SCIENTIFIC ARTICLE

Approach to the Perpendicular Fixation of

a Scaphoid Waist Fracture—A Computer
Analyzed Cadaver Model
Shai Luria, MD,* Samih Badir, MD,* Yonatan Schwarcz, MD,* Eran Peleg, PhD,* Thanapong Waitayawinyu, MD†

Purpose In scaphoid fracture screw fixation, the screw is commonly placed along the long axis
of the bone, without consideration of the fracture plane. This position is not perpendicular to
transverse waist fractures or to the more common horizontal oblique fractures. Our aim was to
examine the feasibility and describe possible approaches to, placing a screw perpendicular
and in the center of the scaphoid waist fracture.
Methods Computed tomography of 12 cadaver wrists was performed in 3 positions to examine
possible approaches in flexion, neutral, and extension of the wrist. The scans were evaluated
using a 3-dimensional model that simulated horizontal oblique (60
) and transverse (90
)
fractures. We examined all possible approaches for screw positioning and their deviation from
the axis perpendicular to the fracture and in the center of its plane.
Results The preferred approaches for a perpendicular screw in a horizontal oblique fracture
were found to be proximal-dorsal in flexion or transtrapezial in the extended or neutral po-
sitions (through the volar-radial trapezium). In transverse fractures, the possible approaches
were proximal-dorsal or transtrapezial in the flexed or neutral positions and distal in the
extended position (volar to volar-radial trapezium). In these approaches, the screw could be
placed perpendicularly (deviating by < 10
) and in the center of the fracture in all specimens.
Conclusions According to this model, it appears feasible to place a perpendicular screw in the
center of a horizontal oblique waist fracture using a proximal-dorsal approach in flexion or a
transtrapezial approach in neutral or extension positions of the wrist. Palpable landmarks may
be used as additional guides to direct these approaches according to the clinical setting.
Clinical relevance Perpendicular screw fixation of horizontal oblique or transverse scaphoid
waist fractures is a possible option, if chosen for its biomechanical advantages. (J Hand Surg
Am. 2019;-(-):1.e1-e10. Copyright Ó 2019 by the American Society for Surgery of the
Hand. All rights reserved.)
Key words Horizontal oblique fracture, perpendicular screw fixation, scaphoid fracture,
transtrapezial approach, waist fracture.

From the *Department of Orthopedic Surgery, Hadassah-Hebrew University Medical Center, Corresponding author: Shai Luria, MD, Department of Orthopedic Surgery, Hadassah-
Jerusalem, Israel; and the †Department of Orthopedic Surgery, Thammasat University, Hebrew University Medical Center, Kiryat Hadassah, POB 12000, Jerusalem 91120, Israel;
Pathumthani, Thailand. e-mail: shail@hadassah.org.il.
Received for publication August 5, 2018; accepted in revised form July 23, 2019. 0363-5023/19/---0001$36.00/0
https://doi.org/10.1016/j.jhsa.2019.07.009
No benefits in any form have been received or will be received related directly or indirectly
to the subject of this article.

Ó 2019 ASSH r Published by Elsevier, Inc. All rights reserved. r 1.e1

1.e2 PERPENDICULAR FIXATION OF SCAPHOID FRACTURE
S
CAPHOID FRACTURES ARE COMMONLY treated by
surgical fixation.1 Placement of a single cannu-
lated screw in the center of the proximal frag-
ment was recommended by Adams et al,2 and later
found to be superior to eccentric positioning of the
screw in a clinical study of fracture nonunions,3 as
well as in a cadaver model of a transverse waist frac-
ture.4 This approach that disregards the fracture and
directs fixation along the scaphoid long axis has
become common practice,5 although its biomechan-
ical superiority has been questioned.6,7 The long-
axis screw was found to be at an angle to transverse
fractures and, to a greater extent, to the majority of
fractures which are horizontal oblique.8,9 Kupperman
et al9 demonstrated that part of the forces of a long-
axis screw will, in fact, be perpendicular to the frac-
ture. The authors examined the longest screw, within
a defined safe zone, from a classic approach. The au-
thors suggest that the surgeon may take the fracture
plane into account by “altering the screw starting
point and trajectory” gaining more screw force
perpendicular to the fracture.9
In basic biomechanical terms, placing a screw
perpendicular and in the center of any fracture plane
has an advantage. A finite element examination
demonstrated this advantage in unstable Herbert B
type scaphoid fractures,10 comparing a screw
perpendicular to the fracture over a long-axis screw.6
Another 3-dimensional simulation of actual fracture
computed tomography (CT) scans demonstrated the
superior compression forces created by the perpen-
dicular screw.7 In these studies, the perpendicular
screw would be in the center of the fracture, not in an
eccentric position.4 An additional advantage to a
perpendicular screw was demonstrated in another
finite element study, indicating that a perpendicular
screw in the horizontal oblique fracture took up a
smaller area of the fracture surface needed for
apposition and healing in comparison with a long-
axis screw.11 Several clinical studies have described
aiming the fixation with a screw perpendicular to the
scaphoid fracture as a standard approach.12,13
Dorsal proximal and volar distal approaches are
the most typical in both open and percutaneous
techniques used today.14e16 As suggested, to place a
screw perpendicular to a horizontal oblique or
transverse fracture, these standard approaches may
require alteration.9,17 The aim of our study was to
examine altering the approaches in order to be able to
place a perpendicular screw in the center of the
fracture. This screw may not be as long as suggested
by Kupperman et al,9 but would not be eccentric. We
chose a cadaver model that allowed us to examine
J Hand Surg Am.
different wrist positions. Our hypothesis was that a
perpendicular screw could be placed in the center of
the fracture plane, either from a proximal approach
with the wrist flexed or from a distal approach in a
neutral position.
METHODS
A sample of convenience of 6 pairs of upper limbs of
embalmed cadavers cut through the midshaft of the
humerus with all soft tissue preserved were assessed
clinically and radiographically to rule out previous
trauma or surgery as well as anatomical abnormal-
ities. The Herbert classification10 was used to simu-
late the most common fractures, specifically,
horizontal oblique and transverse waist fractures (B1
and B2, respectively). These 2 configurations were
chosen based on the typical waist fracture, which is
horizontal oblique,8 or a transverse fracture.18,19
According to the aims of the study, we did not take
into account the degree of displacement or commi-
nution of the fracture but rather its morphology in its
reduced form.
A CT scan of each wrist was carried out in 3 po-
sitions to examine the approaches in positions
mimicking those possible in practice: maximum
flexion, neutral position (without radial or ulnar de-
viation), and maximum extension. The scans were
performed with the wrist held in the extremes of these
positions. The neutral scan was made by laying the
cadaver arm straight on the scan table. The slice
spacing was set at 1 mm with a pixel size of 0.195 
0.195 mm2. Each wrist CT (3 positions for each
wrist) was examined using commercially available 3-
dimensional analysis software (Amira Dev 5.33;
Mercury Computer Systems, Chelmsford, MA) with
the following steps (Fig. 1):
1. Segmentation of the scaphoid geometry was per-
formed with an automated tool followed by minor
manual adjustments for each axial CT image,
resulting in a scaphoid surface mesh.8 Scaphoid
longitudinal axis computation was performed us-
ing a principal component analysis algorithm.8
Volar-dorsal and radial-ulnar directions were
based on the distal radius joint surface as defined
by 4 anatomical markers placed on the joint edges
on the 3-dimensional model—the posterior and
anterior edges of the ridge between the scaphoid
and the lunate facets, the middle of the ulnar
notch, and the farthest edge of the styloid process.
2. Creation of virtual Herbert type B1 and B2 frac-
tures. The B1 (horizontal oblique) fracture was set
at 60
to the longitudinal axis and inclined in the
r Vol. -, - 2019
 PERPENDICULAR FIXATION OF SCAPHOID FRACTURE 1.e3

dorsal direction (distal volar, proximal dorsal)

based on data of the most common fracture
inclination.8 The B2 (transverse) fracture was set
at exactly 90
to the scaphoid longitudinal axis.
Both fractures crossed the longitudinal axis at its
center. A line was marked perpendicular (90
) and
in the computed middle of the fracture plane,
representing the screw in the optimal central and
perpendicular position.
3. The possible approaches for screw placement to
achieve this optimal position were examined,
including measurement of the possible deviations
from the axis trajectory (with a free range for
screw placement options). The screw width was
set to 3.6 mm, comparable with a Mini-Acutrak2
screw (Acumed, Hillsboro, OR). We took into
account the largest diameter, not the change in
screw diameter. For each specimen in each wrist
position, the possible approaches considered the
adjacent bones, including and excluding penetra-
tion into these bones, in an attempt to get as close
as possible to the optimal position. Deviation of
the best possible screw axis from the optimal
position was measured for all approaches and
wrist positions. The approaches achieving the best
position were defined as those that deviated less
than 10
from the optimal position (angle of de-
viation) with as little offset as possible from the
center of the fracture plane.
The reliability of the measures were examined by
repeating all stages of the computation by the same
research assistant on 4 random unmarked specimens,
with 1 week between each measurement session. The
intraobserver error found an insignificant difference
between the pairs of measurements (Wilcoxon
nonparametric paired test) and a high correlation
between these pairs (between 0.8 and 1.0; Spear-
man’s nonparametric correlation coefficient).
For each of these approaches, additional measures
were collected:
1. The length of a virtual screw through the scaphoid
between the scaphoid cortices. In the approaches
achieving a central perpendicular axis, we exam-
ined the proportion of screw length proximal and
distal to the fracture plane. FIGURE 1: A flowchart of the steps carried out for each sample.
2. The distance of each virtual screw entry point
from palpable or recognizable anatomical land- tubercle, a longitudinal axis was drawn along the
marks: (1) for the distal approach: the volar-radial radius, through the tubercle, to the radiocarpal joint
edge of the trapezium (over the scaphoid tuber- and the distance to the entry point was measured
osity) (Fig. 2); (2) for the proximal approach: radial or ulnar to this axis along the joint line
Lister tuberosity of the radius and apex of the (Fig. 3). The apex of the dorsal scaphoid ridge may
dorsal scaphoid ridge (Fig. 3). For Lister’s be appreciated through an open dorsal approach.

J Hand Surg Am. r Vol. -, - 2019

1.e4 PERPENDICULAR FIXATION OF SCAPHOID FRACTURE

FIGURE 2: Three-dimensional view of the volar wrist in neutral position and a transtrapezial approach for placing a perpendicular screw
through the center of a horizontal oblique fracture (white surface within the waist of purple-colored scaphoid). The distance between the
entry point through the volar-radial trapezium and the edge of volar-radial corner of the trapezium, over the scaphoid tuberosity (black
line on trapezium measuring 6.3 mm) is shown. In the scaphoid itself, the blue line is the longitudinal axis of the bone and the red line
depicts the perpendicular screw. Inset: Dorsal view of the same wrist. The screw trajectory (red line) exits through the dorsal-radial
aspect of the distal radioulnar joint.

3. The exit direction was recorded for each approach,
in relation to similar anatomical markers (Figs. 2,
3 insets).
4. The wrist angle was measured for each position in
a coronal view of the 3-dimensional model, be-
tween the radius long axis and the dorsal aspect of
the middle finger metacarpal.
Outcome measures
The outcome measures were the angle and the rela-
tive distance on the fracture plane between the best
axis achieved and the exact central perpendicular
location (Fig. 4). As described previously, we named
these variables the angle of deviation (from the
optimal perpendicular screw axis position)17 and the
offset (distance from the center of the fracture plane).
In the approaches that were found to achieve a central
perpendicular axis, the virtual screw length as well as
the distance of the entry point to specific anatomical
markers were measured.
Statistical methods
Sample size: The sample size was calculated for a 95%
confidence interval (95% CI) for the percentage of
overall success of achieving an optimal location of
the fixation axis (with no offset and < 10
angle
deviation). Using a sample of convenience and given
72 simulations, while defining a minimal success rate
of 90%, a 95% CI level was found to be between 81%
and 96%. We regard the lower limit of the 95% CI as
sufficient for the purpose of this study, considering
the strict criteria of optimal location, high level of
accuracy of the computer model, and as based on a
previous publication that demonstrated lower success
rates and differences in these rates using different
approaches.17
Data analysis: Repeated measures analysis of variance
was used in order to examine the effect of wrist po-
sition and approach as well as these factors and screw
length, for each type of fracture.
RESULTS
The position of the best possible screw axes for all
specimens is summarized in Table 1. Of the 12
specimens examined, 1 was omitted owing to defor-
mity of its distal radius as seen on the CT scan.

J Hand Surg Am. r Vol. -, - 2019

 PERPENDICULAR FIXATION OF SCAPHOID FRACTURE 1.e5

FIGURE 3: Three-dimensional dorsal view of a flexed wrist and a proximal approach for placing a perpendicular screw through the
center of a horizontal oblique fracture (white surface within the waist of a purple-colored scaphoid). The distance between the entry
point into the scaphoid radial base can be measured A from the apex of the dorsal scaphoid ridge (white line on scaphoid, measuring
13.2 mm) or B from Lister tubercles long axis (black line) along the radiocarpal joint line (white line; measuring 2 mm). The yellow
sphere on the radial joint surface marks the dorsal radial joint surface. In the scaphoid itself, the blue line is the longitudinal axis of the
bone and the red line depicts the perpendicular screw. Inset: Volar view of the same wrist. The screw trajectory (red line) exits through
the volar-radial trapezium.

In horizontal oblique and transverse fractures, the
mean angle of deviation from the optimal screw po-
sition was affected by the position of the wrist
(extension, flexion, or neutral) (P < .05) and the
approach for screw placement (proximal, distal, or
transtrapezial) (P < .05 and P ¼ .05, respectively).
The combined effect was significant as well (P ¼ .05
and P < .05, respectively).
In horizontal oblique fractures, the smallest mean
angle of deviation was achieved by placing the virtual
screw in 2 directions: (1) Through a proximal
approach with the wrist flexed (0
deviation and no
offset from the perpendicular axis) or (2) A distal
transtrapezial approach, through the volar-radial
corner of the trapezium, with the wrist extended (0

deviation and no offset) or neutral (7.3
deviation and
no offset). An adequate central axis (< 10
deviation)
could be achieved with a proximal approach with the
wrist extended, although this required an approach
volar to the radius. We consider this less appropriate,
especially for a percutaneous technique.
In transverse fractures, the smallest mean angles
were achieved placing the virtual screw through
different approaches: (1) Proximal approach, with
the wrist in flexion (0
deviation and no offset from
perpendicular axis), or in a neutral position (6.4

deviation and no offset); (2) transtrapezial approach
through the volar-radial angle of the trapezium,
with the wrist flexed (2.7
deviation and no offset)
or neutral (3.2
deviation and no offset); or (3)
distal approach, volar and radial to the trapezium in
an extended wrist (5.9
deviation and no offset).
We did not measure the entry points to the trans-
trapezial approach in extension, although it was
found to be adequate (7.3
deviation with no
offset), given the results with smaller angles ach-
ieved with the transtrapezial approach in a neutral
or flexed wrist.
Virtual screw lengths (Table 2) were measured for
the approaches achieving central perpendicular axes.
We found that the screws would be shorter for hori-
zontal oblique fractures than transverse fractures,
with each of the attempted approaches (P < .05). The
approach or position of the wrist did not affect screw
length. The distance between the axis entry points
and the anatomical landmarks for the approaches are
presented in Table 2. In the examined cases achieving
a central perpendicular axis, a mean of 48% of the
screw (SD, 0.02%) was in the proximal fragment.
The mean wrist angle measured radiographically in
the neutral position was 10
(SD, 4
), in flexion, 106

(SD, 9
), and in extension, 103
(SD, 9).

J Hand Surg Am. r Vol. -, - 2019

1.e6 PERPENDICULAR FIXATION OF SCAPHOID FRACTURE

FIGURE 4: AeC Schematic representation of a lateral view of the scaphoid and the study outcome measures. The fracture is
demonstrated in green and the long axis of the scaphoid in brown. The dashed black line represents the optimal axis, perpendicular and
in the center of the fracture plane. The red line represents the best position achieved from a proximal approach A, C or distal approach B.
The distance between the red line and the dashed black line on the fracture plane is the offset and the angle between these lines, the
deviation angle. A The position is clearly inadequate, yet this is the best position possible when attempting to place a perpendicular
screw from a proximal approach with the wrist in neutral.

DISCUSSION The proximal approach should be performed with

In this computer-analyzed cadaver study, we
demonstrated the possible positioning options of a
screw perpendicular and in the center of a waist
fracture. According to this model, in the common
horizontal oblique fractures, this can be
achieved through a standard proximal approach or
the less commonly used distal transtrapezial
approach.
the wrist fully flexed. In a percutaneous approach
under the specific conditions described here, the entry
point should be slightly (2 mm) radial to the long axis
of Lister tubercle of the radius. In an open approach
(or percutaneously with the aid of fluoroscopy), it
should be 12 mm proximal to the apex of the
scaphoid dorsal ridge. Under these conditions, the
aim of the drilled Kirschner wire should be toward

J Hand Surg Am. r Vol. -, - 2019

 PERPENDICULAR FIXATION OF SCAPHOID FRACTURE 1.e7

the ulnar-volar aspect of the trapezium, palpating the

Offset
volar aspect of the thumb carpometacarpal joint or

0
0
0
scaphoid tuberosity (Fig. 3 inset). A transtrapezial

Transtrapezial
approach with the wrist in a neutral position can
achieve the same trajectory through the volar-radial

Deviation

3 ( 6)‡
7 ( 7)
3 ( 5)
angle
corner of the trapezium (8 mm from its volar edge
Deviation and Offset of Best Axis for Each Wrist Position From the Exact Central Perpendicular Approach for Screw Placement

in this model) and aimed toward the dorsal-radial

aspect of the distal radioulnar joint (Fig. 2 inset).
The transtrapezial approach in full extension may be
Transverse (90
) Fracture

Offset
performed through the volar-radial corner of the tra-
0
0
0
pezium (7 mm from its edge in this model) aiming
Distal

toward the volar aspect of the scaphoid fossa. In cases

10 ( 10)† in which preliminary fixation of a displaced fracture
Deviation

†
6 ( 7)
16 ( 9)
angle

limits the positioning of the wrist, the approach for

final screw placement can be chosen according to
wrist position.
Entry point *volar to radius, †volar and radial to trapezium, ‡through volar radial trapezium, §radial and dorsal to trapezium, or kdorsal to radius.

It should be noted that the use of Lister tubercle or

Offset

the dorsal aspect of the distal radioulnar joint is

1 ( 1)
0
0

highly dependent on the position of the wrist. All of

Proximal

the landmarks presented were measured in order to

examine their use as possible additional tools when
Deviation

30 ( 8)*

aiming the fixation wire. Considering the consistency

angle

6 ( 7)
0

in their position in this model, it is possible that they

be used as additional guides for an approach, under
the conditions described here, in neutral position or
Offset

full flexion or extension of the wrist with no devia-

0
0
0
Transtrapezial

tion. These guides cannot replace the use of fluo-

roscopy or direct visualization for wire positioning.
Deviation

Owing to the variability of wrist position, variability

‡
angle

14 ( 9)
7 ( 8)‡

in the ease of palpation of these landmarks, and other

clinical considerations, these measures may not be
‡
0

relevant in many clinical cases.

Horizontal Oblique (60
) Fracture

A similar attempt at examining the feasibility of a

0 ( 1)
e1 ( 1)
0 ( 0)
Offset

perpendicular screw has recently been published.17

The authors examined CT scans of actual patients
Distal

simulating fractures and different positions of the

wrist using the computer model. The authors found
†

26 ( 5)†
Deviation

22 ( 9)
30 ( 2)

the best possible nonobstructed screw placement in

angle

horizontal oblique fractures to be 40
in the standard

volar approach and 14
in the dorsal approach.17 In
contrast, in the current report, we have demonstrated
approaches enabling placement of perpendicular and
Offset

0 ( 1)

0 ( 0)
0

central screws. There are several differences between

Proximal

the 2 studies. In the current study, we examined

actual positions of cadaver wrists for the creation of
Deviation

the computer model and not computer simulations of

9 ( 14)*

24 ( 8)k
angle

wrist position. We chose to position the screw in the

center of the fracture and not the waist area, as in the
previous report.17 We examined the transtrapezial
Approach

approach and not only the standard volar approach

TABLE 1.

position
Wrist

Extended

and attempted to describe and quantify the entry point

Neutral
Flexed

and exit direction of the optimal approaches. Finally,

there may be a difference in the anatomical features

J Hand Surg Am. r Vol. -, - 2019

1.e8 PERPENDICULAR FIXATION OF SCAPHOID FRACTURE

TABLE 2. Screw Length and Entry and Exit Points of Preferable Approaches to Place a Central Perpendicular
Fixation Screw
Horizontal Oblique
(60
) Fracture Transverse (90
) Fracture

Extended Approach Transtrapezial Distal

Screw length, mm 23 ( 2) 25 ( 2)
Entry point distance to volar-radial 7 ( 2) 6 ( 1)
edge of trapezium, mm
Exit landmark Volar/distal to Ulnar side of radial
scaphoid fossa shaft, proximal to
distal radioulnar
joint
Flexed Approach Proximal Proximal Transtrapezial
Screw length, mm 23 ( 3) 26 ( 2) 25 ( 3)
Entry point distance to
Volar-radial edge of the trapezium 8 ( 3)
Longitudinal axis of Lister tuberosity of 2 ( 2) 3 ( 2)
the radius along radiocarpal joint line,
mm
apex of dorsal scaphoid ridge, mm 12 ( 2) 9 ( 2)
Exit landmark Volar/ulnar trapezium Volar/radial Ulnar/distal to Lister
trapezium tubercle
Neutral Approach Transtrapezial Proximal Transtrapezial
Screw length, mm 24 ( 3) 26 ( 2) 26 ( 2)
Entry point distance to
Volar-radial edge of the trapezium, mm 8 ( 2) 5 ( 1)
Longitudinal axis of Lister tuberosity of 4 ( 2)
the radius along radiocarpal joint line,
mm
Apex of dorsal scaphoid ridge, mm 9 ( 1)
Exit landmark Dorsal/radial aspect of Volar/radial Ulnar/distal to Lister
distal radioulnar edge of the tubercle
joint trapezium

and joint range of motion of live European patients17
and cadavers of Thai origin, which were used in our
study.
The use of a transtrapezial approach may not be a
standard approach in most clinical practices, yet this
is not novel.20,21 The transtrapezial approach may
enable superior screw positioning in comparison with
a standard volar approach that necessitates excision
of part of the trapezium or manipulation of the joint
and has been reported to have a higher risk of injuring
the scaphotrapezial ligament.22 It has been demon-
strated that the transtrapezial approach does not result
in significant arthritic changes to the scaphotrapezial
joint.23 In order to achieve superior screw posi-
tioning, this approach may find new popularity.
We found virtual screw lengths were significantly
shorter for perpendicular fixation of horizontal
oblique fractures in comparison with transverse
fractures. We believe this to be of limited significance
when comparing different screw trajectories. In gen-
eral, a longer screw for the fixation of the scaphoid
fracture has been considered superior biomechani-
cally. Yet the evidence to support this is a cadaver
study24 that tested 2 screw lengths along the same
trajectory in the same specimens and screw holes,
examining long-axis screw fixation of transverse
fractures. In contrast, a different cadaver study
examining perpendicular versus long-axis screw fix-
ation of horizontal oblique fractures found the
perpendicular position to be as strong, although
significantly shorter than the long-axis screw.18 It is
reasonable to consider the position of the screw in
relation to the fracture plane and not only its length as
factors affecting fixation stability.18 This has recently

J Hand Surg Am. r Vol. -, - 2019

 PERPENDICULAR FIXATION OF SCAPHOID FRACTURE
been termed effective compression length, a combi-
nation of screw length and angle to fracture plane, in
a long-axis screw.9 In this study, we did not examine
the implications of placing a long-axis screw. We
attempted to examine the option of perpendicular and
central fixation, even if this meant we needed to
approach these fractures differently and that the screw
may be shorter.
The position of the perpendicular screw may have
an additional biomechanical advantage. According to
a finite element analysis of cadaver bone density and
cortical thickness, more rigid fixation can be achieved
between the midcarpal side of the tuberosity and the
radial side of the proximal pole25 (Fig. 3). Although a
similar direction has been demonstrated in the current
study, we did not examine bone density and cortical
thickness along the screw axis to demonstrate this.
The limitations of this study relate mainly to the
model. Soft tissue constraints might result in more
difficult positioning of an implant than in the com-
puter model, although this was taken into account to
some extent with respect to the screw width. This
computer model cannot take into account other con-
straints dealt with in clinical practice, such as limited
range of wrist motion, significant swelling, or bony
deformity. Wrist position examined in this study on
radiographs may approximate clinical flexion or
extension,26,27 but may not always be achievable. As
mentioned, the study was conducted on cadaver
wrists of Thai origin. Differences in wrist range of
motion have been demonstrated between sexes, sides,
and geographical locations.28,29 It is also possible that
wrist flexion or extension in actual anesthetized pa-
tients would differ in comparison with patients who
are awake, although we found no such comparison in
the literature. If these wrist positions are not possible
during surgery, a distal transtrapezial approach may
be used in neutral wrist positioning to achieve the
desired screw position.
As mentioned, the use of the different landmarks
as additional guides to wire insertion may not be
practical in all cases. The landmarks may not be
easily palpable and suggested measurements only
good in the positions represented in the cadaver-
computer model. By examining these landmarks,
we attempted to add tools that may, in certain situa-
tions, aid in directing the fixating wire, even without
using precise measurements.
This study had a specific aim and did not address
the biomechanical characteristics of screw position,
only the feasibility of screw positioning. Less com-
mon fracture configurations were not examined, such
as vertical oblique fractures. Obviously, each case
J Hand Surg Am.
1.e9
must be examined specifically when choosing the
approach. Finally, we did not evaluate wrist devia-
tion, which is more difficult to simulate consistently.
We chose to examine reproducible positions, which
are commonly used and are easy to replicate.
There are several advantages to a central perpen-
dicular screw, and in some locations, it is in clinical
use. This computer-analyzed cadaver study clearly
showed that a central perpendicular screw is a
promising alternative in horizontal oblique as well as
transverse fractures by implementing slight variations
to standard approaches. Application of these specific
approaches in clinical practice may not be simple at
times, yet the additional tools described may direct
the surgeon when attempting to place a central
perpendicular screw in the scaphoid waist fracture.
ACKNOWLEDGMENTS
We recieved a grant from the Thammasat University
Research Fund. We would also like to acknowledge
Dr. Tali Bdolah-Abram of the Hadassah-Hebrew
University Medical School for her help with the
statistical analysis.
REFERENCES
1. Lee SK. Fractures of the Carpal Bones. 7th ed.In: Wolfe SW,
Pederson WC, Hothckiss R, Kozin SH, Cohen MS, eds. Green’s
Operative Hand Surgery Vol 1. Philadelphia: Elsevier; 2016:
588e652.
2. Adams BD, Blair WF, Reagan DS, Grundberg AB. Technical factors
related to Herbert screw fixation. J Hand Surg Am. 1988;13(6):
893e899.
3. Trumble TE, Clarke T, Kreder HJ. Non-union of the scaphoid.
Treatment with cannulated screws compared with treatment with
Herbert screws. J Bone Joint Surg Am. 1996;78(12):1829e1837.
4. McCallister WV, Knight J, Kaliappan R, Trumble TE. Central
placement of the screw in simulated fractures of the scaphoid waist: a
biomechanical study. J Bone Joint Surg Am. 2003;85(1):72e77.
5. Verstreken F, Meermans G. Transtrapezial approach for fixation of
acute scaphoid fractures: rationale, surgical techniques, and results:
AAOS exhibit selection. J Bone Joint Surg Am. 2015;97(10):
850e858.
6. Luria S, Hoch S, Liebergall M, Mosheiff R, Peleg E. Optimal fixation
of acute scaphoid fractures: finite element analysis. J Hand Surg Am.
2010;35(8):1246e1250.
7. Swanstrom MM, Morse KW, Lipman JD, Hearns KA, Carlson MG.
Effect of screw perpendicularity on compression in scaphoid waist
fractures. J Wrist Surg. 2017;6(3):178e182.
8. Luria S, Schwarcz Y, Wollstein R, Emelife P, Zinger G, Peleg E. 3-
Dimensional analysis of scaphoid fracture angle morphology. J Hand
Surg Am. 2015;40(3):508e514.
9. Kupperman A, Breighner R, Saltzman E, Sneag D, Wolfe S, Lee S.
Ideal starting point and trajectory of a screw for the dorsal approach
to scaphoid fractures. J Hand Surg Am. 2018;43(11):993e999.
10. Herbert TJ, Fisher WE. Management of the fractured scaphoid using
a new bone screw. J Bone Joint Surg Br. 1984;66(1):114e123.
11. Hart A, Mansuri A, Harvey EJ, Martineau PA. Central versus
eccentric internal fixation of acute scaphoid fractures. J Hand Surg
Am. 2013;38(1):66e71.
r Vol. -, - 2019
1.e10 PERPENDICULAR FIXATION OF SCAPHOID FRACTURE
12. Garcia RM, Leversedge FJ, Aldridge JM, Richard MJ, Ruch DS.
Scaphoid nonunions treated with 2 headless compression screws and
bone grafting. J Hand Surg Am. 2014;39(7):1301e1307.
13. McInnes CW, Giuffre JL. Fixation and grafting after limited
debridement of scaphoid nonunions. J Hand Surg Am. 2015;40(9):
1791e1796.
14. Cosio MQ, Camp RA. Percutaneous pinning of symptomatic
scaphoid nonunions. J Hand Surg Am. 1986;11(3):350e355.
15. McQueen MM, Gelbke MK, Wakefield A, Will EM, Gaebler C.
Percutaneous screw fixation versus conservative treatment for frac-
tures of the waist of the scaphoid: a prospective randomised study.
J Bone Joint Surg Br. 2008;90(1):66e71.
16. Slade JF III, Dodds SD. Minimally invasive management of scaphoid
nonunions. Clin Orthop Relat Res. 2006;445:108e119.
17. Ten Berg PWL, Dobbe JGG, Brinkhorst ME, et al. Scaphoid screw
fixation perpendicular to the fracture plane: comparing volar and dorsal
approaches. Orthop Traumatol Surg Res. 2018;104(1):109e113.
18. Faucher GK, Golden ML III, Sweeney KR, Hutton WC, Jarrett CD.
Comparison of screw trajectory on stability of oblique scaphoid
fractures: a mechanical study. J Hand Surg Am. 2014;39(3):
430e435.
19. Russe O. Fracture of the carpal navicular. Diagnosis, non-operative
treatment, and operative treatment. J Bone Joint Surg Am. 1960;42-
A:759e768.
20. Meermans G, Verstreken F. Percutaneous transtrapezial fixation of
acute scaphoid fractures. J Hand Surg Eur Vol. 2008;33(6):
791e796.
21. Verstreken F, Meermans G. Transtrapezial approach for fixation of
acute scaphoid fractures. JBJS Essent Surg Tech. 2015;5(4):e29.
J Hand Surg Am.
22. Meermans G, Van Glabbeek F, Braem MJ, van Riet RP, Hubens G,
Verstreken F. Comparison of two percutaneous volar approaches for
screw fixation of scaphoid waist fractures: radiographic and biome-
chanical study of an osteotomy-simulated model. J Bone Joint Surg
Am. 2014;96(16):1369e1376.
23. Geurts GF, Van Riet RP, Meermans G, Verstreken F. Volar percu-
taneous transtrapezial fixation of scaphoid waist fractures: surgical
technique. Acta Orthop Belg. 2012;78(1):121e125.
24. Dodds SD, Panjabi MM, Slade JF III. Screw fixation of scaphoid
fractures: a biomechanical assessment of screw length and screw
augmentation. J Hand Surg Am. 2006;31(3):405e413.
25. Lee SB, Kim HJ, Chun JM, et al. Osseous microarchitecture of the
scaphoid: cadaveric study of regional variations and clinical impli-
cations. Clin Anat. 2012;25(2):203e211.
26. Han KD, Kim JM, DeFazio MV, et al. Changes in wrist motion after
simulated scapholunate arthrodesis: a cadaveric study. J Hand Surg
Am. 2016;41(9):e285ee293.
27. McVeigh KH, Murray PM, Heckman MG, Rawal B, Peterson JJ.
Accuracy and validity of goniometer and visual assessments of
angular joint positions of the hand and wrist. J Hand Surg Am.
2016;41(4):e21ee35.
28. Allander E, Bjornsson OJ, Olafsson O, Sigfusson N, Thorsteinsson J.
Normal range of joint movements in shoulder, hip, wrist and thumb
with special reference to side: a comparison between two pop-
ulations. Int J Epidemiol. 1974;3(3):253e261.
29. Gunal I, Kose N, Erdogan O, Gokturk E, Seber S. Normal range of
motion of the joints of the upper extremity in male subjects, with
special reference to side. J Bone Joint Surg Am. 1996;78(9):
1401e1404.
r Vol. -, - 2019