SCIENTIFIC ARTICLE

Biomechanics of the Acute Boutonniere

Deformity
Luis Grau, MD,* Hasan Baydoun, MD,‡ Kevin Chen, MD,‡ Seth T. Sankary, BA,†
Farid Amirouche, PhD,‡ Mark H. Gonzalez, MD, PhD‡

Purpose To demonstrate which structures of the extensor mechanism create a boutonniere

deformity, when damaged, in a cadaver model. An analysis of how damage to these anatomical
structures affects the biomechanical performance of the extensor mechanism was also performed.
Methods We secured 18 fresh cadaveric hands onto an apparatus consisting of a computer-
controlled motor and tensiometer attached in series to the extensor communis tendon of the
ring and middle digits. The central slip, transverse, and oblique fibers of the interosseous hood
and the triangular ligament were sequentially divided. After each structure was divided, the
motors were activated to provide a constant tendon displacement force. The angular displace-
ment at the proximal interphalangeal (PIP) and distal interphalangeal joints was recorded.
Results In all digits, detachment of the central slip from the middle phalanx produced a decrease
in extension of the PIP joint. When the transverse and oblique fibers of the interosseous hood
were also divided, extension at the PIP joint was further decreased. A boutonniere deformity
occurred only when all 3 structures were damaged.
Conclusions The boutonniere deformity requires subluxation of the lateral bands volar to the
axis of rotation of the PIP joint. This study demonstrates that damage to the central slip alone
does not cause the deformity. Combined injury of the central slip, triangular ligament, and
transverse and oblique fibers of the interosseous hood causes a boutonniere deformity.
Clinical relevance Division of the central slip leads to loss of extension at the PIP joint. A more
substantial loss of extension after injury or development of a boutonniere deformity should
alert clinicians that other structures of the extensor mechanism are also damaged. (J Hand
Surg Am. 2017;-(-):1.e1-e6. Copyright Ó 2017 by the American Society for Surgery of the
Hand. All rights reserved.)
Key words Boutonniere, biomechanics, hand, finger, acute.

T
HE BOUTONNIERE DEFORMITY IS characterized by
flexion at the proximal interphalangeal (PIP)
From the *Department of Orthopaedics, †University of Miami Miller School of Medicine, joint with hyperextension at the distal inter-
Miami, FL; and the ‡Department of Orthopaedics, University of Illinois College of Medicine,
Chicago, IL. phalangeal (DIP) joint. This occurs as a result of lateral
Received for publication October 23, 2013; accepted in revised form July 12, 2017.
band (LB) subluxation below the axis of rotation of the
PIP joint, where the LBs function as paradoxical
No benefits in any form have been received or will be received related directly or
indirectly to the benefits of this article. flexors. The deformity may be caused by open or
Corresponding author: Mark H. Gonzalez, MD, PhD, Department of Orthopaedics, closed traumatic injuries of the central slip (CS),1e4
University of Miami Miller School of Medicine, PO Box 016960 (D-27), Miami, FL 33101; rheumatoid arthritis,5 or a congenital abnormality.6
e-mail: hiphand15@gmail.com. The current study focused on traumatic injuries.
0363-5023/17/---0001$36.00/0 Using a cadaveric model, Grundberg and Reagan7
http://dx.doi.org/10.1016/j.jhsa.2017.07.011
found that a surgically created central slip laceration

Ó 2017 ASSH r Published by Elsevier, Inc. All rights reserved. r 1.e1

1.e2 ACUTE BOUTONNIERE DEFORMITY
did not cause subluxation of the LB and an acute
boutonniere deformity to occur. Another study by
Mercer et al8 showed that contributions from the LB
largely governed PIP extension. They reported that
extensor mechanism repair with the LBs placed in a
dorsal position, without reattachment of the central slip,
prevented the development of a boutonniere deformity.
The purposes of our study were to create a
reproducible model for boutonniere deformity in a
cadaver and to analyze quantitatively how sequential
damage to key anatomical structures of the extensor
mechanism affects the biomechanics of extension.
We hypothesized that damage to the CS, triangular
ligament, and transverse and oblique fibers of the
interosseous hood running between the central slip
and the LBs is necessary to create an acute bouton-
niere deformity.9 We also hypothesized that damage
to each of these structures decreases the biomechan-
ical efficiency of the extensor mechanism leading to
decreased PIP extension.
MATERIALS AND METHODS
Anatomic study
We thawed 10 fresh cadaveric hands in normal saline,
sectioned them at the distal third of the forearm, and
fixed them palmar side down on the test apparatus. We
randomly selected and used either the middle or ring
digit in each cadaveric hand and excluded digits with
poor passive range of motion. The ring and the middle
digits are more commonly injured than other digits,
accounting for more than 50% of the cases in a large
case series.10 We elected against using the index and
little digits because of contributions of the extensor
indicis and extensor digiti quinti, respectively. Skin
and subcutaneous tissue were removed from the digits
tested. We then dissected and identified the CS, LB,
transverse and oblique fibers of the interosseous hood,
and the triangular ligament (Fig. 1A).
For all 10 digits we divided the CS at its insertion to
the base of the middle phalanx. In group A (5 digits in
total), we sequentially lacerated the triangular liga-
ment and the transverse and oblique fibers of the
interosseous hood running between the CS and LB. In
group B (5 digits in total), we reversed the sequence,
lacerating the transverse and oblique fibers of the
interosseous hood before the triangular ligament.
Figure 1A is a schematic of the anatomy of the
extensor mechanism; Figure 1B depicts the sequence
of structures damaged in groups A and B.
Based on previous descriptions in the literature,
we simulated the action of the flexors and intrinsic
muscles by applying loads to the superficial (100 g)
J Hand Surg Am.
and deep (100 g) flexor tendons as well as the radial
(400 g) and ulnar (400 g) interosseous tendons by
sutures through a low-friction pulley system.11 The
extensor communis tendon was then sutured to a
computer-controlled, winch-type servomotor and
tension was varied to obtain multiple initial flexion
angles at the PIP joint. A constant tendon displacement
force of 30 mm/min was then applied. Consistent
with physiological loads described in the literature,
the force never exceeded 20 N.12,13 Two separate
investigators independently used goniometers to
measure the relative flexion-extension at the PIP and
DIP joints after extensor tendon excursion.
Biomechanical study
Eight additional intact, fresh cadaveric hands were
thawed in normal saline, sectioned at the distal third of
the forearm and fixed palmar side down on the test
apparatus. After randomly selecting either the ring or
middle digit, we made a midline dorsal incision over
each digit beginning 1 cm distal to the DIP joint to 4
cm proximal to the metacarpal head. We dissected and
identified the CS, LB, transverse and oblique fibers of
the interosseous hood and the triangular ligament
(Fig. 1C). We then sutured the extensor communis
tendon to the same computer-controlled, winch-type
servomotor with a tensiometer connected in series,
4 cm proximal to the metacarpal head. To measure
angular displacement of the PIP relative to the starting
position of 35 flexion, we used a potentiometer,
which acted as a rheostat (Fig. 2). The potentiometer
was secured to the soft tissues of the PIP joint by
sutures (Fig. 2). As rotation at the PIP joint rotated the
potentiometer, a precalibrated angular displacement
was displayed and recorded.
Four total configurations were tested for each digit.
In configuration 1, all structures remained intact. In
configuration 2, the CS was detached from its inser-
tion on the middle phalanx. In configuration 3, the CS
was detached and the transverse and oblique fibers of
the interosseous hood were sectioned on either side of
the CS. In configuration 4, the triangular ligament
was sectioned longitudinally, the CS was detached,
and the transverse and oblique fibers of the inter-
osseous hood were sectioned on both sides of the CS.
Figure 1A is a schematic of the anatomy of the
extensor mechanism; Figure 1C depicts the sequence
of structures divided in each of the configurations.
Using the same apparatus as before, a computer
recorded data from the potentiometers and a
modified version of LabVIEW Software (National
Instruments Corporation, Austin, TX) calculated
angular displacement at the PIP joint.13 The data
r Vol. -, - 2017
 ACUTE BOUTONNIERE DEFORMITY 1.e3

FIGURE 1: A Schematic representation of the anatomy of the extensor mechanism. B Schematic representation of the sequence of
lesioned structures in each group for the anatomic portion of the experiment. C Schematic representation of the sequence of lesioned
structures in each group for the biomechanical portion of the experiment.

FIGURE 3: At initial PIP flexion angles of 35 and greater, the

LBs were observed to subluxate below the axis of rotation of the
PIP joint when the dorsal tethering structures were damaged and
paradoxical flexion occurred with excursion of the extensor
tendon.

deformity and extension at the PIP and DIP joints

FIGURE 2: Cadaver hand affixed to the test apparatus. The
extensor communis tendon was sutured to a computer-controlled,
winch-type servomotor and tensiometer in series. A constant
tendon displacement force of 30 mm/min was applied. A
potentiometer was sewn centered over the PIP joint to measure
angular displacement.
points were plotted against time at 1/1,000-second
intervals.
RESULTS
Anatomic Study
Detaching the CS from its insertion on the middle
phalanx was not sufficient to create a boutonniere
was observed in all digits. Similarly, neither damage
to the CS and triangular ligament in group A nor
damage to the CS and transverse and oblique fibers of
the interosseous hood in group B was sufficient to
create a boutonniere deformity. All 10 digits exhibi-
ted extension at the PIP and DIP joints. Only after
damage to the CS, triangular ligament, and transverse
and oblique fibers of the interosseous hood was a
boutonniere deformity created. Excursion of the
extensor tendon led to PIP joint flexion and DIP joint
hyperextension in all digits. At initial PIP flexion
angles of 35 and greater, the LBs were observed to
subluxate below the axis of rotation of the PIP joint
when the dorsal tethering structures were damaged
and paradoxical flexion occurred with excursion of
the extensor tendon (Fig. 3).

J Hand Surg Am. r Vol. -, - 2017

1.e4 ACUTE BOUTONNIERE DEFORMITY

FIGURE 4: Diagram representing the degree of relative motion at the PIP in each configuration after excursion of the extensor tendon. In
configurations 2 and 3 there was decreased extension at the PIP joint compared with the intact digit. The negative values in configuration
4 represent how damage to the CS, the oblique and transverse fibers of the interosseous hood, and the triangular ligament led to relative
flexion of the PIP joint with excursion of the extensor tendon.

Biomechanical study extensor hood.9,14 When the triangular ligament is

The PIP joint behaved similarly in all digits after
damage to structures of the extensor mechanism
(Fig. 4). Extension at the PIP joint was greatest in the
intact digit 17.1  7.9 . Detachment of the CS from
the middle phalanx resulted in an additional exten-
sion lag at the PIP joint of 2.4  1.3 (0.14  0.08),
compared with the intact phalanx (Table 1). Division
of the transverse and oblique fibers of the inteross-
eous hood and the CS caused extension at the PIP
joint to decrease further by 4.1  1.5 (0.24  0.07).
As demonstrated in our anatomic study, paradoxical
flexion at the PIP joint was achieved only after disrup-
tion of the CS and triangular ligament, and of the
transverse and oblique fibers of the interosseous hood
with 12.1  4.1 of flexion relative to neutral position
(Fig. 5). Negative values in Figure 4 and Table 1 indi-
cate flexion at the PIP joint relative to the initial starting
position after excursion of the extensor tendon.
DISCUSSION
Detachment of the CS eliminates the extension
moment it exerts on the PIP joint.9 When intact, the
triangular ligament functions to support the LB in a
dorsal position relative to the PIP joint even when
there is disruption to the LBs’ attachments to the
also damaged, the LB migrates below the axis of
rotation of the PIP joint.9,14 Flexion deformity of the
PIP joint is initiated by the imbalanced pull of the
flexor digitorum superficialis tendon in the setting
of an incompetent CS and volarly subluxated LBs.
Through the LBs in their volarly displaced position,
the pull of the intrinsics and the extensor digitorum
further augment the flexion force in the PIP joint.
With the PIP joint in flexion, decreased tension in the
deep flexor tendon leads to DIP hyperextension
owing to transmission of the intrinsic and extrinsic
muscle force through the LBs and onto their insertion
on the dorsal aspect of the distal phalanx.
We expected change in angle over time at the PIP
joint between the digits we tested. We used different-
sized, fresh-frozen cadaver hands with variable de-
grees of inherent stiffness and placed them under
similar loads. Despite this, our study demonstrated a
decrease in extension, because each structure was
sequentially damaged and exhibited paradoxical
flexion at the PIP joint characteristic of boutonniere
deformity when all structures were damaged. This
was consistent across all digits.
Results from the anatomic portion of the experi-
ment support the idea that even with loading of the
intrinsic muscles, all 3 key structures consistently

J Hand Surg Am. r Vol. -, - 2017

 ACUTE BOUTONNIERE DEFORMITY 1.e5

TABLE 1. Relative Loss of Extension at PIP Joint Compared With Intact Digit
Hand CS CS and Interosseous Fibers CS, Interosseous Fibers, and Triangular Ligament

1 3.7 (0.30) 4.4 (0.36) e20.8 (1.72)

 
2 1.5 (0.10) 2.3 (0.16) e35.3 (2.37)
3 2.1 (0.14) 4.9 (0.34) e24.7 (1.71)
 
4 1.5 (0.14) 2.7 (0.24) e23.2 (2.13)
5 1.2 (0.07) 5.3 (0.30) e32.7 (1.83)
 
6 3.2 (0.09) 5.7 (0.17) e46.9 (1.39)
7 4.8 (0.21) 5.7 (0.25) e32.6 (1.43)
 
8 1.2 (0.12) 2.1 (0.21) e17.3 (1.79)
Average 2.4 (0.14) 4.1 (0.24) e29.2 (1.17)
 
SD 1.3 (0.08) 1.5 (0.07) 9.6 (0.33)

The table shows the absolute difference in degrees from intact and the relative difference from intact (intact e trial) / intact) in parentheses. The table
format represents the absolute and (fractional) loss of PIP joint extension compared with the intact configuration in all digits tested. Negative angles
and (values greater than 1) in parentheses indicate paradoxical flexion of the PIP joint with excursion of the extensor communis tendon.

inability to reach full extension in the intact digits. For

FIGURE 5: With excursion of the common extensor tendon, a
characteristic boutonniere deformity was observed in all digits
when the CS, triangular ligament, and transverse and oblique
fibers of the interosseous hood were damaged.
needed to be divided to create the boutonniere defor-
mity. We elected against intrinsic muscle loading in
the biomechanic portion of our study to simplify the
number of variables in our model, and because of the
limitations of the experimental apparatus. Additional
incisions and loading of the intrinsics through a pulley
system would have interfered with accurate angle
measurements by the potentiometer sewn onto the
skin. The volarly directed force of the intrinsics likely
contributed to subluxation of the LBs, but only
when the dorsal tethering structures were damaged.9
Another observation of the experiment was the
full extension of the PIP joint, a delicate balance is
required between intrinsic and extrinsic muscle force
and metacarpophalangeal joint position.9,14,15 The
absence of intrinsic function in our model may have
contributed to these findings.
Although damage to the CS is necessary, our
findings are consistent with other studies that sug-
gested that it is insufficient to create a boutonniere
deformity in an acute setting.7,15,16 Mercer et al8 re-
ported that failing to reinsert the CS does not
predispose to a boutonniere deformity. Our results
support their conclusion and show that even in the
setting of CS detachment, a boutonniere deformity
would not ensue if the triangular ligament or the at-
tachments between the LB and the CS were intact.
However, CS detachment may lead to increased
tension on the dorsal tethering structures, which in
turn may gradually predispose to failure of the
triangular ligament and oblique and interosseous
fibers of the extensor hood.9 This finding is supported
by the clinical literature in which an early CS injury
is associated with localized swelling and mild
extensor lag, and subsequently a boutonniere defor-
mity develops after 7 to 14 days or longer.17
Division of the CS created an extensor lag of 14%.
Division of the CS and the transverse and oblique
fibers created a marked increase in the extensor lag
(24%) but did not produce paradoxical flexion. Once
the triangular ligament was also released, paradoxical
flexion was noted in all digits tested. Our study
confirms that division of the CS, the transverse and
oblique fibers of the interosseous hood, and the trian-
gular ligament reproducibly creates a boutonniere

J Hand Surg Am. r Vol. -, - 2017

1.e6 ACUTE BOUTONNIERE DEFORMITY
deformity. Furthermore, it confirms that a combination
of the CS and transverse and oblique fibers or the
central slip and the triangular ligament were not
sufficient to create a boutonniere deformity, but it
decreased the efficiency of the extensor mechanism.
Our model emphasizes that a closed injury to the
extensor apparatus of the digit can show a variable
loss of extensor function based on the number of
structures affected. An isolated CS injury produced
only a modest loss of extension in our study. Pre-
sumably, this is why closed injuries to the CS can be
missed initially. Clinicians should remain aware that
even a mild degree of extensor lag can represent a
complete rupture of the CS. Conversely, a closed
injury to the extensor apparatus of the digit with a
marked extensor lag should alert the clinician of an
injury to other structures of the extensor mechanism
in addition to the CS.
Despite several proposed mechanisms for
boutonniere deformity in hand surgery textbooks,
there are only a few studies specifically describing
progression of the boutonniere deformity after a CS
injury. Moreover, the conclusions of these studies are
inconsistent regarding which structures must be
damaged to create the deformity, and their purpose
was not to demonstrate reproducibly how damage to
the structures affects extensor mechanics or to vali-
date an anatomic model for creating the deformity.
The current study provides a quantitative approach to
understanding progression of the boutonniere defor-
mity and validates many previous descriptions in the
literature, proposed pathomechanics and years of
clinical observations.4,7-9,16,18 The results of our
study also stress the significance of the transverse and
oblique fibers of the interosseous hood as dorsal
tethering structures and their role in developing the
deformity, which are seldom emphasized in hand
surgery textbooks or scientific articles.16
Our findings describe a reproducible model of
boutonniere deformity and of extensor mechanism
mechanics that can be used to predict and test repairs
and reconstructions in future studies. Our biome-
chanical data suggest that if surgical intervention is
needed for a passively correctible boutonniere
deformity, reattachment of the CS as well as repair or
J Hand Surg Am.
reconstruction of the LBs in an anatomic position
dorsal to the center of rotation of the PIP should be
performed to avoid extensor lag and progression of
the deformity. Physicians can use the degree of
extensor lag and deformity found on physical ex-
amination in their surgical planning for repair or
reconstruction.
REFERENCES
1. Kozin SH, Nissenbaum M, Berlet AC. Surgical treatment of
boutonniere deformity. Oper Techn Orthop. 1993;3(4):313e317.
2. Imatami J, Hashizume H, Wake H, Morito Y, Inoue H. The central
slip attachment fracture. J Hand Surg Br. 1997;22(1):107e109.
3. Souter WA. The boutonniere deformity: a review of 101 patients with
division of the central slip of the extensor expansion of the fingers.
J Bone Joint Surg Br. 1967;49B:710e721.
4. Littler JW, Eaton RG. Redistribution of forces in correction of
boutonniere deformity. J Bone Joint Surg Am. 1967;49(7):
1267e1274.
5. Ferlic DC. Boutonniere deformities in rheumatoid arthritis. Hand
Clinic. 1989;5(2):215e222.
6. Carneiro RS. Congenital attenuation of the extensor tendon central
slip. J Hand Surg Am. 1993;18(6):1004e1007.
7. Grundberg AB, Reagan DS. Central slip tenotomy for chronic mallet
finger deformity. J Hand Surg Am. 1987;12(4):545e547.
8. Mercer D, Fitzpatrick J, Firoozbakhsh K, Carvalho A, Moneim M.
Extensor tendon repair with and without central slip reattachment to
bone: a biomechanical study. J Hand Surg Am. 2009;34(1):108e111.
9. Smith RJ. Balance and kinetics of the fingers under normal and
pathological conditions. Clin Orthop Relat Res. 1974;(104):92e111.
10. Le Bellec Y, Loy S, Touam C, Alnot JY, Masmejean E. Surgical
treatment for boutonniere deformity of the fingers: retrospective
study of 47 patients. Chir Main. 2001;20(5):362e367.
11. Klasson SC, Adams BD. Biomechanical evaluation of chronic
boutonniere reconstructions. J Hand Surg. 1992;17(5):868e874.
12. Qian K, Traylor K, Lee SW, Ellis B, Weiss J, Kamper D. Mechanical
properties vary for different regions of the finger extensor apparatus.
J Biomech. 2014;47(12):3094e3099.
13. Chow JC, Sensinger J, McNeal D, Chow B, Amirouche F,
Gonzalez M. Importance of proximal A2 and A4 pulleys to main-
taining kinematics in the hand: a biomechanical study. Hand.
2014;9(1):105e111.
14. Williams K, Terrono AL. Treatment of boutonniere finger deformity
in rheumatoid arthritis. J Hand Surg Am. 2011;36(8):1388e1393.
15. Harris C, Rutledge GL. The functional anatomy of the extensor
mechanism of the finger. J Bone Hand Surg. 1972;54(4):713e726.
16. Rubin J, Bozentka DJ, Bora FW. Diagnosis of closed central slip
injuries: a cadaveric analysis of non-invasive tests. J Hand Surg Br.
1996;21(5):614e616.
17. Clavero JA, Golanó P, Fariñas O, Alomar X, Monill JM, Esplugas M.
Extensor mechanism of the fingers: MR imagingeanatomic corre-
lation. Radiographics. 2003;23(3):593e611.
18. Fox A, Kang N. Reinserting the central slip—a novel method for
treating boutonniere deformity in rheumatoid arthritis. J Plast
Reconstr Aesthet Surg. 2009;62(5):e91ee92.
r Vol. -, - 2017