Veterinary Surgery

27:301-306, 1998

Holding Power of Different Pin Designs and Pin Insertion

Methods in Avian Cortical Bone
LAUREL A. DEGERNES, DVM, Diplomate ABVP, SIMON c. ROE, BVSC, PhD,
and C. FRANK ABRAMS, Jr, P ~ DPE
,

Diplomate

ACVS,

Objective-To measure pullout strength of four pin types in avian humeri and tibiotarsi bones
and to compare slow-speed power and hand insertion methods.
Study Design-Axial pin extraction was measured in vitro in avian bones.
Animal Population-Four cadaver red-tailed hawks and 12 live red-tailed hawks.
Methods-The pullout strength of four fixator pin designs was measured: smooth, negative
profile threaded pins engaging one or two cortices and positive profile threaded pins. Part 1:
Pins were placed in humeri and tibiotarsi after soft tissue removal. Part 2: Pins were placed in
tibiotarsi in anesthetized hawks using slow-speed power or hand insertion.
Results-All threaded pins, regardless of pin design, had greater pullout strength than smooth
pins in all parts of the study (P < .OOOl). The cortices of tibiotarsi were thicker than the cortices
of humeri ( P < .OOOl). There were few differences in pin pullout strengths between threaded
pin types within or between bone groups. There were no differences between the pullout strength
of pins placed by slow-speed power or by hand.
Conclusions-There is little advantage of one threaded pin type over another in avian humeri
and tibiotarsi using currently available pin designs. There were few differences in pin pullout
strengths between humeri and tibiotarsi bones. It is possible that the ease of hand insertion in
thin cortices minimizes the potential for wobbling and therefore minimizes the difference between
slow-speed drill and hand insertion methods.
Clinical Relevance-Threaded pins have superior bone holding strength in avian cortices and
may be beneficial for use with external fixation devices in birds.
OCopyright 1998 by The American College of Veterinary Surgeons

IRDS HAVE UNIQUE skeletal modifications

to reduce weight during flight, including loss or
fusion of bones, reduction in cortical bone thickness,
and pneumatization of medullary cavities.' The thin,
brittle cortices, however, often complicate avian
fracture repair because of poor pin or screw holding
power and potential problems with iatrogenic fractures.2 Other complications of avian orthopedics include small patient size, scant soft tissues covering
distal extremities, and fractures near joints that often

lead to reduced joint m ~ b i l i t yMost

. ~ of the pioneering work in avian orthopedics has been done in raptors using techniques adapted from small animal^.^.^
A few controlled studies of avian fracture healing
and stabilization have been conducted, using pigeons
as a m0de1.~-~
External skeletal fixators (ESF) have been successfully used in a variety of avian fracture^.^.^.'^."
Premature pin loosening at the fixator pin-bone interface, however, is one of the most common

From the Department of Companion Animals and Special Species Medicine, College of Veterinary Medicine, and the Department
of Biological and Agricultural Engineering, College of Agriculture and Life Sciences, North Carolina State University, Raleigh,
NC .
This study was supported by a Department of Companion Animals and Special Species Medicine Grant, College of Veterinary
Medicine, North Carolina State University.
Address reprint requests to Laurel Degernes, DVM, Department of Companion Animals and Special Species Medicine, College
of Veterinary Medicine, NCSU, Raleigh, NC 27606.
OCopyright 1998 by The American College of Veterinary Surgeons
016 1-3499/98/2704-0002$3.00/0

301

302

HOLDING POWER OF PINS IN AVIAN CORTICAL BONE

causes of ESF failure in birds2 and

Biomechanical studies in mammals have investigated the influence of pin design and pin insertion
methods on the bone-holding power of pins.
Smooth pins have less resistance to pullout compared with end-threaded pins with negative profile
threads (threads cut into the shaft).I4-l6 Endthreaded pins with double cortical contact provide
better holding power than single cortical contact
end-threaded pins.I4 Pins with a positive thread
profile (outer thread diameter greater than shaft
diameter) have been shown to have increased
strength and stiffness and superior bone-holding
power.'6317Other elements of the pin design, such
as the diameter of the pin and the tooth profile and
pitch (number of threads per unit length) of the
threaded portion, also influence the holding
power."
The method of pin insertion has also been identified as a major influence on premature pin loosening
in dog^.'^,'^ High-speed power drilling or slow-speed
power drilling without predrilling the hole can cause
thermal necrosis and osteocyte damage. 12~19*20Hand
insertion (eg, Jacobs chuck) may cause more structural damage at the pin-bone interface because of
wobbling during pin insertion. '83'9 Slow-speed power
drilling or predrilling with a twist bit, followed by
hand insertion, has been shown to result in greater
holding power as measured by the axial force required to extract the pin from the bone.I4.l9
The objective of this study was to measure pullout
resistance of four pin types in avian humeri and
tibiotarsi bones and to compare slow-speed power
and hand insertion methods. Commercially available
orthopedic pins were designed for mammals, so it
was not known if these pins would have similar
biomechanical properties in avian cortices. We hypothesized that the pin type, bone type, and method
of pin insertion would affect the resistance to pin
pullout from avian cortical bone. This study was
divided into two separate parts. The objective of the
first part of the study was to measure and compare
the resistance to pin pullout of four different fixator
pin types in the humerus (pneumatic) and tibiotarsus
(medullary) bones. The four pin types were bicortical
smooth pins, bicortical pins with negative profile
threads engaging one (NEG- 1) or two (NEG-2) cortices, and bicortical pins with positive profile threads
(POS). The objective of the second part of the study
was to measure and compare the resistance to pin

pullout of each of the four pin types after pin insertion into the tibiotarsal bone using slow-speed power
(150 rpm) or hand insertion.

MATERIALS AND METHODS

Four frozen red-tailed hawk (Buteo jamaicensis) carcasses and 12 permanently disabled but otherwise healthy
red-tailed hawks were obtained from a raptor rehabilitation center. All of the birds were older than 1 year of
age as determined by plumage. Red-tailed hawks were
selected because of their large size (900 to 1,200 g) and
availability from wildlife rehabilitation centers. These
hawks are suitable as a model for avian orthopedic research because they are representative of birds commonly
admitted to wildlife rehabilitation centers with orthopedic
injuries. The live hawks were housed in a large flight pen
and were fed thawed laboratory rodents once daily and
provided fresh water ad libitum. None of the birds had
humerus or tibiotarsus fractures, as determined by physical examination before the study and gross examination
of the bones after pin placement and soft tissue removal.
This study was approved by the North Carolina State
University Institutional Animal Care and Use Committee.

Part I : Pin Pullout Tests on the Humerus and

Tibiotursus Bones
Eight humeri (group 1) and tibiotarsi (group 2) were
dissected from the thawed hawk carcasses, and all soft
tissues were removed. The bones were wrapped in salinesoaked gauze sponges and frozen at -20C until used in
the study. Commercially available stainless steel pins with
trochar points (IMEX Veterinary, Inc, Longview, TX)
were used in the study (1 2 of each pin type per group):
(1) smooth intramedullary pin; (2) partial end-threaded,
negative profile intramedullary pin with single cortical
contact (NEG- 1); ( 3 ) partial end-threaded, negative profile intramedullary pin with double cortical contact (NEG2); and (4) partial end-threaded, positive profile acrylic
half-pin (POS; 1.75 mm outside thread diameter). All of
the pins were new and were used only once during the
study. All of the pins had the same shank diameter (1.6
mm), and all of the threaded pins had the same thread
pitch (0.55 mm or 1.8 threaddmm). Ellis pins (Kirschner
Medical Corporation, Timonium, MD) were not available
in the size required for these birds, so the partial endthreaded, negative profile pins were placed with either
single cortical contact (with approximately 1.5 cm
threaded pin exposed on the far cortex to allow for thread
contact with the far cortex only) or double cortical contact
(with threads engaging both cortices). Six pins were
placed at 1.5-cm intervals along the diaphysis of each

303

DEGERNES. ROE. AND ABRAMS

bone. The pins were placed perpendicular to the long axis
of the bone, from medial to lateral in the tibiotarsus and
lateral to medial in the humerus. Two pins of each pin
type were placed at each of the six locations, to account
for possible differences in the cortices at different sections
of the bones. All pins were inserted with a slow-speed
power drill (<150 rpm) until the trochar point extended
completely through the far cortex. Before placement of
the POS pins, the pin site was predrilled with a smooth
1.6-mm trochar point pin.
The bones were placed in a specially designed holding
device during pin extraction. Pins were connected to the
load cell by a drill chuck. The maximal force required to
pull the pin from the bone was measured using a universal
testing machine (Instron Model 1122; Instron Corp, Canton, MA), at an extraction rate of 100 m d m i n . The peak
load for each pin extraction was recorded from the strip
chart. After pin removal, cross-sections of the bones were
cut on a band saw next to each pin insertion site (within
2 mm). Cortical thickness was measured with calipers at
four locations next to the pin insertion site, on the near
and far cortices, and averaged.

Part 2: Pin Pullout Tests on Tibiotarsus Bones

Using Slow-Speed Power Drill and Hand Insertion
Twelve live, anesthetized hawks were used to simulate
a clinical surgical procedure in which muscles obscure
the approach to the bone and make bone stabilization
during pin insertion more difficult. The hawks were anesthetized with an intramuscular injection of ketamine (10
mgkg) and xylazine (2 mgkg). The feathers over each
tibiotarsus were plucked from the stifle to the tarsal joint.
The 12 birds were divided into two groups: Group 1 had
pins inserted using a slow-speed power drill (150 rpm),
and group 2 had pins inserted by hand using a Jacobs
chuck without a pin guard. Pins of each of the four types
were placed in the tibiotarsus from medial to lateral. Three
pins of each type were placed at each of six pin locations
(six pins per bone at 1.5-cm intervals for a total of 18
pins of each type). The pin sites were predrilled with a
smooth 1.6-mm trochar point pin before placement of the
POS pins. The hawks were euthanatized immediately
after pin placement. The soft tissues were dissected off
the tibiotarsi, and the bones were frozen in saline-soaked
sponges until testing. The bones were thawed for 2 to 4
hours before testing. Testing of pullout forces and measurement of cortical thickness were done as described in
Part 1. Cortical thickness data at each pin site were combined from the eight tibiotarsi in Part 1 and 24 tibiotarsi
in Part 2.

STATISTICAL ANALYSIS

A separate analysis of variance was performed for

each bone type in Part 1 or pin insertion method in

Part 2 to determine differences between pin types

within each group. Pin pullout data were transformed
(square root of the pullout variable) to equalize variance between groups (PROC GLM, SAS 6.09). The
pin type, bone side (left or right), and cortical thickness (average of four cortical measurements at each
pin location) were included in the model to adjust for
variation resulting from them. A pin type-cortical
thickness interaction term was included in the original model to determine if changes in the pullout
strength with cortical thickness were the same for
all pin types. Because significant differences were
not found in this interaction (P 2 .22), the final
analysis did not contain this term. To determine
which pin types were different, pairwise t-tests were
carried out on the least square means of the pullout
data. t-Tests were used to compare pin pullout data
within each pin type between the humerus and tibiotarsus groups and between the slow-speed drill and
hand drill pin insertion groups (PROC t-test of SAS).
Pairwise t-tests were done on the least square means
of the cortical thickness to determine differences in
cortical thicknesses at the six pin locations. The
overall significance was set at P I.05. When multiple comparisons were made between the six pin locations, a P value s .0167 was used to reduce the
chance of a type I1 error.

RESULTS
Part 1: Pin Pullout Tests on the Humerus and
Tibiotarsus Bones
Bone side (left or right) did not account for any
effects (P 2 .15), so the data were combined within
groups. The mean cortical thickness was significantly greater in the tibiotarsus than the humerus
(Fig 1; P 5 .0001). Mean cortical thickness at the
six pin locations ranged from 0.55 to 0.81 mm in
the humerus (overall mean, 0.70 mm) and from 0.88
to 1.O mm in the tibiotarsus (overall mean, 0.95 mm).
There was no correlation in any of the groups between cortical thickness and pin pullout (humerus,
R2 = .28; tibiotarsus, R2 = .06).
There was a significant difference in pin pullout
in the humeri and tibiotarsi based on pin type (Fig
2). The smooth pins had a lower pin pullout than
any of the threaded pins for both long bones (P I
.0001 for all comparisons). Within the humerus
group, the NEG-2 pins had a higher pin pullout than

304

HOLDING POWER OF PINS IN AVIAN CORTICAL BONE

4.2

'0
350

Smoom

Fig 1. Cortical thickness by pin location. Mean 2 SD cortical thickness (mm) of the avian humerus and tibiotarsus at
six pin locations, spaced at 1.5-cm intervals from proximal
(1) to distal (6). The tibiotarsus cortex was significantly
thicker than the humerus cortex at each pin location (P 5
.0001 for all comparisons).

the other threaded pins ( P I.009 for all comparisons). There was no difference between NEG-1 pins
and the POS pins (P = .27). Threaded pins were
two to three times more resistant to pullout from the
humerus than smooth pins.
Within the tibiotarsus group, there were no differ-

NEG-1

NEG-2

~-

pln T~~

&ow
~

POS
7

Speed Drill

.Hand I n s e r I i t i
~

Fig 3. Pin pullout resistance using slow-speed power drill

or hand insertion methods. Pullout resistance (Newtons,
mean 2 SD) of four pin types (n = 18) in the avian tibiotarsus: smooth, negative profile with single cortical contact
(NEG-l), negative profile with double cortical contact
(NEG-2), and positive profile pins (POS). The smooth pins
had significantly lower pin pullout resistance than the three
threaded pins in both pin insertion groups (P 5 .0001 for
all comparisons).

ences between the pin pullout forces for any of the

threaded pins (Fig 2; P 2 .92 for all comparisons).
Threaded pins were three times more resistant to
pullout from the tibiotarsus than smooth pins.
The pin pullouts were compared between the humerus and the tibiotarsus to determine if there were
differences between pneumatic versus medullary
bones. When resistance to pullout for each pin type
was compared between the two groups of bones,
only the NEG- 1 pins had greater strength in the tibiotarsus (Fig 2; P 5 .02).
Part 2: Pin Pullout Tests on Tibiotarsus Bones
Using Slow-Speed Power Drill and Hand Insertion

Fig 2. Pin pullout resistance in the humerus and tibiotarsus. Pullout resistance (Newtons, mean 2 SD) of four pin
types (n = 12): smooth, negative profile with single cortical
contact (NEG-l), negative profile with double cortical contact (NEG-2), and positive profile pins (POS). The smooth
pins had significantly lower pin pullout resistance than the
three threaded pins in both sets of bones (P 5 .0001 for
all comparisons). The NEG-2 pin had greater pin pullout
resistance than the NEG-1 or POS pins in the humerus (P
5 .009). The NEG-1 pin had greater pin pullout resistance
in the tibiotarsus compared with the humerus (P 5 .02).

In both groups, all of the smooth pins had less

resistance to pullout than the threaded pins (Fig 3;
P I.0001 for all comparisons). None of the threaded
pins were different from each other within the hand
insertion group. The only difference between
threaded pin types in the slow-speed power drill
group was that the NEG-2 pins had a significantly
greater pullout strength than the POS pins ( P I.006
for all comparisons). Threaded pins were three to
four times more resistant to pullout than smooth pins
using either pin insertion method. There were no

305

DEGERNES, ROE, AND ABRAMS

differences between the slow-speed power drill and
hand insertion groups when comparing pin pullout
strength within each pin type.

DISCUSSION
Threaded external fixator pins provided more resistance to pullout than smooth pins in avian bones.
This finding was in agreement with studies conducted in marnmal~.~-~*
An unexpected finding of
the study is that there was little difference between
pins with different thread designs. Because pins with
larger diameter should have greater bone-holding
strength, we anticipated that the positive profile pins
would resist pullout better than negative profile
threaded pins.16 The positive profile pin could have
caused damage to the bone at the pin entry site.
Microscopic studies of the bone at the pin entry site
have shown that the bone surface is damaged by the
threads until the threads engage and begin to cut
a tract. Because avian cortices are thin, a larger
proportion of the bone may be damaged by the pin
entry before the threads are formed. Microfissures at
the bone-pin interface could have developed, despite
predrilling. Although no visible cracks were noted
in the bone adjacent to the pins, we did not evaluate
the pin-bone interface microscopically or radiographically in this study.
End-threaded, negative profile pins with double
cortical contact provide better holding power than
similar pins with single cortical contact (eg, Ellis
pins) in canine bone.14 Ellis pins were designed to
minimize the potential for breakage at the weak
threaded-nonthreaded portion of the pin because this
weak section of the pin is protected within the medullary ~ a v i t y . In
~ ,our
~ ~ study, we observed greater
bone-holding strength in the NEG-2 pins in the humerus compared with the NEG-1 pins.
Bone-holding strength of threaded pins increases
linearly with increasing cortical bone thi~kness.~
We expected to find differences in pin pullout
strength between tibiotarsi and humeri because tibiotarsi cortices were an average of 26% thicker than
humeri cortices. The only difference we observed
was in the NEG-1 pin group, which had greater pin
pullout strength in the tibiotarsus when compared
with the humerus. Because only half of the threads
were engaged in cortical bone in the NEG-1 pin
group, compared with the NEG-2 and POS pin
groups, the contribution of each thread engaged in

bone could be important. When the number of engaged threads was decreased even more in the thin
humerus cortex, the difference between threaded pin
types became statistically significant. The differences between the humerus and tibiotarsus in pin
pullout strengths of the other pin types could have
been masked by wide sample variation.
The method of pin insertion has been identified
as a major influence on premature pin loosening in
Our study did not show any differences
within each pin type between the two pin insertion
methods. Little effort is required to drive pins manually through thin avian cortices when compared with
manual pin insertion in thicker cortices. Because of
this, wobbling during pin insertion is less likely to
occur. Proper technique must be used during pin
insertion to avoid wobbling. We did not measure pin
tip temperatures during pin insertion and did not
evaluate bone-holding strength in a long-term study.
It is possible that if thermal necrosis occurred during
pin insertion, more differences would have been evident at a later time.
Threaded pins, regardless of their design, have
better resistance to pullout than smooth pins in avian
long bones. There is reduced pin pullout strength
when single cortical contact negative profile pins
are placed in the avian humerus compared with the
tibiotarsus. For all other pin types studied, however,
the pin pullout resistance from the humerus is comparable to that from the tibiotarsus despite the thicker
cortex in the tibiotarsus. It is likely that the thin
cortices of avian bone diminish the influence of different threaded pin designs. The influence of pitch
and thread depth on the pullout strength of implants
in thin cortical bone deserves further study. There is
no difference in the pullout strength of pins inserted
in avian long bones by either slow-speed power or
hand methods.

ACKNOWLEDGMENT
The authors thank Debbie Whitt Smith and Timothy
Seaboch for their technical expertise and Judith Jayawickrama for statistical support.

REFERENCES
1. King AS, McLelland J: Birds: Their Structure and Function.
Philadelphia, PA, Baillibre Tindall, 1984, pp 43-78
2. Bennett RA, Kuzma AB: Fracture management of birds. J
Zoo Wild1 Med 23538, 1992

306

HOLDING POWER OF PINS IN AVIAN CORTICAL BONE

3. MacCoy DM: Treatment of fractures in avian species. Vet

Clin North Am Small Anim Pract 22:225-238, 1992
4. Redig PT, Roush JC: Orthopedic and soft tissue surgery in
raptorial species, in Fowler ME (ed): Zoo and Wild Animal Medicine. Philadelphia, PA, Saunders, 1978, pp 246253
5. Redig PT: A clinical review of orthopedic techniques used
in the rehabilitation of raptors, in Fowler ME (ed): Zoo
and Wild Animal Medicine (ed 2). Philadelphia, PA,
Saunders, 1986, pp 388-401
6. Bush M, Montali RJ,Novak GR, et al: The healing of avian
fractures: A histological xeroradiographic study. J Am
Anim Hosp Assoc 12:768-773, 1976
7. Newton CD, Zeitlin S: Avian fracture healing. J Am Vet
Med Assoc 170:620-625, 1977
8. MacCoy DM, Haschek WM: Healing of transverse humeral
fractures in pigeons treated with ethylene oxide-sterilized,
dry-stored, onlay cortical xenografts and allografts. Am
J Vet Res 49:106-111, 1988
9. Matthews KG, Wallace LJ, Redig PT,et al: Avian fracture
healing following stabilization with intramedullary polyglycolic acid rods and cyanoacrylate adhesive vs. polypropylene rods and polymethylmethacrylate. Vet Comp
Orthop Traumatol 7:158-169, 1994
10. Grimm F: Fracture healing in predatory birds, with special
consideration given to a modified external fixator, in Redig PT,Cooper JE, Remple JD, et al (eds): Raptor Biomedicine. Minneapolis, MN, University of Minnesota
Press, 1993, pp 136-140
11. Satterfield WC, ORourke KI: External skeletal fixation in
avian orthopedics using a modified through-and-through
Kirschner-Ehmer splint technique (Boston technique). J
Am Anim Hosp Assoc 17:635-637, 1981
12. Egger EL: Complications of external fixation-a problem
oriented approach. Vet Clin North Am Small Anim Pract
21:705-733, 1991
13. Harari J: Complications of external skeletal fixation. Vet
Clin North Am Small Anim Pract 22:99-107, 1992

14. Bennett RA, Egger EL, Histand MB, et al: Comparison of

the strength and holding power of 4 pin designs for use
with half pin (type 1) external skeletal fixator. Vet Surg
16:207-211, 1987
15. Halsey D, Fleming B, Pope MH, et al: External fixator pin
design. Clin Orthop Re1 Res 278:305-312, 1992
16. Anderson MA, Mann FA, Wagner-Mann C, et al: A comparison of nonthreaded, enhanced threaded, and Ellis fixation pins used in type 1 external skeletal fixators in dogs.
Vet Surg 22:482-489, 1993
17. Palmer RH, Hulse DA, Hyman WA, et al: Principles of
bone healing and biomechanics of external skeletal fixation. Vet Clin North Am Small Anim Pract 22:45-69,
1992
18. Egger EL, Histand MB, Blass CE, et al: Effect of pin insertion on the bone-pin interface. Vet Surg 15:246-252, 1986
19. Gumbs JM, Brinker WO, Decamp CE, et al: Comparison
of acute and chronic pull-out resistance of pins used with
the external fixator (Kirschner splint). J Am Anim Hosp
Assoc 24:231-234, 1986
20. Matthew LS, Green CA, Goldstein SA: The thermal effects
of skeletal fixation-pin insertion in bone. J Bone Joint
Surg 66A:1077-1083, 1984
21. Anderson MA, Mann FA, Kinden DA, et al: Evaluation of
cortical bone damage and axial holding power of nonthreaded and enhanced threaded pins placed with and
without drilling of a pilot hole in femurs from canine
cadavers. J Am Vet Med Assoc 2082383-887, 1996
22. Clary EM, Roe SC: In vitro biomechanical and histological
assessment of pilot hole diameter for positive-profile external skeletal fixation pins in canine tibiae. Vet Surg
25~4.53-462,1996
23. Palmer RH, Aron DN: Ellis pin complications in seven
dogs. Vet Surg 19:440-44.5, 1990
24. Koranyi E, Bowman CE, Knecht CD, et al: Holding power
of orthopedic screws in bone. Clin Orthop 72:283-286,
1970