Reconstruction of the Ulnar

Collateral Ligament of the Thumb

Metacarpophalangeal Joint:
A Cadaver Study
Christopher J. Hogan, MD, Durham, NC, Robert T. Ruland, MD,
Portsmouth, VA, L. Scott Levin, MD, Durham, NC
Purpose: The purpose of this study was to compare the stiffness and strength of the native ulnar
collateral ligament with 4 methods of static ulnar collateral ligament (UCL) reconstruction at
the thumb metacarpophalangeal (MCP) joint.
Methods: Eleven fresh-frozen cadaver specimens were amputated at the carpometacarpal and
interphalangeal joints and all soft tissues were removed except for the extensor pollicis brevis
tendon, the proper and accessory collateral ligaments, and the volar plate. Each thumb
metacarpal was potted in cement and the native UCL was loaded to failure at 30 of MCP
exion. Ulnar collateral ligament reconstructions as described by Strandell, Osterman,
Fairhurst, and a modication of the Glickel procedure then were performed. Each specimen
was again loaded to failure and the moment at failure, stiffness, and angle at failure were
calculated.
Results: None of the reconstructions duplicated the strength or stiffness of the native UCL. The
modication of the Glickel procedure with interference knot xation had a signicantly higher
moment at failure and was signicantly stiffer than any of the other procedures. The differences
in strength and stiffness between the Strandell, Osterman, and Fairhurst reconstructions were
not statistically signicant. There were no signicant differences in angle at failure for any of
the reconstructions.
Conclusions: No static ligament reconstruction restores the normal stability characteristics of the
thumb UCL. The anatomic reconstruction of the UCL with interference knot xation of the tendon
graft has far better strength and stiffness than any of the other reconstructions tested. These
characteristics may allow for early motion at the MCP joint. (J Hand Surg 2005;30A:394399.
Copyright 2005 by the American Society for Surgery of the Hand.)
Key words: Gamekeepers thumb, skiers thumb, thumb ulnar collateral ligament reconstruction.
Division of Orthopedic Surgery and Plastic and Reconstructive Surgery, Duke University Medical Center, Durham, NC; and the Department of
Orthopedic Surgery, Charette Health Sciences Center, Portsmouth, VA.
Received for publication February 10, 2004; accepted in revised form September 27, 2004.
No benets in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.
The views expressed in this article are those of the authors and do not necessarily reect the ofcial policy or position of the Department of the Navy,
Department of Defense, nor the U.S. Government.
Reprint requests: Christopher J. Hogan, MD, Dept. of Orthopaedic Surgery, 620 John Paul Jones Circle, Portsmouth, VA 23708.
Copyright 2005 by the American Society for Surgery of the Hand
0363-5023/05/30A02-0024$30.00/0
doi:10.1016/j.jhsa.2004.09.012
394 The Journal of Hand Surgery
Injuries to the ulnar collateral ligament (UCL) of the
thumb metacarpophalangeal (MCP) joint frequently
are seen after athletic injuries or falls in which a
signicant radial deviation force is applied to the
abducted thumb. Nonsurgical management of incom-
plete tears usually results in full recovery of thumb
stability and motion.
13
In contrast complete ruptures
often are associated with a Stener lesion and have
little potential for regaining stability without surgical
intervention. In these cases acute anatomic repair of
the UCL almost uniformly restores strength, stability,
and motion to the thumb MCP joint.
410
If neglected,
however, chronic UCL insufciency ensues, increas-
ing the likelihood of pain, weakness, and premature
degenerative arthritis.
Many treatments have been described to restore
the stability of the thumb in these chronic injuries
including repair using local tissue, dynamic stabili-
zation, and static ligament reconstruction using free
tendon grafts. The clinical results of these interven-
tions vary, ranging from 50% to greater than 90%
restoration of joint stability.
1,3,1121
The biomechanical characteristics of the native
and acutely repaired UCL have been described.
22
The purpose of this study was to compare the stiff-
ness and strength of the native UCL with the char-
acteristics of the ligament reconstruction techniques
described by Strandell,
11
Osterman et al,
12
Fairhurst
and Hansen,
13
and a modication of the procedure by
Glickel et al.
14
We hypothesized that none of the
ligament reconstructions would approximate the
stiffness and strength of the native UCL.
Materials and Methods
Eleven fresh-frozen cadaver thumbs were disarticu-
lated through the carpometacarpal and interphalan-
geal joints, preserving only the collateral ligaments,
volar plate, and extensor pollicis brevis tendon. The
metacarpal was potted into a loading jig with poly-
methylmethacrylate cement to allow for testing.
Once mounted the MCP joint was exed to 30 as
measured by a hand-held goniometer to isolate the
proper collateral ligament.
A radial deviation force oriented perpendicular to
the long axis of the proximal phalanx in the antero-
posterior and lateral planes was applied at a rate of 20
mm/s until UCL failure occurred. For each trial the
applied force and displacement of the proximal pha-
lanx were recorded continuously. Before loading the
distance between the applied force and the edge of
the proximal phalanx was measured using calipers.
This distance, and the fact that the load was applied
perpendicular to the bone, allowed us to calculate the
angle at which UCL failure occurred, the moment at
failure, and the rotational stiffness of the ligament.
Visualization of the specimen throughout testing
conrmed that additional exion or extension of the
MCP joint and/or rotation of the specimen did not
occur.
The UCL reconstructions described in the litera-
ture not only vary in the number and location of bone
tunnels but also in the orientation of the tendon graft.
Appropriate graft material was obtained by harvest-
ing the exor digitorum supercialis and exor digi-
torum profundus tendons from each cadaver. To min-
imize the variability that would result from
differences in graft caliber or bone tunnel dimensions
these were standardized for all specimens. Each ten-
don was passed through a series of calibrated rings
and selected for use only if it measured between 3
and 4 mm in diameter. The bone tunnels in the
proximal phalanx and metacarpal were drilled se-
quentially from 2.5 mm to a nal diameter of 4 mm.
Further standardization was achieved by securing all
of the reconstructions with the same number of 3-0
sutures (Ethibond, Johnson & Johnson, Piscataway,
NJ).
Four separate static ligament reconstructions were
performed. Three of these used a tendon graft woven
through 2 holes in the proximal phalanx and 2 holes
in the metacarpal. The described procedures differed
only in the orientation of the tendon graft. The Stran-
dell
11
reconstruction placed a free tendon graft in a
gure-of-eight pattern through the bone tunnels (Fig.
1). Osterman et al
12
modied this reconstruction by
passing a third limb of the graft parallel to the edge
of the volar plate and suturing the ends of the tendon
graft to the volar plate (Fig. 2). Fairhurst and Han-
sen
13
used a distally attached extensor pollicis brevis
tendon looped though the bone tunnels in a triangular
fashion (Fig. 3).
Figure 1. Strandell
11
reconstruction. The free tendon graft is
passed in gure-of-8 fashion through bone tunnels and su-
tured to itself.
Hogan, Ruland, and Levin / Ulnar Collateral Ligament Reconstruction 395
Each of these procedures was performed sequen-
tially on 4 specimens. Care was taken to excise only
the proper and accessory collateral ligaments and
leave the volar plate intact. The order of the recon-
struction was varied on each individual thumb to
ensure that differences in strength and stiffness were
not caused by the effects of cumulative testing.
The remaining ligament reconstruction used 3 drill
holes, reproducing the origin and insertion of the
proper and accessory collateral ligaments. In 1993
Glickel et al
14
described a procedure in which a
tendon graft was pulled through the phalangeal holes
and advanced into a single metacarpal tunnel (Fig. 4).
A modication of this procedure in which the free
ends of the tendon graft are tied into a square knot on
the radial side of the metacarpal (Fig. 5) provides
stout irretrievable interference xation without the
shortcomings that accompany transdermal xation.
This reconstruction was performed on 5 specimens.
Statistical Analysis
The moment at failure, angle at failure, and stiffness
at failure for the native UCL and for each reconstruc-
tion were calculated and subjected to analysis of
variance single-factor testing and Tukey testing. A p
value of less than .05 was considered signicant.
Results
Biomechanical testing of the native UCL revealed an
average failure moment of 2,311.7 N-mm 1,152.8
and an average rotational stiffness of 76.7 N-mm/
57.3. Failure occurred at an average of 33 of radial
deviation. During this portion of the experiment 2
specimens failed by fracture through the metacarpal
head, preventing any further testing.
The results of testing the static UCL reconstruc-
tions are shown in Table 1. All of the reconstructions
as described by Strandell,
11
Osterman et al,
12
and
Fairhurst and Hansen
13
failed by suture pull-through
of the tendon, with the exception of one specimen for
the Osterman et al
12
reconstruction that failed by
fracture through the bone holes. All of the specimens
that used interference knot xation failed by fracture
through the bone tunnels in the proximal phalanx.
This modication of the procedure by Glickel et al
14
had a signicantly higher moment at failure (924.7
N-mm 363.4 vs 176.7 N-mm 61.2, p .00001)
Figure 2. Osterman et al
12
reconstruction. The free tendon
graft is passed in a gure-of-8 fashion and sutured to the volar
plate.
Figure 3. Fairhurst and Hansen
13
reconstruction. The exten-
sor pollicis brevis tendon is left attached at its insertion and
passed from dorsal to volar through the proximal phalanx,
then dorsal to volar through the metacarpal and sutured to
itself.
Figure 4. Modied Glickel et al
14
reconstruction. The free
tendon graft is passed dorsal to volar through a tunnel in the
proximal phalanx and both limbs are passed through a single
metacarpal tunnel.
Figure 5. Interference xation. The tendon graft is tied into a
square knot on the radial aspect of the metacarpal.
396 The Journal of Hand Surgery / Vol. 30A No. 2 March 2005
and was signicantly stiffer (22.6 N-mm/ 15.1 vs
4.4 N-mm/ 2.2, p .0005) than any of the other
reconstructions. There were no signicant differ-
ences in the angle at failure for any of the surgeries
tested.
None of the reconstructions approached the
strength and stiffness of the native UCL. Statistical
analysis of the data showed that serial testing of the
specimens did not have any signicant effect on the
moment at failure, rotational stiffness, angle at fail-
ure, or mode of failure of each reconstruction.
Discussion
Incomplete tears and acutely repaired complete rup-
tures of the UCL of the thumb MCP joint almost
universally have excellent clinical outcomes. In con-
trast patients with neglected injuries that result in
laxity of the UCL do not receive the same favorable
prognosis. Loss of ulnar-sided stability compromises
lateral pinch, distorts the mechanics of the MCP
joint, and predisposes the thumb to premature degen-
erative arthritis. Despite improved physician aware-
ness and renements in the physical examination,
chronic UCL insufciency continues to challenge the
orthopedic surgeon.
In an effort to restore stability, numerous inge-
nious reconstructions have been described. These
surgical procedures vary in the number and location
of bone tunnels, as well as the choice, orientation,
and xation of the tendon graft. The ideal surgery
would reproduce the anatomic origin and insertion of
the native ligament, have sufcient strength to permit
early range of motion, and be technically easy to
perform.
A review of the literature shows that the use of
local tissues to restore ulnar-sided stability has led to
mixed results.
1,2,1121
In contrast to the predictable
recovery of function after acute repair, the poorer
tissue quality seen in chronic injuries leads to a
higher frequency of recurrent instability. Although
small numbers have been reported, the long-term
integrity is restored in 50%
1
to 75%,
19
with dimin-
ished recovery of pinch strength. Controversy re-
mains, however, because other investigators state no
difference between acute and chronic repairs using
local tissue.
13,17
These uncertainties have led to the description of
various ligament reconstructions using dynamic,
static, or a combination of both dynamic and static
restraints to prevent abnormal joint motion. Dynamic
reconstructions use tendon advancement (adductor
pollicis)
12,13,16
or tendon transfer (extensor indicis,
extensor pollicis longus, or extensor pollicis bre-
vis)
18
to increase the adduction moment on the
thumb proximal phalanx during pinch, thereby re-
storing joint stability. The results from these proce-
dures are variable, with some reports approaching the
favorable results seen after acute repairs.
12,13,16,18
This dynamic contribution, however, is difcult to
reproduce accurately in the cadaver model so these
methods were not evaluated in our study.
Many innovative methods for static ligament
reconstruction of chronic UCL-decient thumbs
have been described. They all use tendon graft,
usually the palmaris longus, passing through bone
tunnels in the proximal phalanx and the metacar-
pal. Clinical results after these reconstructions
range from 70% to 92% restoration of joint stabil-
ity and strength, with an average loss in motion of
20 at the MCP joint.
1,1115,20,21
Most of these reconstructive techniques, however,
use tissue congurations that do not mimic the native
anatomy and rely on xation that does not allow
early motion of the MCP joint. Biomechanical stud-
ies of acute UCL repair show that nonanatomic res-
toration of the phalangeal insertion leads to restricted
MCP exion.
23
In addition most surgeons, keenly
aware that a stiff MCP joint usually is well tolerated,
are willing to relinquish motion to maintain stability.
Therefore the graft is protected with several weeks of
immobilization and occasional K-wire xation of the
MCP joint. These 2 factors may contribute to the
decit in motion that is seen clinically.
Table 1. Loading Characteristics for Native UCL and Ligament Reconstructions
Procedure
Failure Moment
N-mm
Rotational Stiffness
N-mm/ Angle at Failure ()
Native UCL (N 11) 2,311.7 1,152.8 76.7 57 33 9
Strandell
11
(N 4) 235.6 18.2 5.6 2 43 10
Osterman et al
12
(N 4) 160.2 76.7 4.2 3 43 13
Fairhurst and Hansen
13
(N 4) 134.1 5.9 3.5 2 44 18
Interference xation (N 5) 924.7 363.4 22.6 15 42 13
Hogan, Ruland, and Levin / Ulnar Collateral Ligament Reconstruction 397
In an attempt to restore the normal anatomy more
accurately, Glickel et al
14
described a procedure in
which the tendon graft was advanced through drill
holes located at the origin and insertion of the proper
and accessory collateral ligaments. In this recon-
struction the free tendon ends were pulled through
the skin and tied over holes in a large button on the
radial aspect of the thumb.
Although this procedure follows the normal ana-
tomic course of the thumb ligaments and the graft is
xed securely, the stiffness of the construct poten-
tially is compromised by the intervening soft tissue
between the button and the bone. In addition there is
a possible increase in the risk for infection resulting
from the breach in the normal skin barrier by the
tendon graft and from possible skin breakdown from
prolonged pressure from the cutaneous button. The
modication described here uses an interference knot
as the sole means of graft xation. Other reports
document the use of a knot in conjunction with
buttons, rubber tubing, or suture anchors.
24
One potential limitation of this study was the
choice of a 3- to 4-mm exor tendon as the tendon
graft. Although the standardization of tendon diam-
eter allows for comparisons to be made between the
different techniques, the dimensions of this graft may
not accurately reect the size of the palmaris longus
tendon usually used for these reconstructions. Each
of the nonanatomic reconstructions, however, failed
by suture pull-through and not tendon rupture, indi-
cating that xation of the graft was the weakest link.
For the anatomic reconstruction with interference
knot xation, failure consistently occurred by frac-
ture through the bone tunnels. This suggests that the
actual strength and stiffness of this procedure likely
would be greater than the values seen in this study
because a smaller tendon would require smaller-
diameter bone tunnels.
This experiment shows that anatomic graft place-
ment with interference knot xation results in signif-
icantly greater initial strength and stiffness than non-
anatomic reconstructions. The experimental design
mimics the graft characteristics that would be seen
immediately after surgery, before healing into the
bone tunnels or periosteum could take place. With
healing, alterations in the stiffness and strength of the
reconstructions undoubtedly occur, but these changes
cannot be determined by using a cadaver model.
Previous work by Firoozbakhsh et al
22
showed
that acute repair using suture anchors could with-
stand only 32% of the load of the intact ligament.
Despite these inferior characteristics, a clinical series
in which an early mobilization protocol was used
after acute ligament repair resulted in improved mo-
tion without any cases of recurrent joint instability.
25
These results suggest that this modication of the
reconstruction by Glickel et al
14
may provide suf-
cient stability to allow for early motion at the MCP
joint and minimize the need for prolonged casting.
By decreasing perioperative immobilization time and
duplicating the normal anatomy of the native true and
accessory UCL, improved MCP motion may result.
The authors acknowledge the assistance of Roger Nightengale and
Danielle Ottaviano for technical support and statistical analysis.
References
1. Smith RJ. Post-traumatic instability of the metacarpopha-
langeal joint of the thumb. J Bone Joint Surg 1977;59A:
1421.
2. Coonrad RW, Goldner JL. A study of the pathological nd-
ings and treatment in soft-tissue injury of the thumb meta-
carpophalangeal joint. With a clinical study of the normal
range of motion in one thousand thumbs and a study of post
mortem ndings of ligamentous structures in relation to
function. J Bone Joint Surg 1968;50A:439451.
3. Bowers WH, Hurst LC. Gamekeepers thumb. Evaluation by
arthrography and stress roentgenography. J Bone Joint Surg
1977;59A:519524.
4. Bostock S, Morris MA. The range of motion of the MP joint
of the thumb following operative repair of the ulnar collat-
eral ligament. J Hand Surg 1993;18B:710711.
5. Kozin SH. Treatment of thumb ulnar collateral ligament
ruptures with the Mitek bone anchor. Ann Plast Surg 1995;
35:15.
6. Weiland AJ, Berner SH, Hotchkiss RN, McCormack RR Jr,
Gerwin M. Repair of acute ulnar collateral ligament injuries
of the thumb metacarpophalangeal joint with an intraosseous
suture anchor. J Hand Surg 1997;22A:585591.
7. Downey DJ, Moneim MS, Omer GE Jr. Acute gamekeepers
thumb. Quantitative outcome of surgical repair. Am J Sports
Med 1995;23:222226.
8. Derkash RS, Matyas JR, Weaver JK, Oden RR, Kirk RE,
Freeman JR, et al. Acute surgical repair of the skiers thumb.
Clin Orthop 1987;21:2933.
9. Saetta JP, Phair IC, Quinton DN. Ulnar collateral ligament
repair of the metacarpo-phalangeal joint of the thumb: a
study comparing two methods of repair. J Hand Surg 1992;
17B:160163.
10. Gerber C, Senn E, Matter P. Skiers thumb. Surgical treat-
ment of recent injuries to the ulnar collateral ligament of the
thumbs metacarpophalangeal joint. Am J Sports Med 1981;
9:171177.
11. Strandell G. Total rupture of the ulnar collateral ligament of
the metacarpophalangeal joint of the thumb. Acta Chir
Scand 1959;118:7280.
12. Osterman AL, Hayken GD, Bora FW Jr. A quantitative
evaluation of thumb function after ulnar collateral repair and
reconstruction. J Trauma 1981;21:851861.
13. Fairhurst M, Hansen L. Treatment of gamekeepers thumb
398 The Journal of Hand Surgery / Vol. 30A No. 2 March 2005
by reconstruction of the ulnar collateral ligament. J Hand
Surg 2002;27B:542545.
14. Glickel SZ, Malerich M, Pearce SM, Littler W. Ligament
replacement for chronic instability of the ulnar collateral
ligament of the metacarpophalangeal joint of the thumb.
J Hand Surg 1993;18A:930941.
15. Breek JC, Tan AM, Van Thiel TPH, Daantje CRE. Free
tendon grafting to repair the metacarpophalangeal joint of
the thumb. J Bone Joint Surg 1989;71B:383387.
16. Neviaser RJ, Wilson JN, Lievano A. Rupture of the ulnar
collateral ligament of the thumb (gamekeepers thumb).
J Bone Joint Surg 1971;53A:13571364.
17. Stener B. Displacement of the ruptured ulnar collateral lig-
ament of the metacarpo-phalangeal joint of the thumb. A
clinical and anatomical study. J Bone Joint Surg 1962;44B:
869879.
18. Sakellarides HT. Treatment of recent and old injuries of the
ulnar collateral ligament of the MP joint of the thumb. Am J
Sports Med 1978;6:255262.
19. Parikh M, Nahigian S, Froimson A. Gamekeepers thumb.
Plast Reconstr Surg 1976;58:2431.
20. Mitsionis GI, Varitimidis SE, Sotereanos GG. Treatment of
chronic injuries of the ulnar collateral ligament of the thumb
using a free tendon graft and bone suture anchors. J Hand
Surg 2000;25B:208211.
21. Melone CP Jr, Beldner S, Basuk RS. Thumb collateral
ligament injuries. An anatomic basis for treatment. Hand
Clin 2000;16:345357.
22. Firoozbakhsh K, Yi IS, Moneim MG, Umada Y. A study of
ulnar collateral ligament of the thumb metacarpophalangeal
joint. Clin Orthop 2002;403:240247.
23. Bean CHG, Tencer AF, Trumble TE. The effect of thumb
metacarpophalangeal ulnar collateral ligament attachment
site on joint range of motion: An in vitro study. J Hand Surg
1999;24A:283287.
24. Glickel SZ, Barron OA, Eaton RG. Dislocation and ligament
injuries in the digits. In: Green DR, Hotchkiss RN, Pederson
WC, eds. Greens Operative Hand Surgery. 4th ed. Churchill
Livingstone, 1999; 792794.
25. Lane LB. Acute grade III ulnar collateral ligament ruptures.
A new surgical and rehabilitation protocol. Am J Sports Med
1991;19:234238.
Hogan, Ruland, and Levin / Ulnar Collateral Ligament Reconstruction 399