other side. This configuration was not
intended to apply tension band principles since the wire
was on the compression surface in our testing mode,
but instead represented a technique combining multiple
stiff and flexible intraosseous wires. Fixation with
crossed Kirschner wire used two I mm Kirschner wires
The Journal of
HAND SURGERY
Fig. 3. Apex palmar three-point bend.
applied from side to side in the midlateral plane across
the oblique osteotomy.
As previously mentioned, unstable fractures of the
proximal phalanx tend to displace into apex palmar
angulation, combined with a variable degree of rotation
and shortening. Internal fixation in these bones must
resist the forces creating these deformities; thus testing
in apex palmar bending was considered most important.
Testing was accomplished with a standard screw-driven
and load cell-equipped mechanical testing machine.
The testing mode was an apex palmar three-point bend
with a span of 30 mm, and the third point was centered
over the osteotomy site (Fig. 3) . Crosshead speed was
kept constant at 0.5 mm/min. All specimens were taken
to failure with recording of the load-deformation be-
havior from which the structural rigidity, yield point,
and maximum bending moment were calculated. Visual
observation of the apparent failure mode for each type
of fixation was also made. Specimens were kept moist-
ened with saline solution during storage and testing.
Results
Data on rigidity in an apex palmar three-point bend
are presented in Fig. 4. For each of the five fixation
techniques and controls, nine specimens were tested,
and Fig. 4 shows the mean values and one standard
deviation about the mean . The formula used to calculate
Vol. IIA , No.5
September 1986 Stability of internal fixation in proximal phalanx 675

70
.40

-;;; 60
N
G;
.,
~

Ai
;; ::;:
:::' .30 g 5.0
c
o i.,
i., z
z
>- ~ 4.0
t:::
o
w
::;:
c; .20 o
::;:
a::
~ 3.0
Cl
z
oz
oz
w
w II)
II)
::;: 2.0
:::>
~ . 10 ~
::;: x
a:
...J <t
::;: 1.0
x
w
Cl.
<t
DORSAL DORSAL PLATE
PLATE +INTERfRAG .
LAG SCREW

Fig. 4. Bending rigidity of fixation techniques with mean and
range of one standard deviation about the mean depicted.
bending rigidity was EI = UF/48Y where L = span
between supports, F = force measured at the center
point, and Y = bending deflection measured. Both of
the techniques that used interfragmentary lag screws
across the oblique osteotomies provided significantly
more rigidity than did dorsal plating alone or crossed
Kirschner wires (p < 0.001) . A dorsal plate combined
with an interfragmentary lag screw was more rigid than
were two interfragmentary lag screws (p < 0.01). Dor-
sal plates alone provided rigidity comparable with
crossed Kirschner wires while the tension band tech-
nique resulted in intermediate rigidity. Rigidity of
the intact proximal phalanges in the controls was sig-
nificantly greater than all fixation techniques tested
(p < 0.001).
Maximum bending moments were obtained for each
proximal phalanx during testing to failure in an apex
palmar three-point bend and are shown in Fig. 5. The
equation used was M = FLl4 where F = force mea-
sured at the center point and L = the span between
supports. Results for all fixation techniques were sta-
tistically similar, but strength of the intact proximal
phalanges in the control group was significantly greater
(p < 0.00 I). A dorsal plate alone was similar to a plate
bending test and the standard deviation was small, al-
lowing for a statistical difference in strength between
Fig. 5. Maximum bending moments of fixation techniques.
Normal range for control intact proximal phalanges of two
standard deviations is represented by area between hatched
lines .
a dorsal plate alone and a dorsal plate combined with
an interfragmentary lag screw, two interfragmentary lag
screws, or the tension band technique to be calculated
(p < 0.01).
Yield moment, energy absorbed to yield, and fracture
site angulation at yield may also be relevant in com-
paring stability for each technique, but their relative
importance is difficult to assess (Table I). Yield moment
was calculated in a manner similar to that of maximum
bending moment after determination of the yield point.
Energy absorbed to yield was derived from the area
under the load-deformation curve to yield, and fracture
site angulation at yield was derived trigonometrically
assuming that all angulation occurred at the osteotomy
site. No statistical differences in moments at yield be-
tween fixation techniques were found. Fracture site an-
gulation allowed at yield was the least with the tech-
niques that used interfragmentary lag screws, resulting
in less calculated energy absorbed to yield for these
techniques.
Failure modes for each configuration were observed
visually and documented photographically during test-
ing to failure in apex palmar three-point bending. Prox-
imal phalanges fixed with a dorsal plate alone failed by

676 Black et al.
Table I. Calculated moment at yield, energy absorbed to yield, and fracture site angulation at yield of fixation
techniques
Moment at yield
(Nn'S)
Intact proximal phalanges 2.583 1.218
Dorsal plate 0.144 0.035
Dorsal plate and interfragmen- 0.260 0.116
tary lag screw
Two interfragmentary lag screws 0.141 0.052
Tension band 0.182 0.079
Crossed Kirschner wires 0.123 0.034
simple bending of the plate with no contribution from
the bone. Dorsal plates combined with an interfrag-
mentary lag screw failed in the cortical purchase of the
lag screw, which then allowed bending of the plate.
Metaphyseal or epiphyseal fractures did not occur in
any of the plated specimens. Fixation with two inter-
fragmentary lag screws failed in the cortical purchase
of one screw, allowing rotation about the other screw.
Techniques that used interfragmentary lag screws al-
lowed minimal angulation at the fracture site at failure.
The tension band technique failed by sliding of the
Kirschner wires in their bony channels with the flexible
figure of eight wire at a mechanical disadvantage on
the compression surface. No untwisting of the flexible
wires occurred. Crossed Kirschner wires failed by slid-
ing and twisting of the stiff wires in their bony channels.
Failure occurred through sliding and not bending of the
Kirschner wires.
Discussion
We compared the stability provided by five com-
monly used fixation techniques in proximal phalanges
after oblique osteotomy. Standard mechanical tests
were performed to determine apex palmar bending ri-
gidity, yield and maximum bending moment, and en-
ergy absorbed and angulation at yield. No standard
formula exists for weighing each of these factors in
arriving at an overall stability; thus it is best to judge
each separately. Values for intact proximal phalanges
were also studied to establish a normal range. If early
joint motion is desired, restoration of stability com-
parable with intact bones would seem desirable. Our
study used an oblique osteotomy that allowed for ap-
plication of interfragmentary lag screws, whereas all
previously published studies on internal fixation in the
hand have used a transverse osteotomy. Significant dif-
ferences in the stability provided by the various fixation
techniques evaluated were found. It should be recog-
The Journal of
HAND SURGERY
Energy absorbed to yield Fracture site angulation
(joules) at yield (degrees)
0.1015 0.0630 4.29 0.81
0.0093 0.0035 7.33 2.26
0.0040 0.0029 1.59 0.67
0.0021 0.0012 1.59 0.54
0.0081 0.0067 4.33 2.15
0.0100 0.0055 9.48 4.43
nized that the fracture configuration often dictates the
technique used and that the optimum rigidity and
strength for fracture healing are unknown. 2 -5 . 10 Bio-
mechanical analysis also does not assess factors such
as the extent of soft tissue dissection required or the
need for subsequent hardware removal.
Our results showed that both techniques that used
interfragmentary lag screws across the oblique osteo-
tomies provided significantly more rigidity than did dor-
sal plating alone or the wired configurations. Results
from the strength tests including maximum bending
moment and moment at yield were not statistically dif-
ferent among the fixation techniques tested. Dorsal
plates alone were at a mechanical disadvantage on the
compression surface in our apex palmar bend test and
therefore only provided rigidity and strength compa-
rable with the wired techniques. The tension band tech-
nique in our phalangeal model and testing mode rep-
resented a combination of stiff and flexible intraosseous
wires, and tension band principles were not strictly
applied. The use of the tension band technique resulted
in intermediate rigidity and strength. Similar results
might be expected with other techniques not tested that
involve multiple wires or combinations of stiff and flex-
ible intraosseous wires. 9 . 11-13.16 Rigidity and strength in
an equal number of intact proximal phalanges in the
control group were significantly greater than that found
in all fixation techniques tested. Fracture site angulation
allowed at yield was the least with the techniques that
used an interfragmentary lag screw, resulting in less
calculated energy absorbed to yield for these tech-
niques. Failure modes were also observed and docu-
mented for all techniques tested and differed signifi-
cantly.
Fyfe and Mason's8 work on fixation in proximal pha-
langes has been widely quoted in the field of hand
surgery. Their choice of an apex palmar cantilever bend
test is appealing, but rigid and reproducible fixation of
Vol. IIA, No.5
September 1986
the specimen at its base is difficult in our experience.
Statistical analysis of their results was incomplete. Al-
though they conclude that crossed Kirschner wires and
an intraosseous wire combined with a Kirschner wire
provided more rigidity than did dorsal or lateral plating,
review of the data provided in their article suggests that
the test results may not have been significantly different
among any of the techniques evaluated except for a
single intramedullary Kirschner wire that provided the
least stability. 8. 10 Our present study of proximal pha-
langes used an oblique osteotomy instead of the trans-
verse osteotomy used by Fyfe and Mason, which makes
comparison difficult. However, crossed Kirschner wires
and dorsal plating techniques again showed similar ri-
gidity and strength.
The publication by Massengill et al. II represents the
most extensive work to date on proximal phalangeal
fixation. Mechanical analysis of various fixation tech-
niques in an apex palmar four-point bend was accom-
plished and failure modes were recorded. Their plated
techniques performed well compared with various
wired configurations, probably because they were at a
mechanical advantage on the palmar and lateral sur-
faces. Unfortunately, they did not study dorsal plating,
which we believe to be the most practical plate location
on phalanges and metacarpals. They attributed their
good results with plates and Fyfe and Mason's poorer
results with plates to a possible calculation error in the
previous work. They did not mention the important
difference that they had used palmar plates, whereas
Fyfe and Mason had used dorsal plates. Comparison
with our present study of proximal phalanges is difficult
because we used an oblique osteotomy instead of a
transverse osteotomy and also studied different fixation
techniques for the most part. However, both studies
showed that fixation with multiple Kirschner or intraos-
seous wires improved stability compared with crossed
Kirschner wires alone. Plates or wires placed at ad-
vantage on the tension surface also provided superior
stability.
Stability of internal fixation in proximal phalanx 677
REFERENCES
I. Barton NJ: Fractures of the shafts of the phalanges of
the hand. Hand 11:I 19-33, 1981
2. Belsole R: Physiological fixation of displaced and unsta-
ble fractures of the hand. Orthop Clin North Am 11:393-
404, 1980
3. Brown PW: The management of phalangeal and meta-
carpal fractures. Surg Clin North Am 53: 1393-437, 1973
4. Heim V, Pfeiffer KM: Small fragment set manual. Berlin,
1982, Springer-Verlag
5. Meyer VE, Chiv DT, Beasley RW: The place of internal
skeletal fixation in surgery of the hand. Clin Plast Surg
8:51-64, 1981
6. Alexander H, Langrana N, Massengill 18, Weiss AB:
Development of new methods for phalangeal fracture
fixation. J Biomechanics 14:377-87, 1981
7. Black DM, Mann RJ, Constine R, Daniels AU: Com-
parison of internal fixation techniques in metacarpal frac-
tures. J HAND SURG lOA:466-72, 1985
8. Fyfe IS, Mason S: The mechanical stability of internal
fixation of fractured phalanges. Hand 11 :50-4, 1979
9. Gould WL, Belsole RJ, Skelton WH: Tension band sta-
bilities of transverse fractures: an Experimental analysis.
Plast Reconstr Surg 73: 111-6, 1984
10. Mason SM, Fyfe IS: Comparison of rigidity of whole
tubular bones. J Biomechanics 12:367-72, 1979
II. Massengill 18, Alexander H, Langrana N, Mylod A: A
phalangeal fracture model-quantitative analysis of ri-
gidity and failure. J HAND SURG 7:264-79, 1982
12. Massengill 18, Alexander H, Parson JP, Schecter MJ:
Mechanical analysis of Kirschner wire fixation in a pha-
langeal model. J HAND SURG 4:351-6, 1979
13. Rayhack JM, Belsole RJ, Skelton WH: A strain recording
model: analysis of transverse osteotomy fixation in small
bones. J HAND SURG 9A:383-7, 1984
14. Vanik RK, Weber RC, Matloub HS, Sanger JR, Gingrass
RP: The comparative strengths of internal fixation tech-
niques. J HAND SURG 9A:216-21, 1984
15. Swanson SA: Biomechanical characteristics of bone. Adv
Biomed Eng 1:137-87, 1971
16. Lister G: Intraosseous wiring of the digital skeleton. J
HAND SURG 3:427-35, 1978