Phẫu Thuật Gân Bàn Tay (Tendon Surgery of the Hand)

388
Section 4:  Extensor Tendon Repair and Reconstruction
reconstruction has a high incidence of deformity recur­
rence, it may be indicated as substantial functional
improvements can be achieved.10 EPL tendon rerouting
coupled with synovectomy of the MCP joint will lead
to an increased extensor moment at the MCP joint via
the dorsal MCP joint capsular attachment of the EPL.9
In moderate type I deformities, a fixed MCP joint
deformity is present. Appropriate planning includes
assessment of the CMC and IP joints. The relative condi­
tion of these adjacent joints may guide a decision to
proceed with arthroplasty or arthrodesis of the MCP
joint. If the CMC or IP joint pathology necessitates or
has resulted in fusion, it may be desirable to preserve
motion at the MCP joint with an implant arthroplasty,
avoiding arthrodesis at consecutive joints. Extensor
reconstruction, including EPL rerouting, can be per­
formed concomitantly with arthroplasty of the thumb
MCP joint to augment extensor and abductor forces.23
Following extensor reconstruction and MCP joint
arthroplasty, the thumb MCP joint is splinted in exten­
sion for 4 to 6 weeks, allowing for CMC and IP joint
exercises. If arthrodesis of the MCP joint is performed,
the joint is positioned for fusion in approximately 15°
of flexion and in slight pronation. Several surgical
options are available for arthrodesis of the thumb MCP
joint, including tension band wire fixation, crossing
Kirschner wires, headless compression screws, or plate
and screw fixation.24 Postoperatively, interphalangeal
joint motion is encouraged; however, the fusion site
is protected until radiographs demonstrate evidence
of union.
Severe type I deformities are characterized by fixed
deformities of both the MCP and interphalangeal joints.
Often, the condition of the interphalangeal joint will
warrant arthrodesis. Options for MCP joint arthrodesis
and arthroplasty are the same as for moderate type I
rheumatoid thumb deformities. The CMC joint should
be considered carefully in severe deformities as it is
affected commonly in this stage of the disease process.
Motion sparing procedures are preferred, and include
resection or hemiresection arthroplasty with ligament
reconstruction and soft tissue interposition arthro­
plasty.24 The use of a trapezial implant arthroplasty
should be considered carefully in the rheumatoid popu­
lation due to the higher risk for dislocation or implant
failure.24
Type III rheumatoid swan-neck deformities are
divided into subtypes: mild, moderate, and severe. Mild
type III deformities consist of isolated CMC joint patho­
logy. Therefore, surgical correction is directed toward
alleviation of CMC joint discomfort and dysfunction. If
conservative treatment fails to provide symptomatic
relief, CMC hemitrapeziectomy or trapeziectomy and
ligament reconstruction with soft tissue interposition
arthroplasty are considered. Moderate type III deformi­
ties are characterized by mild MCP joint involvement
PIP joint is near normal, DIP joint arthrodesis may be
considered. Postoperatively, the DIP joint is protected
in a mallet-finger splint and the PIP joint is left free to
promote mobilization. In patients with type I swan-
neck deformities who are unable to initiate PIP joint
flexion from a resting hyperextended position, soft
tissue reconstructive procedures that prevent PIP joint
hyperextension may be considered. These procedures
include volar skin dermodesis, oblique retinacular liga­
ment reconstruction, lateral band tenodesis,19 and PIP
joint flexor tenodesis.2,20
In type II swan-neck deformities, intrinsic tightness
results in the MCP joint being held in extension and PIP
joint motion is limited both actively and passively. As
with type I deformities, the presence of MCP joint
pathology and intrinsic tightness should be considered
prior to surgical intervention for correction of the swan-
neck deformity. If MCP joint arthroplasty or intrinsic
release is performed, flexor tenodesis of the PIP joint
may be necessary. Intrinsic release may be performed as
described by Nalebuff, in which the lateral band and
extensor hood are exposed via a dorsal approach, and a
1-cm segment of the lateral band and sagittal band
fibers is excised.7,14
Type III deformities generally require more extensive
soft tissue reconstructive procedures, as PIP joint con­
tractures and lateral band adhesions have developed.
There is decreased active and passive motion of the PIP
joint with fixed positioning of the lateral bands, dorsal
to the PIP joint axis of rotation. Reconstruction involves
lateral band release and volar translocation, dorsal
PIP joint capsulectomy, collateral ligament release, and
extensor tenolysis.16,19,21,22
Type IV deformities are characterized by degenerative
changes of the PIP joint articular surfaces in combina­
tion with a fixed hyperextension deformity of the PIP
joint. Typically, soft tissue procedures alone will not
achieve improved function or provide substantial pain
relief. Options for type IV deformities include implant
arthroplasty and arthrodesis.16,22
Rheumatoid Thumb Deformity
Nonoperative Management
The treatment includes pharmacologic therapy, includ­
ing intra-articular corticosteroid injections and oral
anti-inflammatory medications to decrease joint syno­
vitis. Functional splinting using custom-molded ther­
moplast materials or neoprene may be beneficial for
stabilization of the thumb for pinch and grasp activities
and for positioning of the thumb tip for dexterity-
related tasks.
Operative Management
Surgical correction of type I, or boutonnière, thumb
deformities is based on the subtype: mild, moderate, or
severe. In mild type I deformities, although soft tissue

389
Chapter 36:  Treatment of Boutonnière and Swan-Neck Deformities
Central Slip Reconstruction Using Local Tissue
Several methods have been described for central slip
reconstruction using local tissues when insufficient
central slip is available for direct repair and a flexible
deformity is present:
1. Snow’s Technique.28 The proximal stump of the
central slip is isolated from surrounding tissues. A
distally based flap of extensor tendon, sufficient to
span the central slip deficit is elevated sharply and
is turned back on itself. The flap is repaired to the
lateral bands using nonabsorbable suture. Follow­
ing repair, passive PIP joint flexion of greater than
60° without excessive tension on the repair site
should be possible.
2. Aiche’s Technique.29 The radial and lateral bands/
conjoined lateral bands are isolated by longitudi­
nal division from the trifurcation of the extrinsic
extensor tendon to the triangular ligament. The
dorsal half of each lateral band / conjoined lateral
band are mobilized dorsally and are sutured
together using nonabsorbable suture. Relocation
of the remaining lateral band is recommended if
it remains volar to the PIP joint axis of rotation.
3. Littler and Eaton’s Technique.30 The radial and ulnar
lateral bands/conjoined lateral bands are isolated
and incised over the middle phalanx. The ORL
must be preserved; otherwise, DIP joint extension
will be compromised. The divided lateral bands
are mobilized dorsally and sutured into the inser­
tion of the central slip. This method is not indi­
cated in the presence of severe attenuation of the
central slip.
4. Matev’s Technique.31 The lateral bands/conjoined
lateral bands are isolated longitudinally. The ulnar
lateral band is divided at the level of the DIP joint
and the radial lateral band is incised at the mid-
point of the middle phalanx. The proximal stump
of the ulnar lateral band and the distal stump of
the radial lateral band are sutured over the dorsal
digit, thereby lengthening the lateral band. The
proximal stump of the radial lateral band is sutured
into the remaining central slip and the base of the
middle phalanx in order to assist with PIP joint
extension. The PIP joint is held in extension for
6 weeks postoperatively, with or without transar­
ticular Kirschner wire fixation (Figure 36-5).
Central Slip Reconstruction Using Tendon Graft
A dorsal exposure of the digit is used, elevating full-
thickness skin flaps and preserving the extensor
paratenon. The central slip remnant is isolated and ele­
vated from the surrounding tissues. An autologous
tendon graft, often the ipsilateral palmaris longus (PL)
tendon, is harvested and is passed through an osseous
tunnel created through the dorsal base of the middle
(flexible deformity) in addition to CMC joint patho­
logy. Progressive MCP joint hyperextension should be
addressed concurrently with surgical intervention for
CMC joint involvement. MCP joint volar plate capsu­
lodesis, sesamoidesis, or volar tenodesis can be coupled
with CMC hemitrapeziectomy or trapeziectomy and
ligament reconstruction with soft tissue interposition
arthroplasty. Transarticular pin stabilization of the MCP
joint in 20° to 30° of flexion for 3 to 4 weeks postop­
eratively protects the joint while permitting early motion
of the interphalangeal joint.
The severe type III deformity consists of CMC joint
dislocation, adduction contracture of the metacarpal,
and fixed MCP joint hyperextension. Each of these com­
ponents should be addressed in the surgical interven­
tion performed for severe type III rheumatoid thumb
deformities. Generally, arthrodesis of the MCP joint is
warranted to correct the rigid MCP joint hyperexten­
sion. Treatments options for the CMC joint are the same
as those for mild and moderate deformities, including
CMC resection arthroplasty and ligament reconstruc­
tion or tendon interposition arthroplasty. Often, adduc­
tion contracture of the thumb metacarpal is corrected
with metacarpal base resection during resection arthro­
plasty of the CMC joint; however, if the correction is
inadequate, fasciotomy of the first dorsal interosseous
and adductor muscles and, rarely, first web to space
reconstruction with z-plasties, may be performed.24
SURGICAL PROCEDURES
Boutonnière Reconstruction
Primary Central Slip Repair
A dorsal approach is used, preserving the extensor
paratenon. The central slip is isolated and redundant
fibrous tissue is excised after assessing central slip length
with the PIP joint held in extension. In certain cases, a
V-Y advancement is required if tissue is insufficient. If
an avulsion fracture is present, the fragment is carefully
elevated, preserving the central slip attachment. If the
fragment is smaller and fixation with Kirschner wires
or screws is inappropriate, then the fragment is excised
and the central slip is repaired directly to the dorsal
base of the middle phalanx using a suture anchor or
pullout suture method. If a larger fragment, amenable
to fixation, is present, then it is anatomically reduced
and stabilized with two small Kirschner wires or screws.
The lateral bands are restored to their anatomical
location. Incising the transverse retinacular ligaments
may be required to permit mobilization dorsal to the
axis of rotation of the PIP joint. The lateral bands are
stabilized in their anatomical position using non­
absorbable suture. Following repair, the PIP joint is
maintained in full extension for 6 weeks. Often, a trans­
articular Kirschner wire is used to temporarily stabilize
the joint.15,17,18,21,25-27

390
Section 4:  Extensor Tendon Repair and Reconstruction
of the flexor sheath, from the distal margin of the A2
pulley to the proximal margin of the A4 pulley is ele­
vated, exposing the flexor tendons. One slip of the FDS
tendon is identified and divided at the level of the
decussation, preserving its insertion into the middle
phalanx. The free, proximal end of the divided FDS
slip is passed from deep to superficial (dorsal to volar)
through a transverse incision in the A2 pulley, approxi­
mately 3mm from the distal margin of the pulley. The
tendon slip is sutured back onto itself using nonabsorb­
able suture, tensioning the repair to hold the PIP joint
in approximately 20° of flexion (Figure 36-7).
After surgery, the repair is protected in a dorsal block­
ing splint with the PIP joint at approximately 30°
flexion for 6 to 8 weeks; protected flexion exercises are
started at 2 to 3 weeks.
Lateral Band Tenodesis
The extensor apparatus is exposed via a dorsal curvilin­
ear incision. Cleland’s ligaments are divided and the
flexor sheath is accessed. The dorsally subluxated lateral
band is dissected free from the central slip and the tri­
angular ligament; the lateral band is left intact proxi­
mally and distally. The lateral band is translocated volar
to the PIP joint axis of rotation and is stabilized by a
phalanx. The two limbs of the tendon graft are weaved
into the lateral bands with the digit held in neutral. A
temporary, transarticular Kirschner wire may be placed
to stabilize the joint and to protect the reconstruction.
Extensor Tenotomy
It is indicated in the presence of a flexible deformity and
in patients for whom prior PIP joint surgery has been
unsuccessful.26 This procedure can be considered in con­
junction with PIP joint arthrodesis. The terminal tendon
is isolated and is elevated proximally over a distance of
1.5 cm from the underlying DIP joint and middle
phalanx.
The terminal tendon is incised transversely, distal to
the triangular ligament. The ORL must be preserved so
as to not compromise DIP joint extension. The incised
tendon ends are separated by passively extending the PIP
joint and passively flexing the DIP joint (Figure 36-6).
Swan-Neck Reconstruction
PIP Joint Flexor Tenodesis
A check-rein to PIP hyperextension can be created using
a slip of the FDS tendon.17,18,21,32-34 The flexor sheath is
exposed from the base of the digit to the A4 pulley via
a Bruner or mid-axial incision. The membranous portion
Figure 36-6  Location of extensor tenotomy (solid vertical
line “I”) distal to the triangular ligament and proximal to the
ORL insertion at the terminal tendon. Lateral view of the
digit demonstrating the coalescing fibers of the lateral slip
(LS) and oblique fibers of the extensor apparatus (O) which
combine to form the conjoined lateral band (clb). The two
conjoined lateral bands combine to form the terminal
tendon (TT), which inserts into the dorsal base of the distal
phalanx. The transverse retinacular ligament (TRL) prevents
dorsal subluxation of the lateral bands. The oblique
retinacular ligament (ORL) passively links the PIP and DIP
joints as it travels from volar to dorsal from the fibro-
osseous gutter (middle third of the proximal phalanx and A2
pulley) to the proximal aspect of the distal phalanx through
the extensor tendon. (Reprinted with permission: © 2004
Leversedge FJ, Goldfarb CA, Boyer MI.)
LS
O
TRL
ORL
TT
clb
Figure 36-5  Illustration of Matev’s technique for extensor
reconstruction.
Matev technique

391
Chapter 36:  Treatment of Boutonnière and Swan-Neck Deformities
graft is passed volarly, deep to the radial neurovascular
bundle, and is sutured to the distal margin of the A2
pulley after appropriate tensioning. The A2 pulley may
be stabilized by passing the tendon graft from deep to
superficial through a small incision 2 to 3 mm proximal
to the distal and lateral margin of the pulley, before
folding it back upon itself and securing it with nonab­
sorbable suture. Prior to suture repair, the DIP joint is
held in full extension and the PIP joint in 25° of flexion
using temporary, transarticular Kirschner wires that are
left in place for approximately 6 weeks.
Type III Swan-Neck Reconstruction
Via a dorsal curvilinear incision, the extensor apparatus
is exposed. The following procedures are considered, as
indicated: (1) lateral band release from the central
tendon and from the triangular ligament; (2) dorsal PIP
joint capsulectomy; (3) release of the radial and ulnar
collateral ligaments symmetrically from dorsal to volar
as needed to permit passive flexion of the PIP joint to
90°; (4) lateral band stabilization volar to the PIP joint
axis of rotation is not required, typically, as they trans­
late passively with PIP joint flexion; and (5) the PIP
joint in 20° flexion is stabilized using a transarticular
Kirschner wire that is removed 2 to 3 weeks postopera­
tively. The reconstruction is protected in a forearm-
based splint, which may be removed to permit MCP and
DIP joint motion.
SUMMARY
The treatment of boutonnière and swan-neck deformi­
ties of the digits is guided by the etiology and staging
of the condition, primarily differentiated as being the
sequelae of trauma or as a part of the spectrum of rheu­
matoid disease. Treatment may range from splinting
dorsally based flap of the flexor sheath at the level of
the PIP joint.
Alternatively, the lateral band may be incised at its
distal insertion into the terminal tendon and rerouted
from proximal to distal through a 0.5- to 1.0-m segment
of the flexor sheath. The tendon slip is repaired subse­
quently to its insertion at the terminal extensor tendon
(Figure 36-8).
The repair is protected postoperatively with a dorsal
blocking splint, which maintains the PIP joint in greater
than 30° flexion for 6 to 8 weeks. Early protected digital
flexion exercises are encouraged.
Oblique Retinacular Ligament Reconstruction
ORL reconstruction may be indicated in the presence of
a well-preserved DIP joint when a flexible swan-neck
deformity develops as the result of a chronic mallet
injury. The digit is exposed via an incision from the
MCP joint flexion crease along the radial midaxial line,
curving dorsally to end over the DIP joint. The radial
neurovascular bundle is protected and the A2 pulley
and the terminal tendon slip are identified. A suitable
graft (ideally the ipsilateral PL tendon, when present) is
harvested. The graft is sutured to the terminal tendon
slip using nonabsorbable suture and the free end of the
Figure 36-7  Illustration of the volar digit demonstrating
(A) the incised FDS slip proximally at the FDS decussation in
relation to the A2 and A4 pulleys and (B) the passage of the
FDS slip through a transverse incision in the A2 pulley and
repair back onto itself, holding the PIP joint in 20° to 30°
flexion.
A
B
A4
A2
Flexor
digitorum
superficialis
(FDS)
tendon
Figure 36-8  Illustration of the lateral digit demonstrating:
A, Incision of the lateral band at its insertion into the
terminal tendon; B, the lateral band is elevated, brought
proximally to be re-routed through a 1-cm portion of the A2
pulley before being repaired back to the terminal tendon
insertion over the dorsal DIP joint.
A4
Lateral
band
A
B

392
Section 4:  Extensor Tendon Repair and Reconstruction
with rheumatoid disease, the swan-neck deformity can
be secondary to pathology at the MCP, PIP, or DIP
joints. Determination as to the specific type of defor­
mity is critical to successful treatment in this setting.
Nonoperative treatment rarely leads to a favorable
outcome in patients with an advanced swan-neck defor­
mity. However, the flexible swan neck deformity may be
treated successfully in a figure-of-eight splint. Surgical
options for rheumatoid and post-traumatic swan neck
deformities include volar positioning of the lateral band
with tenodesis, oblique retinacular ligament reconstruc­
tion, DIP arthrodesis, and PIP joint arthroplasty or
arthrodesis.
Boutonnière or swan-neck deformity of the thumb
can develop in the presence of rheumatoid disease. It is
imperative to assess both individually and collectively
the status of the CMC, MCP, and IP joints in formulat­
ing a treatment plan. Operative intervention varies,
with options ranging from soft tissue reconstruction
procedures to arthroplasty and arthrodesis. In the CMC
joint, implant interposition arthroplasty may have an
increased failure rate due to poor soft tissue restraints
and an increased risk of dislocation.
Ultimately, the primary goals of treatment of swan
neck and boutonnière deformities are to alleviate pain
and to improve function. Preoperative patient educa­
tion will help to avoid unrealistic expectations and or
the consequences of unanticipated outcomes. Patient
selection and thorough preoperative evaluation are criti­
cal to guiding appropriate intervention and to achieving
successful outcomes.
and therapy to operative intervention and, therefore, the
careful assessment of the patient and the deformity is
critical to the formulation of an appropriate and indi­
vidualized treatment strategy. A thorough and global
evaluation of the extremity should include consider­
ation for the current or potential pathology of adjacent
joints, the active and passive motion of involved joints,
subjective concerns for pain or instability, and pertinent
radiographic assessment. In patients with rheumatoid
disease, reconstruction for wrist and/or MCP joint
pathology may influence decision-making for PIP and
DIP joints, including surgical techniques and the
sequence of reconstruction.
A boutonnière deformity, post-traumatic or rheuma­
toid, develops from an injury to the central slip at the PIP
joint. The deformity may evolve in a progressive fashion
secondary to trauma and, therefore, may not be imme­
diately appreciated clinically following injury. Outcomes
are improved with early diagnosis, particularly utilizing
the Elson test. However, if the diagnosis is within 2 to 3
months of the time of injury, extension splinting (with
or without use of a temporary transarticular Kirschner
wire holding the PIP joint in extension) is an effective
treatment. Surgical intervention for boutonnière defor­
mities that have failed conservative treatment range from
central slip reconstruction with lateral band reposition­
ing to PIP joint arthroplasty and arthrodesis.
Swan-neck deformities are characterized by laxity of
the PIP volar plate and an inability of the terminal slip
to extend the DIP joint, leading to hyperextension of
the PIP joint and flexion of the DIP joint. In patients
References
1. Carducci T: Potential boutonniere deformity: its recognition
and treatment, Orthop Rev 10:121–123, 1981.
2. Lovett WL, McCalla MA: Management and rehabilitation of
extensor tendon injuries, Orthop Clin North Am 14:811–826,
1983.
3. Lattanza L, Hattwick A: Extensor tendon repair and recon­
struction. In Hand, Elbow Shoulder: Core Knowledge in Ortho­
paedics, Philadelphia, 2006, Mosby, pp 201–211.
4. Elson RA: Rupture of the central slip of the extensor hood of
the finger. A test for early diagnosis, J Bone Joint Surg (Br)
68:229–231, 1986.
5. Rubin J, Bozentka DJ, Bora FW: Diagnosis of closed central
slip injuries. A cadaveric analysis of non-invasive tests, J Hand
Surg (Br) 21:614–616, 1996.
6. Coons MS, Green SM: Boutonniere deformity, Hand Clin
11:387–402, 1995.
7. Nalebuff EA: The rheumatoid swan-neck deformity, Hand
Clin 5:203–214, 1989.
8. Nalebuff EA, Millender LH: Surgical treatment of the bouton­
niere deformity in rheumatoid arthritis, Orthop Clin North Am
6:753–763, 1975.
9. Nalebuff EA: Diagnosis, classification and management of
rheumatoid thumb deformities, Bull Hosp Joint Dis 29:119–
137, 1968.
10. Terrono A, Millender L, Nalebuff E: Boutonniere rheumatoid
thumb deformity, J Hand Surg (Am) 15:999–1003, 1990.
11. Ratliff AH: Deformities of the thumb in rheumatoid arthritis,
Hand 3:138–143, 1971.
12. Flatt AE: The Care of the Arthritis Hand, St Louis, 1995, Quality
Medical Publishing, pp 263-264.
13. Burton RI, Melchoir JA: Extensor tendons: late reconstruction.
In Green DP, Hotchkiss RN, Pederson WC, editors: Green’s
Operative Hand Surgery, ed 4, New York, 1999, Churchill
Livingstone, pp 215–221.
14. O’Brien ET: Surgical Principles and planning for the rheuma­
toid hand and wrist, Clin Plast Surg 23:407–420, 1996.
15. Urbaniak JR, Hayes MG: Chronic boutonniere deformity: an
anatomical reconstruction, J Hand Surg (Am) 6:379–383,
1981.
16. Boyer MI, Gelberman RH: Operative correction of swan-neck
and boutonniere deformities in the rheumatoid hand, J Am
Acad Orthop Surg 7:92–100, 1999.
17. Swanson AB, Maupin BK, Gajjar NV, et al: Flexible implant
arthroplasty in the proximal interphalangeal joint of the
hand, J Hand Surg (Am) 10:796–805, 1985.
18. Takigawa S, Meletiou S, Sauerbier M, et al: Long-term assess­
ment of Swanson implant arthroplasty in the proximal inter­
phalangeal joint of the hand, J Hand Surg (Am) 29:785–795,
2004.
19. Gainor BJ, Hummel GL: Correction of rheumatoid swan-neck
deformity by lateral band mobilization, J Hand Surg (Am)
10:370–376, 1985.

393
Chapter 36:  Treatment of Boutonnière and Swan-Neck Deformities
27. Towfigh H, Gruber P: Surgical treatment of the boutonniere
deformity, Oper Orthop Traumatol 17:66–78, 2005.
28. Snow JW: Use of a retrograde tendon flap in repairing a
severed extensor in the PIP joint area, Plast Reconstr Surg
51:555–558, 1973.
29. Aiche A, Barsky AJ, Weiner DL: Prevention of boutonniere
deformity, Plast Reconstr Surg 46:164–167, 1970.
30. Littler JW, Eaton RG: Redistribution of forces in the correction
of boutonniere deformity, J Bone Joint Surg (Am) 49:1267–
1274, 1967.
31. Matev I: Transposition of the lateral slips of the aponeurosis
in treatment of long-standing “boutonniere deformity” of the
fingers, Br J Plast Surg 17:281–286, 1964.
32. Thompson JS, Littler JW, Upton J: The spiral oblique retinacu­
lar ligament (SORL), J Hand Surg (Am) 3:482–487, 1978.
33. Kleinman WB, Petersen DP: Oblique retinacular ligament
reconstruction for chronic mallet finger deformity, J Hand
Surg (Am) 9:399–404, 1984.
34. Tonkin MA, Hughes J, Smith KL: Lateral band translocation
for swan-neck deformity, J Hand Surg (Am) 17:260–267,
1992.
20. Curtis R: Sublimis tenodesis. In Edmonson AS, Crenshaw AH,
editors: Campbells’s Operative Orthopaedics, ed 6, St Loius,
1980, CV Mosby, p 319.
21. Kiefhaber TR, Strickland JW: Soft tissue reconstruction for
rheumatoid swan-neck and boutonniere deformities: long
term results, J Hand Surg (Am) 18:984–989, 1993.
22. Strickland JW, Boyer M: Swan neck deformity. In Strickland
JW, editor: The Hand. Master Techniques in Orthoapedic Surgery
Series, Philadelphia, 1998, Lippincott-Raven, pp 459–470.
23. Figgie MP, Inglis AE, Sobel M, et al: Metacarpal-phalangeal
joint arthroplasty of the rheumatoid thumb, J Hand Surg
(Am) 15:210–216, 1990.
24. Stein AB, Terrono AL: The rheumatoid thumb, Hand Clin
12:541–550, 1996.
25. Wilczynski M, Boyer MI, Leversedge FJ: Operative reconstruc­
tion of boutonnière and swan-neck deformities. In Wiesel S,
editor: Operative Techniques In Orthopaedic Surgery, Philadel­
phia, 2011, Wolters Kluwer/Lippincott Williams & Wilkins,
pp 2619–2633.
26. Meadows SE, Schneider LH, Sherwyn JH: Treatment of the
chronic boutonniere deformity by extensor tenotomy, Hand
Clin 11:441–447, 1995.

394
CHAPTER
37 
VASCULARIZED TENDON
GRAFT FOR EXTENSOR
TENDON RECONSTRUCTION
Roberto Adani, MD, Luigi Tarallo, MD,
Massimo Corain, MD, and Jean Claude Guimberteau, MD
can be included are the palmaris longus (PL) and a strip
of the brachioradialis tendons along with fascia and
skin, and a slip of flexor carpi radialis (FCR) tendon.21,22
The ulnar island flap of the forearm allows for the inclu­
sion of the PL and a strip of the flexor carpi ulnaris
(FCU) tendons.23,24
OPERATIVE TECHNIQUES
Dorsalis Pedis Cutaneotendinous Free Flap
The surgical procedure uses two operative teams: one to
harvest the cutaneotendinous free flap and the other to
prepare the recipient site. A pattern outlining the soft
tissue defect on the dorsum of the hand is transferred
to the dorsum of the foot centered on the second meta­
tarsal bone using Doppler examination. It is helpful to
also outline the venous drainage. Flap elevation begins
distally between the first and the second toes, identify­
ing the first dorsal metatarsal artery that is tied and
sutured to the distal skin margin. The flap is elevated
from distal to proximal and laterally to medially below
the level of the first dorsal metatarsal artery. The flap is
dissected to include in continuity, the long toe extensors
to the second through fifth toes tendons, keeping the
areolar tissue around the tendons, vessels, and nerves
with the flap. The extensor digitorum tendons are fol­
lowed proximally and are divided, according to the
number of destroyed tendons in the hand. Care should
be taken to preserve the paratenon on the extensor hal­
lucis longus tendon, which is left on the foot. During
elevation, the long saphenous vein or median dorsal
vein is preserved, In addition, the superficial peroneal
nerve must be included in the flap. Finally, the extensor
retinaculum is divided and the dorsalis pedis artery can
be traced until a sufficient pedicle length is obtained.
Radial Artery Forearm Island Flap With
Vascularized Tendons
This is a modification of the radial artery forearm flap.
Preoperatively, an Allen test must be done to ensure the
OUTLINE
This chapter describes different vascularized compo­
site tissue graft procedures incorporating tendons to
reconstruct the tendons and overlying soft tissue
defects in the hand. The options are the dorsalis pedis
flap, the radial forearm flap, and the ulnar artery–
based island flap. The radial island fasciotendinous
flap is an excellent evolution of the radial forearm flap
for coverage of soft tissue defects of the hand; we rec­
ommend this composite fascial flap to reconstruct the
dorsum of the hand that requires simultaneous grafts
of two or three tendons.
Composite tissue loss of the hand involving tendon
defects represents a great clinical challenge. These injuries
require restoration of both skin coverage and tendon
function. These injuries are approached by different ways:
multistage reconstructions of soft tissue then tendon, or
single-stage vascularized composite tissue grafting.1
Multistage reconstructions include skin coverage
with distant flaps and tendon grafting in a later stage.2
Nonvascularized tendon grafts in conjunction with ped­
icled flaps or free tissue transfer are termed partially
vascularized tissue transfer.1 Skin,3-8 fascial,9-11 and muscle
flaps12,13 can be used for this purpose. A completely
vascularized single-stage reconstruction uses a compos­
ite flap in which different tissue components (skin,
tendon, and nerve) are included.
In 1979, Taylor and Towsend14 described for the first
time a composite free flap, with attached vascularized
segments of the extensor hallucis brevis to the great toe
and the extensor digitorum longus to the second toe.
Subsequently, several authors15-19 have described good
results for the treatment of compound injuries with the
dorsalis pedis composite flap; this flap provides four
vascularized tendons (extensor digitorum communis
[EDC]) of adequate length.
Reid and Moss20 modified the radial artery forearm
flap to include flexor tendons from forearm. The tendons

395
Chapter 37:  Vascularized Tendon Graft for Extensor Tendon Reconstruction
Then, the ulnar pedicle is separated from the ulnar nerve
on its dorsal aspect along its whole length from the
lower third of the forearm to Guyon’s canal. This ulnar
skin flap is raised together with the FCU tendon.
The FCU tendon is longitudinally split into two parts.
One is maintained in place with the muscle and the
other one (i.e., the medial part) is transected at its distal
insertion on the triquetrum. The proximal portion is
transected at the musculotendinous junction. All the
other branches of the ulnar artery are ligated, except
branches to the skin flap on the anterior side and to the
periosteum in the event of a bone transfer. The ulnar
pedicle is dissected distally and the combined tendon–
skin transfer is performed (Figure 37-2).
The FCU and ulnar skin transfer is generally used as
a retrograde flap by preserving the inflow from the distal
portion of the ulnar artery. One method is by transpos­
ing to the anterior and radial side for thumb and index
reconstruction. The other method is transposing to the
ulnar and posterior direction. Because the transfer is
oriented on the lateral side of the hand, caution has to
be taken concerning the sensory branches of the radial
nerve. Sutures for extensor tendon reconstruction are
various, from simple sutures to Pulvertaft weaves.
POSTOPERATIVE CARE
We perform split-thickness skin grafting on the defect
immediately following flap harvesting. When the dor­
salis pedis flap is used, the ankle, foot, and toes are
splinted to prevent movement under the graft. When the
radial forearm flap is used, the wrist and the fingers are
immobilized for 10 days with the wrist in 20° to 30° of
extension with the metacarpophalangeal (MCP) joints
in 50° of flexion and the interphalangeal (IP) joints in
extension.4
Initially, the hand is splinted with the wrist at 30° of
extension and MCP and IP joints in 0°. In our case
series, these injuries have been managed with variable
periods of immobilization after repair; nevertheless,
presence of a complete vascular palmar arch. The pres­
ence of a PL tendon is also confirmed. The pattern of
the dorsal defect is outlined around the radial vessels,
and the flap is raised including skin and fascia. Two 7- to
8-cm-long strips of FCR tendon and brachioradialis
tendons might be included in the flap with the PL
tendon, if it is present. Care is taken to preserve the
areolar tissue around the tendons, which receive their
blood supply from the deep fascia. The radial artery and
its venae comitantes are ligated proximally, and the flap
is mobilized distally. Due to cosmetic reasons, this flap
is less favored by many surgeons in recent years.
Ulnar Artery–Based Island Flap With
Vascularized Tendons (Used by J.C.G.)
At the distal third of the wrist, just proximal to the flexor
retinaculum, the ulnar artery on its lateral side gives off
one or two branches of around 1 mm in diameter to the
distal part of the FCU tendon. Two types of arterial
branching can be observed, either directly from the
ulnar artery or from the dorsal branch of the ulnar
artery. These vascular branches are constant and easily
identified on the lateral side of the pedicle. It is possible
to carry out not only tendinous vascular transfers but
also a cutaneotendinous transfer, and even the triple
transfer of skin, tendon, and bone.
Preoperative evaluation includes the Allen test and
Doppler testing to ascertain that the radial artery pro­
vides adequate blood supply to the hand. Angiography
of the arm is also advisable. A bayonet-shaped incision
is first traced and then made on the medial side of the
forearm, the axis of the incision overlying the lateral
border of the flexor carpi ulnaris (Figure 37-1). The
ulnar pedicle is dissected and all its branches are care­
fully separated and divided.
First, the cutaneous branches between the ulnar
artery and the skin, emerging from the volar aspect of
the pedicle and which are the principal components
of the ulnar forearm flap, are then carefully isolated.
Figure 37-1  Skin incision for harvesting the ulnar artery-
based skin flap incorporating half of the FCU tendon.
Figure 37-2  Ulnar artery island skin flap with the FCU
tendon connecting to the flap by fascia.

396
Section 4:  Extensor Tendon Repair and Reconstruction
rehabilitation of the extensor tendon has recently been
settled in favor of early mobilization.25
Mobilization is started after 1 week, allowing 0° to
30° movements at MCP joints with IP joints free to
move during exercise. The flexion movement of MP
joints is gradually increased, and after 5 weeks MCP and
IP joints are left free to move during the day with night
splinting. The splint is continued for 6 weeks. Patient
motivation is very important: without the patient’s
active participation, functional recovery could be very
limited.
OUTCOMES
Between 1988 and 2008, the lead author (R.A.) and his
colleagues treated 21 patients with composite loss of
tendon and skin on the dorsum of the hand. The skin
defects ranged from 8 × 5 cm to 11 × 13 cm. The interval
from injury to flap transfer ranged from 2 to 40 days,
with an average of 14 days. The time between injury and
reconstruction was long because initial trauma in most
cases were treated elsewhere and referred to us later. The
number of transferred tendons in each patient was three
on average. We used cutaneotendinous dorsalis pedis
flap in 8 cases, and the radial tendinous island flap in
13 cases (a cutaneous flap in 10 cases and a fascioten­
dinous flap in 3 cases).
The aim of extensor tendon reconstruction is to
achieve free gliding of the tendons and normal finger
movement. The transferred tendons function well in
our patients. In one case the flexion of the fingers
and the extension of MCP joints were full, but there were
IP joint extension deficits of the middle, ring, and little
finger due to the original interosseous muscle avulsion.
Tendon adhesions markedly limiting hand motion
were seen only in one case; tenolysis was necessary. In
the donor site, toe extension was weak. However, toe
motion was not impaired because the short extensors of
the toes were intact. Partial skin graft loss at the donor
site occurred in the majority of the cases treated with
dorsalis pedis flap but did not require additional surgery
except that skin grafting was redone in one patient with
radial artery flap. Hypertrophic scarring occurred in all
patients at the donor site; however, this improved with
time, particularly in the foot.
Ulnar artery–based composite flap tendon transfer
was performed in 14 patients by J.C.G. and his col­
leagues. There were no cases of complete necrosis, but
there was one case of partial skin loss because the flap
was passed in a subcutaneous tunnel. The sacrifice of the
ulnar artery is the main disadvantage. In the long term,
the flap may remain somewhat swollen for 3 or 4 months.
It may be necessary to reshape the skin flap in the months
that follow. The tissue quality is excellent, being fine, fat
free, and virtually hairless (Figure 37-3). There are
minimal concerns at the donor site. The scar on the
forearm is usually very limited and not hypertrophic.
Figure 37-3  A case of skin defect and EPL tendon defect
over the proximal part of the thumb. A and B, An island
ulnar artery flap with the FCU tendon was transferred to
cover skin defect and reconstruct the EPL tendon. C and
D, Postoperatively, the cosmetic appearance is excellent
and thumb extension was restored.
A
B
C
D

397
Chapter 37:  Vascularized Tendon Graft for Extensor Tendon Reconstruction
DISCUSSION
The dorsum of the hand is covered by a thin skin with
little subcutaneous tissue, underneath which lie the
extensor tendons; traumatic loss of skin is often associ­
ated with extensor tendon defects. These injuries are
conventionally treated with primary distant flap transfer
(axial and random pattern flaps, groin flap, etc.) and
secondary tendon grafts or transfers.2 The treatment
does not yield satisfactory results either aesthetically or
functionally because of the need for immobilization of
the hand in a nonfunctional position and bulky flap
coverage.14,15 Moreover, the risk of tendon adhesions
may require the staged tendon reconstruction preceded
by silicone rubber tendon implants in the first stage.26
Multiple operations also require a prolonged period of
physiotherapy and functional recovery.
The partially vascularized tissue transfer allows the
hand to be placed in a functional position and ensure
early mobilization. Flaps of various forms can be
used: island or free skin flaps, island or free fascial
flaps, island or free muscle flaps. Fascial and muscle
flaps do not always allow immediate tendon repair.9-12
Tendon reconstruction is more difficult under muscle
or fascial than with cutaneous flaps.27 Muscle flaps are
too bulky, requiring subsequent thinning, and they
adhere to tendon grafts extensively.6,7 This have been
partially solved by using partial superior latissimus or
partial medial rectus flaps; in this way, the debulking
rate has decreased, which may allow one-stage tendon
reconstruction.13
Distally based radial3,4 and ulnar5 flaps sacrifice a
major vascular supply to a hand, and this may jeopar­
dize blood flow to an already compromised area.6
Scheker and colleagues6 used primary tendon grafts
covered by a free flap from either the lateral arm or groin.
Free tendon grafts were placed through a subcutaneous
tunnel under the flap. This technique may damage the
flap vessels, which has resulted in hematomas, causing
tendon adhesion.8 To minimize this problem, the groin
flap can be harvested along with a sheet of external
oblique aponeurosis preserving the vascular connec­
tions between it and the under surface of the groin flap.
This aponeurosis is used to create a gliding surface for
tendon grafts, reducing tendon adhesion.8
The vascularized flap incorporating extensor tendons
allows one-stage reconstruction of all lost structures.
The vascularized tendon grafts may heal faster with less
adhesion.14 Hand motion can be initiated earlier because
union of the tendon junction is achieved sooner than
with conventional tendon grafts.21 The dorsalis pedis
cutaneotendinous flap has been used most commonly.
It can provide four vascularized tendons of adequate
length completely surrounded by paratenon and loosely
attached to skin as thin as that on the dorsum of the
hand (Figure 37-4).
Figure 37-4  A case of severe friction injury to the dorsum
of the hand caused by road accident. A, Loss of skin and
EDC tendons of index, middle and ring fingers and extensor
indicis proprius tendon. B, Transfer of dorsalis pedis artery
flap with extensor tendons. C and D, Aesthetic and
functional results.
A
B
C
D

398
Section 4:  Extensor Tendon Repair and Reconstruction
In our cases, tenolysis was not necessary and a defat­
ting procedure was never used with the dorsalis pedis
flap. Recently, the dorsalis pedis flap has been less
used as a free skin flap transfer because of variations
in donor site anatomy, delayed wound healing, tedious
dissection, and scar formation on the dorsum of the
foot.21,27-29 However, the cosmetic appearance of the
donor site of the foot is rather acceptable in our patients
after dorsalis pedis flap transfer (Figure 37-5). These
patients believe the overall benefits of the procedure
outweighed the donor-site morbidity. There was only
minimal functional deficits. Delayed donor site healing
was common in this series of patients and late wound
breakdown was also common. However, this rarely
produced a significant long-term functional deficit.27
None of our patients had ulcerations, and all were able
to walk normally. Only two patients had contractures
of the skin at the grafted area overlying the extensor
hallucis longus tendon and the extensor digiti minimi
tendon with minimal deficits of toe plantar flexion
(Figure 37-6).
We believe that some technical pearls can help reduce
donor site problems associated with dorsalis pedis flap.
Figure 37-5  A patient with severe friction injury on the dorsum of the hand. A, Loss of skin and extensor tendons to the
middle and ring fingers. B, Dorsalis pedis artery flap of 7 × 11 cm with two extensor digitorum longus tendons of the toe was
harvested and transferred to the hand. The extensor digitorum longus tendons were sutured to the extensor communis
tendons in the dorsum of the hand. C, Immediate postoperative view. D–F, Aesthetic and functional results.
A
B
C
D
E
F
The size of the flap should be limited and surgeons
should avoid raising the flap too distally: the morbidity
of the donor site is related to both the site and the loca­
tion of the flap harvested. In the literature, increased
morbidity seems to accompany harvested flaps that
extend to within 2 cm of the metatarsophalangeal
(MTP) joint crease.28 In contrast, when the flap was
raised 2 cm or more proximal to the MTP joint crease,
there were no problems with the distal portion of the
donor site.
Some authors20, 21 preferred a radial forearm cutaneo­
tendinous flap. The flap transfer is quick and simple, but
only a limited number of tendons are available. The PL
tendon can be taken completely with strips of the FCR
tendon and the brachioradialis tendon as vascularized
grafts (Figure 37-7). The sacrifice of the radial artery
does not generally cause significant problems. In cases
of severe hand trauma, use of the reverse radial forearm
flap can be contraindicated.6 In our cases, there were few
donor site complications. To minimize donor site mor­
bidity, the radial border of the flap should not be
extended beyond the radial border of the forearm and
the FDS muscle must be imbricates over the FCR tendon

399
Chapter 37:  Vascularized Tendon Graft for Extensor Tendon Reconstruction
A
B
C
D
E
F
G
Figure 37-6  A patient with hot press injury. A, Loss of
dorsal skin and all extensor tendons to the four fingers.
B, Defects of all extensor tendons in the dorsum of the
hand. C, Harvest of a free cutaneous tendinous dorsalis
pedis flap including four extensor digitorum longus
tendons: intraoperative view showing tendon
reconstruction. D, Immediate postoperative result.
E–G, Aesthetic and functional results of the hand.
to prevent exposure of this tendon.4 The cosmetic
appearance of the flap donor site could be improved
covering it with artificial dermis (Integra, Integra Life­
sciences, Plainsboro, NJ, USA).30 However, in recent
years, more surgeons prefer not to use this site as a
donor because of obvious cosmetic reasons in the
forearm.
An evolution of the radial forearm cutaneotendinous
flap is the radial island fasciotendinous flap.31 This radial
fascial flap in combination with vascularized extensor
tendons has not been popular. Reid and Moss20 used a
radial artery–based flap of fascia and tendon alone,
leaving the forearm skin unattached. The fascia flap
includes nerve and tendon units to facilitate true com­
posite tissue reconstructive efforts,10 which provide thin,
pliable tissue with no or minimal donor site morbidity.
We have used radial forearm cutaneotendinous flap in
three cases thus far (Figure 37-8). We obtained excellent
functional and aesthetic results in the hand with
minimal donor site morbidity. None of the patients had

400
Section 4:  Extensor Tendon Repair and Reconstruction
A
B
C
D
E
G
F
Figure 37-7  A, A case of hot press injury on the
dorsum of the hand with section of EDC tendon of
the middle finger and loss of EDC tendons of the
ring and little fingers and extensor digit minimi
proprius. B, Intraoperative débridement of the
dorsal wound. C, Reconstruction with radial
artery cutaneotendinous island flap; detail of
reconstruction of the extensor apparatus of the
little finger with PL tendon. D–F, Aesthetic and
functional results after surgery. G, Donor site
result in the forearm.
symptoms of cold intolerance due to the sacrifice of the
radial artery, or functional deficits due to the harvesting
of the tendons. The outcomes of the use of this flap are
encouraging.
SUMMARY
The dorsalis pedis flap, radial forearm cutaneotendi­
nous flap, and island ulnar artery flap may provide good
options for one-stage simultaneous reconstruction of
skin and tendon defects on the dorsum of the hand. In
our patients, we achieved good cosmetics and function
after these composite flap procedures.
When grafting of three or four tendons is required, the
dorsalis pedis flap can be used with careful management
of the donor site. The radial forearm cutaneotendinous
flap can also be applied in cases of one or two extensor
tendon reconstruction, as much as it is a simpler tech­
nique, with no microsurgical risks, and followed by
minor complications. The dorsalis pedis flap and radial
forearm flap are less often used now because of the donor
site morbidity.
Based on our preliminary results in a series of three
patients, we believe that radial island fasciotendinous
flap represent an excellent evolution for coverage of
soft tissue defects of the hand and we recommend the
use of this composite fascial flap in the reconstruction
of cutaneotendinous injuries of the dorsum of the
hand that require the use of two or three tendon grafts.
Harvesting a small dorsalis pedis flap or ulnar side
island flap for reconstruction of extensor tendon in the

401
Chapter 37:  Vascularized Tendon Graft for Extensor Tendon Reconstruction
Figure 37-8  A patient with injury to the dorsum of the hand. A, Soft tissue defect located on the dorsum of the hand with
absence of extensor tendons from the index and middle finger. B, The forearm fascia flap was raised on a distal pedicle with a
strip of the brachioradialis and flexor carpi radialis tendons. C and D, The flap was transferred to the dorsum of the hand and
is covered with a full-thickness skin graft. E and F, Aesthetic and functional results. G, Appearance of the donor site in the
forearm.
A
B
C
D
E
G
F
dorsum of the digits remains a practical and attractive
option.
The replacement of combined loss of skin and
tendons on the dorsum of the hand with vascularized
composite tissue grafting in one stage is expected to
confer the best opportunity to patients for functional
recovery, allowing patients a relatively rapid return to a
normal life.
1. Dessai SS, Chuang DC, Levin SL: Microsurgical reconstruc­
tion of the extensor system, Hand Clin 11:471–482, 1995.
2. Winspur I: Distant flaps, Hand Clin 1:729–739, 1985.
3. Soutar DS, Tanner NS: The radial forearm flap in the manage­
ment of soft tissue injuries of the hand, Br J Plast Surg 37:18–
26, 1984.
4. Jones NF, Jarrahy R, Kaufman MR: Pedicled and free radial
forearm flaps for reconstruction of the elbow, wrist, and
hand, Plast Reconstr Surg 121:887–898, 2008.
5. Grobbelaar AO, Harrison DH: The distally based ulnar artery
island flap in hand reconstruction, J Hand Surg (Br) 22:204–
211, 1997.
References
6. Schecker LR, Langley SJ, Martin DL, et al: Primary extensor
tendon reconstruction in dorsal hand defects requiring free
flaps, J Hand Surg (Br) 18:568–575, 1993.
7. Sundine M, Scheker LR: A comparison of immediate staged
reconstruction of the dorsum of the hand, J Hand Surg (Br)
21:216–221, 1996.
8. Jeng SF, Wei FC, Noordhoff MS: The composite groin fascial
free flap, Ann Plast Surg 35:595–600, 1995.
9. Weinzweig N, Chen L, Chen ZW: The distally based radial
forearm fasciosubcutaneous flap with preservation of the
radial artery: An anatomic and clinical approach, Plast
Reconstr Surg 94:675–684, 1994.

402
Section 4:  Extensor Tendon Repair and Reconstruction
10. Carty MJ, Taghinia A, Upton J: Fascial flap reconstruction of
the hand: A single surgeon’s 30-year experience, Plast Reconstr
Surg 125:953–962, 2010.
11. Buehler MJ, Pacelli L,Wilson KM: Serratus fascia “sandwich”
free-tissue transfer for complex dorsal hand and wrist avul­
sion injuries, J Reconstr Microsurg 15:315–320, 1999.
12. Brody GA, Buncke HJ, Alpert BS, et al: Serratus anterior
muscle transplantation for treatment of soft tissue defects in
the hand, J Hand Surg (Am) 15:322–327, 1990.
13. Parrett BM, Bou-Merhi JS, Buntic RF, et al: Refining outcomes
in dorsal hand coverage: Consideration of aesthetic and donor
site morbidity, Plast Reconstr Surg 126:1630–1638, 2010.
14. Taylor GI, Townsend P: Composite free flap and tendon trans­
fer: An anatomical study and a clinical technique, Br J Plast
Surg 32:170–183, 1979.
15. Hentz VR, Pearl RM: Hand reconstruction following avulsion
of all dorsal soft tissues: A cutaneo-tendinous free tissue
transfer, Ann Chir Main 6:31–37, 1987.
16. Vila-Rovira R, Ferreira BJ, Guinot A: Transfer of vascularized
extensor tendons from the foot to the hand with a dorsalis
pedis flap, Plast Reconstr Surg 76:421–427, 1985.
17. Caroli A, Adani R, Castagnetti C, et al: Dorsalis pedis flap with
vascularized extensor tendons for dorsal hand reconstruction,
Plast Reconstr Surg 92:1326–1330, 1993.
18. Lee KS, Park SW, Kim HY: Tendocutaneous free flap transfer
from the dorsum of the foot, Microsurg 15:882–885, 1994.
19. Cho BC, Lee JH, Weinzweig N, et al: Use of free innervated
dorsalis pedis tendocutaneous flap in composite hand recon­
struction, Ann Plast Surg 40:268–276, 1998.
20. Reid CD, Moss LH: One-stage flap repair with vascularized
tendon grafts in a dorsal hand injury using the “Chinese”
forearm flap, Br J Plast Surg 36:473–479, 1983.
21. Yajima H, Inada Y, Shono M, et al: Radial forearm flap with
vascularized tendons for hand reconstruction, Plast Reconstr
Surg 98:328–333, 1996.
22. Adani R, Marcoccio I, Tarallo L: Flexor coverage of dorsum of
hand associated with extensor tendons injuries: A completely
vascularized single-stage reconstruction, Microsurg 23:32–39,
2003.
23. Glasson DW, Lovie MJ: The ulnar island flap in hand and
forearm reconstruction, Br J Plast Surg 41:349–353, 1988.
24. Guimberteau JC, Panconi B, Boileau R: Mesovascularized
island flexor tendon: new concepts and techniques for flexor
tendon salvage surgery, Plast Reconstr Surg 92:888–903, 1993.
25. Koul AR, Patil RK, Philip V: Complex extensor tendon inju­
ries: early active motion following single-stage reconstruc­
tion, J Hand Surg (Eur) 33:753–759, 2008.
26. Cautilli D, Schneider LH: Extensor tendon grafting on the
dorsum of the hand in massive tendon loss, Hand Clin
11:423–429, 1995.
27. Tomaino MM: Treatment of composite tissue loss following
hand and forearm trauma, Hand Clin 15:319–333, 1999.
28. Samson MC, Morris SF, Tweed AE: Dorsalis pedis flap donor
site: Acceptable or not? Plast Reconstr Surg 102:1549–1554,
1998.
29. Chen HC, Buchman MT, Wei FC: Free flaps for soft tissue
coverage in the hand and fingers, Hand Clin 15:541–554,
1999.
30. Murray RC, Gordin EA, Saigal K, et al: Reconstruction of the
radial forearm free flap donor site using integra artificial
dermis, Microsurg 31:104–108, 2011.
31. Adani R, Tarallo L, Marcoccio I: Island radial artery fascioten­
dinous flap for dorsal hand reconstruction, Ann Plast Surg
47:83–85, 2001.

403
SECTION
5
REHABILITATION OF
TENDON SURGERY

This page intentionally left blank

405
CHAPTER
38 
STATE OF THE ART FLEXOR
TENDON REHABILITATION
Karen M. Pettengill, MS, OTR/L, CHT, and
Gwendolyn Van Strien, LPT, MSc
success of the repair. In discussing the various rehabilita­
tion programs available, we will focus almost entirely
on the first few weeks, with discussion of basic concepts
and techniques, and rationale for advancing the program
according to individual patient and surgical factors. We
will also examine emerging trends and issues in flexor
tendon rehabilitation, raising questions we hope to see
answered in the years to come.
In 1917, Harmer1 published his results of uncon­
trolled early active flexion and extension following
flexor tendon repair. Over the course of the past 100
years, we have come full circle in our approach: For a
long period, injured flexor tendons in the hand were
immobilized or resected, with later grafting. In the
1960s, passive mobilization programs were in vogue,
and today we have returned with greater sophistication
to active mobilization as the preferred approach. None­
theless, there is limited high-quality evidence support­
ing one form of early mobilization over another.2,3
IMMOBILIZATION
The more conservative approaches of immobilization
and/or resection and grafting were a response to the
complexity of the flexor tendons in zone 2, known as
“no man’s land.” Surgeons were understandably hesi­
tant to perform a repair at that level because of the
unacceptably high risk of rupture or development of
restrictive adhesions. In 1941, Mason and Allen4 found
that in immobilized repairs there was an initial drop in
repair strength, followed by a gradual increase. At 3
weeks, repair strength was equivalent to that on the first
postoperative day. Because of the observed dip in
strength, it was considered necessary to protect the
repair with complete immobilization for 3 weeks. Since
then research has supported the value of controlled
early mobilization of repairs. Even though immobiliza­
tion is no longer the preferred approach, there are many
cases in which this is the treatment of choice (e.g., in
young children).
When a repair must be immobilized, the cast or
splint is designed to keep the repaired flexor tendon on
OUTLINE
There are three approaches to management of the
repaired flexor tendon during the first few weeks:
immobilization, passive mobilization (passive flexion
and active or passive extension), and active flexion
initiated within the first several days. The “classic”
early passive mobilization programs have been modi­
fied, improving results. While the literature now sup­
ports the efficacy of early active mobilization over
early passive and early passive mobilization over
immobilization, there is little evidence behind indi­
vidual techniques. Rehabilitation programs must be
individually designed, based on such factors as char­
acteristics of injury and repair, concomitant injuries,
patient access to therapy, and patient ability to perform
a given program. This requires close surgeon–therapist
collaboration. There are a number of ways to safely
progress active flexion programs and to provide splint
protection. Splints (or orthoses) are dorsal and provide
some protection from excessive extension through
metacarpophalangeal joint and wrist position; the
traditional wrist and finger positions have been modi­
fied over the years. One published active flexion
plan involves a splint with a wrist hinge. Growing
evidence suggests greater incorporation of neuro­
muscular theory and purposeful activity into therapy,
preventing loss of cortical representation of involved
digits and facilitating recovery of fluid functional
motion. We need more specific measures of functional
outcomes for both surgery and therapy, to better rep­
resent the success of each individual’s rehabilitation.
Methods of rehabilitation after flexor tendon repairs are
evolving. Earlier chapters examined the anatomy,
healing, surgical repair and other aspects in great detail,
offering a biological, mechanical, and surgical basis for
establishment of current rehabilitation principles and
methods. It is well recognized that the first few weeks
following repair are crucial in flexor tendon rehabilita­
tion: Management during this phase will determine the

406
Section 5:  Rehabilitation of Tendon Surgery
(Figure 38-1).10 Some immobilize the interphalangeal
joints in extension at night, and others use the splint
design proposed by May and colleagues,11 with the
dorsal finger component ending at the PIP joint for
optimal PIP joint extension. At night a separate volar
piece is strapped on to the splint to keep the IP joints
in extension.11 In the same study, May and colleagues
examined the issue of using dynamic flexion for all four
fingers versus only the involved finger.11 They concluded
that incorporation of all four fingers contributed to
improved results by making passive flexion easier to
attain. Other authors10 have recommended incorporat­
ing only the involved finger in traction. Some practitio­
ners apply traction to only the involved finger(s) for
flexor digitorum profundus (FDP) repairs, in order to
facilitate inadvertent “cheating” with active motion of
the uninvolved providing limited active motion of the
involved finger(s).
To provide greater passive excursion of the FDP
through flexion of the distal interphalangeal (DIP)
joint, a palmar pulley can be added to redirect the
line of traction to the distal palmar crease (Figure
38-2).10,12,13 To mobilize the flexor pollicis longus (FPL),
the thumb MCP joint is immobilized and the IP joint
must be passively flexed, with active or passive exten­
sion (Figure 38-3).14
Duran’s approach7 involved less frequent mobiliza­
tion entirely through passive flexion and extension. His
exercises were designed to provide passive excursion of
the repair relative to surrounding tissues and differen­
tial glide between the FDP and flexor digitorum super­
ficialis (FDS) in zones 2 and 3. In 1980 Strickland and
Glogovac15 published a study showing improved results
in zone 2 repairs mobilized passively compared with
slack by positioning the wrist at neutral to 20° to 30°
of flexion, the metacarpophalangeal (MCP) joints in
40° to 70° of flexion, and the interphalangeal (IP)
joints at 0°. Following 3 weeks of immobilization, there
are two predictable problems. Since immobilized repairs
grow weaker before they return to their immediate post­
operative strength, we can expect a higher risk of rupture
with active motion in these patients. At the same time,
frustratingly, restrictive tendon adhesions will have
developed. Therapy in these cases focuses on increasing
tendon excursion while protecting the repair from
excessive stress. A 1991 article by Cifaldi Collins and
Schwarze5 describes a systematic approach to regaining
tendon function following initial immobilization of
repairs in zones 2 and 3. They provide timing guidelines
for initiating tendon gliding exercises, blocking and
other specifically targeted interventions to restore gliding
without overstressing the immobilized repair.
CONTROLLED EARLY MOBILIZATION
In earlier chapters, authors have covered the beneficial
effects of early motion on tendon healing and biome­
chanics, and the supporting research showing that early
mobilization of the repair provides at least a partial
solution to the problems of immobilization: better
adhesion control and increased repair strength.
EARLY PASSIVE MOBILIZATION
Before any significant research had been performed in
early mobilization of tendon repairs, forward-thinking
surgeons such as Kleinert6 and Duran7 had designed
postoperative management programs incorporating
controlled passive motion of the repair in the first 3 to
4 weeks. These programs involved protecting the tendon
repair by placing it on slack in a dorsal protective splint
in some degree of wrist and MCP joint flexion as
described above. For both programs, elastic traction
held the fingers in flexion. Therapy consisted of passive
flexion and either active or passive extension, to provide
passive tendon gliding of the flexor tendon with minimal
stress to the repair.
Kleinert’s technique, designed for zone 2 and which
he also advocated for zone 1, involved hourly active
extension against dynamic traction, followed by relax­
ation of the finger and passive flexion by means of the
elastic traction, premised on the theory that the flexors
would relax with resisted extension. However, later
research8,9 has shown that the desired relaxation of the
flexors does not occur consistently.
Over the years various authors have proposed modi­
fications of the Kleinert approach. To prevent or mini­
mize the risk of proximal interphalangeal (PIP) joint
flexion contractures, elastic resistance to extension
can be decreased by manually releasing rubber band
traction during active extension or by using a lighter
rubber band for exercises as in the Washington regimen
Figure 38-1  In this modification of the Kleinert program,
the patient manually releases rubber band traction during
extension to facilitate full PIP extension in an effort to avoid
development of a PIP flexion contracture.

407
Chapter 38:  State of the Art Flexor Tendon Rehabilitation
authors have looked at the stresses placed on the tendon
and the amount of excursion attained with SWM.9,17,18
Amadio and Tanaka19,20 have proposed a further modi­
fication on a passive SWM exercise. The wrist is in flexion
initially, with the fingers in extension, followed by
passive finger flexion, wrist extension and MCP joint
extension (holding the IP joints in flexion). The combi­
nation of wrist extension and MCP joint hyperextension
with the DIP and PIP joints fully flexed produces an
increase in flexor tendon tension. They hypothesize that
in vivo this small increase in tendon tension may
increase FDP excursion. An added benefit would be
control of any intrinsic muscle tightness developing in
response to blocking the MCP joints in flexion in the
splint.
EARLY ACTIVE MOBILIZATION
Although early passive mobilization produced improved
results,15 surgeons and therapists questioned whether
they could attain even better functional results with
techniques producing greater excursion of the repair
site. The model of passive excursion is logical and clearly
safe in theory, but in practice the repaired tendon and
adjacent tissues may be edematous, with added bulk
from sutures and related injuries. In those cases passive
flexion may simply fold the tendon or cause it to bunch
up due to the gliding resistance it encounters.
Once again, clinical development preceded support­
ing research: Surgeons and therapists began using con­
trolled active flexion to attempt to provide gentle
proximal tendon glide more effectively than was possi­
ble with passive flexion. As new suture techniques were
developed to provide improved strength and gliding
characteristics, therapists became more and more com­
fortable using active flexion for repair mobilization.
Hitchcock and colleagues21 showed that controlled
active mobilization actually increased the strength of
the repair in the first few postoperative days, in contrast
to the accepted dip in strength demonstrated in immo­
bilized repairs.
Savage22 studied the effect of wrist position on active
tension on the flexor tendon, or the force required to
flex the interphalangeal joints against the passive resis­
tance of the extensors. He found that the smallest
amount of force was required with the wrist in 45° of
extension, and the greatest amount of force was required
when the wrist was in 45° of flexion. Since then, other
authors23-28 have explored this problem in greater depth,
taking into account the total force exerted by the repaired
tendon, due to edema, joint stiffness, and the gliding
resistance at the tendon–pulley interface. In practice,
therapists must deal with the total work of flexion with
all of its components.
Savage’s work22 supported the concept of either posi­
tioning the wrist in extension during flexor mobiliza­
tion or incorporating synergistic wrist motion with
immobilized repairs. For that study they used a modi­
fied version of the Duran approach, now known as
modified Duran. They eliminated the rubber band trac­
tion, splinting fingers in IP joint extension between
exercises, modified the wrist flexion angle in the splint,
and added composite passive flexion and active exten­
sion to the original exercises. Over the years the Strick­
land and Glogovac approach has been further modified
to include more frequent mobilization and other
variations.
A study by Cooney and colleagues16 found that incor­
porating synergistic wrist motion (SWM) would increase
passive flexor tendon excursion in zones 2, 3, and 5:
extending the wrist assists passive digit flexion via teno­
desis, while wrist flexion during digit extension uses
tenodesis to decrease stress on the repaired flexor(s) and
allows greater digit extension. Since then a number of
Figure 38-2  A palmar pulley at the distal palmar crease
redirects the line of traction to attain DIP flexion and thus
attain greater excursion of FDP tendon.
Figure 38-3  For FPL repairs, the MCP joint should be
immobilized in extension for IP joint passive flexion (whether
applied by elastic traction or manually).

408
Section 5:  Rehabilitation of Tendon Surgery
weeks. Some of the interventions used by experienced
clinicians (functional activities, blocking exercises,
splinting to address joint and tendon tightness) were
introduced later than either of these authors would have
expected. An experienced clinician progresses the
program according to evaluation of tendon function:
e.g., slowing down the progression if tendon excursion
is returning rapidly and treating more aggressively in
the presence of severe adhesions. Although it may be
that the therapists overseeing the passive mobilization
program took some liberties with the stated timetable,
the article does not say so, leaving the impression that
all of the passively mobilized tendons were moved with
similar caution. If that is the case, then these patients
might have achieved better results if treated according
to current best practice in passive mobilization. This
diminishes the implications of the study, but not the
importance of its publication.
In choosing an appropriate early active mobilization
technique, there are several factors to consider: patient
finger flexion and extension. As a result, early mobiliza­
tion programs were adapted to bring the wrist into less
flexion, and in some cases, into extension. More recently
a study by Kursa and his colleagues9 provided additional
evidence, at least for FDS, for which greater forces were
recorded with finger flexion and extension when the
wrist was at 30° of flexion.
Active mobilization programs can be divided into
three basic approaches: active,29-32 assisted active,33 and
place-hold (active-hold).34-36 The first two programs are
self-explanatory. In place-hold programs the involved
fingers are placed in flexion and the patient holds the
position with a gentle muscle contraction. Assisted
active and place-hold flexion exercises are designed to
decrease the work of flexion by overcoming passive
resistance manually. All of these approaches assume
that the tendon will be protected through splint posi­
tion, and that work of flexion will be reduced through
edema control and passive exercises performed prior
to active flexion. Most published programs include a
dorsal protective splint with the MCP joints in some
degree of flexion, IP joints allowed to extend to 0° and
the wrist in slight flexion, neutral or slight extension to
decrease the work of flexion.29,32,33,35,36 One published
program, the Indianapolis program,34 adds an exercise
splint with a hinged wrist, allowing wrist extension to
30°, to incorporate SWM and thus both increase excur­
sion and decrease work of flexion (Figures 38-4 to
38-6). Although the authors do not state that the
program is for zone 2 repairs, the program is based on
the issues of excursion and work of flexion crucial to
rehabilitation of zone 2 injuries.
A recent study3 showed clearly superior results in
zone 2 flexor tendon repairs treated with active versus
passive mobilization. The authors used a hinged-wrist
splint program for active mobilization and a modified
Kleinert program for passive mobilization. The passive
mobilization program was conservative, starting active
flexion using place-hold with the wrist at neutral (within
the splint), and delaying true active flexion until 6
Figure 38-4  The patient extends the wrist actively with
simultaneous passive digit flexion. (Modified from Cannon N:
Post flexor tendon repair motion protocol, Indiana Hand
Center Newsletter 1:13-18, 1993.)
Figure 38-5  The patient maintains digit flexion with a
gentle active muscle contraction. (Modified from Cannon N:
Post flexor tendon repair motion protocol, Indiana Hand
Center Newsletter 1:13-18, 1993.)
Figure 38-6  The wrist is allowed to relax into flexion
with simultaneous digit extension (limited to 60° at the
metacarpophalangeal joints). (Modified from Cannon N: Post
flexor tendon repair motion protocol, Indiana Hand Center
Newsletter 1:13-18, 1993.)

409
Chapter 38:  State of the Art Flexor Tendon Rehabilitation
flexion (place-hold) with “minimal active muscle
tension” (MAMT) through partial digital flexion and
slight wrist extension.
Another convenient way to gradually increase the
amount of flexion over the first 4 weeks was proposed
by Sheila Harris (personal communication, 2010). The
patient is asked to place the four fingers of the uninjured
hand in the palm of the injured hand with the little
finger resting at the distal palmar crease and the fingers
perpendicular to the plane of the palm. In the first week
the patient flexes to touch the index finger. In the second
week the patient is asked to move away the index and
try to touch the middle finger (Figure 38-7). In the third
and fourth week the patient is asked to flex to touch the
ring and small finger respectively in the same way. A
useful modification was added by van Strien in order to
add more DIP joint flexion. By leaving the four fingers
of the uninjured hand in place and sliding the fingertip
across the dorsum of the fingers to reach the next finger
more DIP joint flexion is achieved (Figure 38-8). Not
only does this method gradually increase the force
exerted through the tendon, but with the modification
it also encourages the patient to flex both the PIP and
the DIP joints, thus theoretically achieving greater FDP
excursion and differential FDP/FDS glide.
Place-hold (active-hold) flexion may decrease the
force of muscle contraction, but only if the fingers are
partially flexed (as noted earlier), or if the wrist is syner­
gistally extended to assist passive proximal excursion. If
the fingertips are flexed to the DPC passively before con­
tracting the flexors to hold the position, there is a risk of
increasing the stress to the tendon if the passive flexion
produced only buckling or folding of the tendon (which
must now be unfolded, requiring a theoretically greater
amount of force than would be required for unassisted
selection, ways to control force of contraction (primarily
through wrist and digital position), timing of initiating
motion,24,25,28,37 and methods of reducing work of flexion
prior to exercise.
PATIENT SELECTION
Patient characteristics are difficult or impossible to
control, and should dictate the choice of program.
Therefore it is crucial to choose an approach that fits the
patient, rather than imposing unrealistic expectations
on a patient incapable of performing a complex
program. Recent research has found that a high percent­
age of patients are noncompliant with their home reha­
bilitation programs.38,39 In fact, in one study 67% of
patients with flexor tendon repairs removed the splint
against therapist instruction.39 The most common
reason given was hand washing: an all-too-familiar
story to an experienced hand therapist!
CONTROLLING FLEXION FORCE
There are a number of different ways to ensure that the
patient exerts the minimal flexion force needed to
achieve sufficient tendon excursion. Studies have shown
that placing the wrist in flexion increases the force
required to flex the fingers.9,22 Placing the wrist in neutral
or extension reduces the force required to flex the fingers.
Incorporating synergistic wrist extension with finger
flexion favors proximal gliding of the flexor tendons and
decreases the force required to flex the fingers. Forces
transmitted through the flexor tendons increase with
increasing active flexion of the fingers and wrist.9 To
determine the optimum splint position (or to choose
whether or not to use an SWM program), the therapist
must consider zone and type of injury. For example,
placing the wrist in slight extension benefits gliding of
the tendon repaired in zone 2 in the digits and reduces
the forces applied to the tendon during active finger
flexion. Nevertheless, some wrist flexion or a more
limited range of wrist synergistic motion should be
chosen to protect multiple nerve and tendon repairs in
zone 5, avoiding nerve compression by limiting degree
of wrist flexion, and limiting passive traction on nerve
and tendon repairs by limiting degree of synergistic
wrist extension. A lesser degree of MCP joint flexion in
the splint is appropriate in zone 2 and 3 injuries to
avoid development of intrinsic muscle tightness in con­
junction with adhesion formation.
Some authors recommend gradually increasing the
IP joint flexion demands over time. The Belfast proto­
col29,32 for zone 2 repairs sets increasing range of flexion
goals from week to week. Evans and Thompson35 used
a biomechanical model to calculate the force placed on
flexor tendon repairs at various points in the range of
flexion and concluded that a partial fist places consider­
ably less stress on the repair than does a full fist. On this
basis they proposed a program designed to elicit active
Figure 38-7  If the patient flexes to touch the second
finger with the first finger out of the way, the patient will
automatically tend to flex the PIP and MCP joints instead of
the DIP and PIP joints.

410
Section 5:  Rehabilitation of Tendon Surgery
edema cannot be adequately controlled, there is a higher
risk of rupture.
FUNDAMENTAL CONCEPTS AND TECHNIQUES
OF EVALUATION AND TREATMENT
Choosing the most appropriate rehabilitation approach
requires an understanding of all the factors affecting
tendon healing and recovery of function. The mecha­
nism of injury will affect both surgical and therapy deci­
sions. A crush injury, for example, has a higher risk of
formation of restrictive adhesions, whereas a clean lac­
eration may be more easily repaired and may recover
function more easily. The effect of the level of injury is
discussed in Chapter 39.
The therapist must know what kind of repair was
performed. Despite all the recent evidence cited in pre­
vious chapters supporting the use of four-strand core
sutures and strong peripheral sutures for early active
flexion programs, many surgeons use a two-strand
Kessler and a simple running peripheral suture. In a
2006 survey of members of the Irish Hand Surgery
Society, Healy and colleagues40 found that a clear major­
ity preferred the latter to the former. In addition to the
type of repair, the therapist should be informed of any
intraoperative findings that may influence clinical deci­
sions, such as concomitant injuries, repairs performed
under tension, pulley resection, venting or repair, or
other repair characteristics that may affect gliding or
strength of the repair. Gliding resistance from surround­
ing tissues is not the only reason why active or passive
flexion may be difficult to attain. Postoperative edema
and joint stiffness in particular pose resistance to
motion, increasing the risk of rupture or repair deforma­
tion as well as decreasing repair mobility.
Amadio19 has proposed the concept of a safe zone,
which allows the therapist and surgeon to make deci­
sions about mobilizing tendons based on the various
factors increasing work of flexion (Figures 38-9 and
38-10). For example, if the repair is strong but bulky, it
will withstand the stress of active motion but will glide
poorly, and may even “catch” on surrounding tissues,
thereby exceeding its resistance to rupture. This would
lower the upper limit of the safe zone and could pre­
clude active motion. Likewise, if edema and joint stiff­
ness are not well controlled, the passive resistance to
flexion increases, and again may preclude safe active
mobilization. Recent work of Cao and colleagues and
Tang and colleagues has demonstrated increased tendon
repair strength and decreased gliding resistance with
partial excision of the A2 pulley,41,42 which suggests that
venting pulleys may further widen the “safe zone” as
described by Amadio.19
Such factors make it imperative that the therapist
gather all pertinent information through clear commu­
nication with the referring surgeon. Ideally, all hand
therapists should receive a copy of the operative report
active flexion). We propose that assisted active flexion is
a safer way to decrease force of contraction.
Before active flexion exercises, work of flexion must
be minimized by decreasing the resistance to passive
flexion, not only through positioning to minimize resis­
tance from the extensors but also through reducing
resistance caused by edema and passively ranging the
finger joints to reduce stiffness. Cao and Tang24 found
that six cycles of digit motion in a chicken model sig­
nificantly decreased work of flexion.
INITIATING ACTIVE MOBILIZATION
Determining the best time to initiate active mobiliza­
tion is difficult, due to both conflicting evidence in the
literature24,25,27,28 and variability of patient healing char­
acteristics. It is widely accepted that active mobilization
is both safe and effective only if started within the first
postoperative week. In a recent study in a chicken
model, Tang and colleagues24 divided the early postop­
erative phase into three stages: days 0 to 3, increasing
resistance due to postoperative inflammatory edema;
days 4 to 7, further increase in resistance from edema;
and after days 7 to 9, hardening of the subcutaneous
tissue with development of adhesions. On the basis of
this and other studies,25,27,28,37 it is reasonable to assume
that in an otherwise appropriate patient, active flexion
may begin at 3 to 4 days, if edema and wound healing
appear well under control.
All of these studies further underline the importance
of controlling edema in order to decrease work of flexion
in the crucial early days. A patient with an unusually
edematous finger or other complications of healing
presents a dilemma: If motion is delayed for too long,
adhesions will have formed and subcutaneous tissue
will have hardened, but if motion begins early and
Figure 38-8  With the first finger kept in place, the patient
is encouraged to slide the fingertip down to the second
finger, thereby using DIP and PIP joint flexion rather than
just PIP and MCP joint flexion.

411
Chapter 38:  State of the Art Flexor Tendon Rehabilitation
response to intervention as absent, responsive, and unre­
sponsive and suggested clinical observations that may
help assess adhesion severity. This classification is
further clarified and presented in Table 38-1. Groth’s
system helps the therapist decide how to advance up the
proposed pyramid of force progression (graded passive
and active mobilization) safely and effectively (Figure
38-11).43 As published the pyramid does not address the
progression of synergistic wrist motion programs; at the
Annual Meeting of the American Society of Hand Thera­
pists in Boston in 2008, von der Heyde proposed a
modification of the pyramid that incorporates wrist
motion during the early weeks of therapy and proposes
progression to straight fist (DIP joints in extension as
MCP joints and PIP joints flex) and hook fist (MCP
joints in extension, IP joints flexed) before full fist to
decrease work of flexion in the early postoperative
period.
Coert and colleagues44 used positron emission tomog­
raphy (PET) imaging to determine cortical activity fol­
lowing 6 weeks of passive mobilization of a repaired
flexor tendon. In these patients changes in cortical func­
tion were evident, showing inefficient motor control
apparently due to deprivation of sensorimotor input.
Once active flexion was initiated, cortical function grad­
ually returned to normal, reflecting more efficient motor
control. The authors suggest that this inefficiency might
be avoided with active mobilization or use of other
techniques such as mirror therapy (with active flexion
of the contralateral hand) combined with passive mobi­
lization during the first 6 postoperative weeks.
as well as discussing with the surgeon any unusual
aspects of history, injury, and repair.
Observation and palpation are key skills for identify­
ing edema, joint stiffness, and wound complications
that could influence treatment planning. Edema and
range of motion should be measured objectively as dic­
tated by the characteristics of the injury and repair, and
the therapist must repeat measurement periodically to
assess the effectiveness of treatment and to determine
how best to advance or modify the program. Groth43
proposed a classification of adhesion severity and
Figure 38-9  Small safe zone for a four-strand repair with
40 N breaking strength. (Modified from Amadio PC: Friction
of the gliding surface. Implications for tendon surgery and
rehabilitation, J Hand Ther 18:112-129, 2005.)
Max strength in vitro
Max strength in vivo
2 mm gap
Initial gap
In vivo friction
Repair friction
Normal friction
0.2
5
10
20
30
40
Force (N)
Strong grip
Light grip
Light active
Synergistic
Passive motion
Safe zone
Figure 38-10  Larger safe zone with increased suture repair
strength and faster healing, and reduced friction through the
use of lubricants and repairs with smooth gliding surfaces.
(Modified from Amadio PC: Friction of the gliding surface.
Implications for tendon surgery and rehabilitation, J Hand
Ther 18:112-129, 2005.)
Max strength
in vitro
Max strength
at 1 week in vivo
2 mm gap
Initial gap
In vivo repair
In vitro repair
Improve strength:
Faster healing
Stronger repairs
Reduce friction:
Lubricants
Smoother repairs
Normal friction
0.2
5
10
20
30
Force (N)
40
Table 38-1  Assessment of Adhesions Based on
Clinical Findings (Modified Based on Groth)
Adhesions*
Clinical Findings†
Absent
Tendons glide well; <5° discrepancy
between total active and passive
digital flexion.
Responsive
≥10% decrease in active motion lag
is observed between one session and
the next‡
Unresponsive
<10% decrease in active motion lag
is observed between one session and
the next‡
*When adhesions are considered absent or responsive, the therapy
program does not need upgrading. When adhesions are
unresponsive, the therapist should consider upgrading the
program.
†Clinical findings were reworded from the Groth’s description for
clarity based on the clinical judgment of the present authors.
‡Because frequency of therapy appointments varies according to
many factors, therapists should apply the definition of “responsive”
or “unresponsive” with due consideration to the number of days
between sessions.

412
Section 5:  Rehabilitation of Tendon Surgery
and basic science regarding the thermal and nonthermal
effects of ultrasound.
For that matter, there is insufficient evidence support­
ing use of any one tendon rehabilitation protocol over
others.2 With the exception of the Evans protocol for
zone 1 repairs, the published protocols either are
intended for zone 2 repairs or do not specify the zone.
The therapist must choose and adapt as needed the
program most appropriate to the patient. This may be
dependent on many variables, not the least of which is
cultural variation and differences in health care system
from one country to another. For example, some pub­
lished active mobilization protocols require closer mon­
itoring than others (some including postoperative
hospitalization), and may be feasible only within a
health care system that pays for frequent therapy or
postoperative hospitalization. Those protocols should
be adapted if used with a different patient population
or within a different health care system.
In comparing publications on surgical techniques
and rehabilitation programs, the reader must be alert to
the different systems that can be used to assess clinical
outcomes of flexor tendon repair. The most commonly
accepted and most logical is the modified Strickland
formula, in which the total active motion (TAM) of the
involved digit (excluding MCP joint motion, which may
be within normal limits even with severely adherent
repairs) is expressed as a percentage of normal. To
Many patients have a tendency to co-contract intrin­
sic or extrinsic extensor muscles in their first attempts at
active flexion. This can be very frustrating for the patient
and increases the difficulty of attaining flexor tendon
excursion. One way to overcome the problem is to
incorporate purposeful, goal-oriented activities, which
elicit more fluid and less effortful motion than do exer­
cises.45,46 For example, when asked to bend the tip of the
thumb following FPL repair, a patient may tend to
co-contract or use excessive force. When asked to encir­
cle a dowel with thumb and index finger, the same
patient, focusing on the task rather than the motion,
may easily and fluidly incorporate all involved muscles
in a functional tip pinch position including thumb IP
joint flexion. Once this position has been attained, the
patient can be asked to gently scratch the tip of the index
with the thumb fingernail, thus eliciting greater IP joint
flexion. Some patients benefit from performing the
same task first with the uninvolved hand or perhaps
simultaneously with both hands.
QUESTIONS AND ISSUES
There are many exercises, activities, splints, and adjunc­
tive modalities commonly used and anecdotally suc­
cessful. However, there is a dearth of evidence supporting
most of these techniques. For example, many therapists
routinely use ultrasound to aid recovery of tendon glide,
but this is supported only by extrapolation of theory
Figure 38-11  Progressive application of force to the repaired intrasynovial flexor tendon. (Modified from Groth G: Pyramid of
progressive force exercises to the injured flexor tendon, J Hand Ther 17:31-42, 2004.)
Resistive
isolated
joint motion
Resistive hook and
straight fist
Resistive composite fist
Discontinuation of protective splint
Isolated joint motion
Hook and straight fist
Active composite fist
Place and hold
Passive protected extension
Pyramid of progressive force application
Wrist unprotected
Wrist protected

413
Chapter 38:  State of the Art Flexor Tendon Rehabilitation
falling into disfavor, both because the rationale for its
use is in question and because it increases the risk of
PIP joint flexion contractures. Early active flexion is
proving to be the most effective approach, with mobi­
lization starting within the first postoperative week. The
safety of active flexion is contingent on controlling
edema and other elements that pose resistance to tendon
glide and thus dramatically increase the work of flexion.
Safety is also dependent on very gradually increasing the
motion demanded of the finger as tissue healing and
gliding resistance dictate. We can augment excursion
further by incorporating synergistic wrist motion and
other specific maneuvers such as passive MCP joint
extension. The successful flexor tendon repair depends
not only on surgical and therapeutic expertise, but also
on patient selection, ample communication between
surgeon and therapist, and careful and continual reas­
sessment of tendon function to determine when and
how to progress the rehabilitation program.
specifically evaluate FDP tendon function, DIP joint
flexion must be given greater weight, as suggested by
Moieman.47 None of the existing clinical outcomes
methods takes into account the position of the wrist
during measurement of finger flexion. Outcomes assess­
ments based on range of motion alone do not take into
account all of the factors (such as coordination, quad­
riga effect, individual digit differences, and patient per­
ception of function) that determine success. Existing
functional outcomes instruments such as the Disability
of the Arm, Shoulder, and Hand (DASH) questionnaire
are not specific enough to be of much value in evalua­
tion of flexor tendon repairs.
SUMMARY
Several distinct trends are apparent in current postop­
erative management of the repaired flexor tendon.
Dynamic traction into flexion (Kleinert technique) is
References
1. Harmer TW: Tendon suture, Boston Med Surg J 177:808–810,
1917.
2. Thien TB, Becker JH, Theis JC: Rehabilitation after surgery for
flexor tendon injuries in the hand, Cochrane Database Syst Rev
2004:CD003979.
3. Trumble TE, Vedder NB, Seiler JG 3rd, et al: Zone-II flexor
tendon repair: a randomized prospective trial of active place-
and-hold therapy compared with passive motion therapy,
J Bone Joint Surg (Am) 92:1381–1389, 2010.
4. Mason ML, Allen HS: The rate of healing of tendons: An
experimental study of tensile strength, Ann Surg 113:424–
459, 1941.
5. Cifaldi Collins D, Schwarze L: Early progressive resistance
following immobilization of flexor tendon repairs, J Hand
Ther 4:111–113, 1991.
6. Kleinert HE, Kutz JE, Ashbell TS, et al: Primary repair of lacer­
ated flexor tendons in “no-man’s-land,” J Bone Joint Surg (Am)
49:577–584, 1967.
7. Duran R, Houser R: Controlled passive motion following
flexor tendon repair in zones 2 and 3. In AAOS Symposium
on Tendon Surgery in the Hand, St Louis, 1975, CV Mosby,
pp 105–114.
8. van Alphen JC, Oepkes CT, Bos KE: Activity of the extrinsic
finger flexors during mobilization in the Kleinert splint,
J Hand Surg (Am) 21:77–84, 1996.
9. Kursa K, Lattanza L, Diao E, et al: In vivo flexor tendon forces
increase with finger and wrist flexion during active finger
flexion and extension, J Orthop Res 24:763–769, 2006.
10. Dovelle S, Heeter PK: The Washington regimen: rehabilita­
tion of the hand following flexor tendon injuries, Phys Ther
69:1034–1040, 1989.
11. May EJ, Silfverskiold KL, Sollerman CJ: Controlled mobiliza­
tion after flexor tendon repair in zone II: A prospective com­
parison of three methods, J Hand Surg (Am) 17:942–952,
1992.
12. Slattery PG, McGrouther DA: A modified Kleinert Controlled
Mobilization Splint following flexor tendon repair, J Hand
Surg (Br) 9:217–218, 1984.
13. McGrouther DA, Ahmed MR: Flexor tendon excursions in
“no-man’s land,” Hand 13:129–141, 1981.
14. Brown CP, McGrouther DA: The excursion of the tendon of
flexor pollicis longus and its relation to dynamic splintage,
J Hand Surg (Am) 9:787–791, 1984.
15. Strickland JW, Glogovac SV: Digital function following flexor
tendon repair in zone 2: A comparison study of immobiliza­
tion and controlled passive motion, J Hand Surg (Am) 5:537–
543, 1980.
16. Cooney WP, Lin GT, An KN: Improved tendon excursion
following flexor tendon repair, J Hand Ther 2:102–106,
1989.
17. Lieber RL, Amiel D, Kaufman KR, et al: Relationship between
joint motion and flexor tendon force in the canine forelimb,
J Hand Surg (Am) 21:957–962, 1996.
18. Zhao C, Amadio PC, Momose T, et al: Effect of synergistic
wrist motion on adhesion formation after repair of partial
flexor digitorum profundus tendon lacerations in a canine
model in vivo, J Bone Joint Surg (Am) 84A:78–84, 2002.
19. Amadio PC: Friction of the gliding surface. Implications for
tendon surgery and rehabilitation, J Hand Ther 18:112–119,
2005.
20. Tanaka T, Amadio PC, Zhao C, et al: Flexor digitorum pro­
fundus tendon tension during finger manipulation, J Hand
Ther 18:330–338, 2005.
21. Hitchcock TF, Light TR, Bunch WH, et al: The effect of imme­
diate constrained digital motion on the strength of flexor
tendon repairs in chickens, J Hand Surg (Am) 12:590–595,
1987.
22. Savage R: The influence of wrist position on the minimum
force required for active movement of the interphalangeal
joints, J Hand Surg (Br) 13:262–268, 1988.
23. Cao Y, Tang JB: Investigation of resistance of digital subcuta­
neous edema to gliding of the flexor tendon: An in vitro study,
J Hand Surg (Am) 30:1248–1254, 2005.
24. Cao Y, Tang JB: Resistance to motion of flexor tendons and
digital edema: An in vivo study in a chicken model, J Hand
Surg (Am) 31:1645–1651, 2006.
25. Halikis MN, Manske PR, Kubota H, et al: Effect of immobili­
zation, immediate mobilization, and delayed mobilization
on the resistance to digital flexion using a tendon injury
model, J Hand Surg (Am) 22:464–472, 1997.

414
Section 5:  Rehabilitation of Tendon Surgery
combining passive and active flexion, J Hand Surg (Am)
19:53–60, 1994.
37. Xie RG, Cao Y, Xu XF, et al: The gliding force and work of
flexion in the early days after primary repair of lacerated flexor
tendons: An experimental study, J Hand Surg (Eur) 33:192–
196, 2008.
38. Groth GN, Wulf MB: Compliance with hand rehabilitation:
health beliefs and strategies, J Hand Ther 8:18–22, 1995.
39. Sandford F, Barlow N, Lewis J: A study to examine patient
adherence to wearing 24-hour forearm thermoplastic splints
after tendon repairs, J Hand Ther 21:44–52, 2008.
40. Healy C, Mulhall KJ, Bouchier-Hayes DJ, et al: Practice pat­
terns in flexor tendon repair, Ir J Med Sci 176:41–44, 2007.
41. Cao Y, Tang JB: Strength of tendon repair decreases in the
presence of an intact A2 pulley: Biomechanical study in a
chicken model, J Hand Surg (Am) 34:1763–1770, 2009.
42. Tang JB, Cao Y, Wu YF, et al: Effect of A2 pulley release on
repaired tendon gliding resistance and rupture in a chicken
model, J Hand Surg (Am) 34:1080–1087, 2009.
43. Groth GN: Pyramid of progressive force exercises to the
injured flexor tendon, J Hand Ther 17:31–42, 2004.
44. Coert JH, Stenekes MW, Paans AM, et al: Clinical implications
of cerebral reorganisation after primary digital flexor tendon
repair, J Hand Surg (Eur) 34:444–448, 2009.
45. Lin KC, Wu CY, Trombly CA: Effects of task goal on move­
ment kinematics and line bisection performance in adults
without disabilities, Am J Occup Ther 52:179–187, 1998.
46. Ma HI, Trombly CA: The comparison of motor performance
between part and whole tasks in elderly persons, Am J Occup
Ther 55:62–67, 2001.
47. Moiemen NS, Elliot D: Primary flexor tendon repair in zone
1, J Hand Surg (Br) 25:78–84, 2000.
26. Tanaka T, Amadio PC, Zhao C, et al: Gliding resistance versus
work of flexion: two methods to assess flexor tendon repair,
J Orthop Res 21:813–818, 2003.
27. Zhao C, Amadio PC, Paillard P, et al: Digital resistance and
tendon strength during the first week after flexor digitorum
profundus tendon repair in a canine model in vivo, J Bone
Joint Surg (Am) 86A:320–327, 2004.
28. Zhao C, Amadio PC, Tanaka T, et al: Short-term assessment
of optimal timing for postoperative rehabilitation after flexor
digitorum profundus tendon repair in a canine model, J Hand
Ther 18:322–329, 2005.
29. Cullen KW, Tolhurst P, Lang D, et al: Flexor tendon repair in
zone 2 followed by controlled active mobilisation, J Hand
Surg (Br) 14:392–395, 1989.
30. Elliot D, Moiemen NS, Flemming AF, et al: The rupture
rate of acute flexor tendon repairs mobilized by the con­
trolled active motion regimen, J Hand Surg (Br) 19:607–612,
1994.
31. Gratton P: Early active mobilization after flexor tendon
repairs, J Hand Ther 6:285–289, 1993.
32. Small JO, Brennen MD, Colville J: Early active mobilisation
following flexor tendon repair in zone 2, J Hand Surg (Br)
14:383–391, 1989.
33. Sandow MJ, McMahon MM: Single-cross grasp six-strand
repair for acute flexor tenorrhaphy: modified Savage tech­
nique, Atlas Hand Clin 1:41–64, 1996.
34. Cannon N: Post flexor tendon repair motion protocol,
Indiana Hand Center Newsletter 1:13–17, 1993.
35. Evans RB, Thompson DE: The application of force to the
healing tendon, J Hand Ther 6:266–284, 1993.
36. Silfverskiold KL, May EJ: Flexor tendon repair in zone II with
a new suture technique and an early mobilization program

415
CHAPTER
39 
CUSTOMIZING FLEXOR
REHABILITATION BASED ON
ZONE OR TYPE OF INJURY
Fiona H. Peck, MCSP
these injuries require not only specialized knowledge in
the field but also the ability to use advanced clinical
reasoning skills to select appropriate postoperative exer­
cise programmes.3 Early active motion in different forms,
used in conjunction with multistrand repair techniques,
has gained increasing popularity but application must
depend on consideration of a number of variables.1-3
The aim of this chapter is to examine the implications
of these variables on regimen selection and assist the
therapist with the provision of evidence-based practice
following primary flexor tendon repair at all levels.
CUSTOMIZING A REGIMEN
Specialized hand therapy clinics with easy access to
expert medical and nursing staff provide the best envi­
ronment for the treatment of these injuries. Therapists
with the responsibility for the selection of rehabilitation
programs must be well informed with regard to a variety
of different factors. There should be a good working
relationship and excellent communication between the
surgeon and the therapist. Access to accurate medical
records is also vital to the process and clear recording
of operative details will provide the therapist with the
relevant information.
EXTENT OF INJURY
Mechanism of Injury
The mechanism of injury can impact on the outcome of
tendon repairs. Clean cut injuries with the finger in
extension require less extensive surgical exposure and
the tendon can be easily retrieved and repaired with
minimal interference with the surrounding tissues or
loss of tendon length. In uncomplicated tendon injuries
and in the presence of a strong multistrand repair, the
therapist can be more confident in selecting an active
protocol. Ragged saw or crushing injuries involving
bone, joint, and neurovascular structures can be very
different and result in significant postrepair edema and
adhesion formation.
OUTLINE
This chapter details the information that must be
assimilated before selecting an appropriate rehabilita­
tion regimen for flexor tendon repairs at all areas. The
postoperative management of these injuries depends
primarily on the extent of the injury, concomitant
damage to surrounding tissues, surgical methods, and
other factors. The type of tendon injury, timing and
strength of the repair, and the age and characteristics
of the patient all affect the details of the regimen.
Within each zone of injury, a number of anatomical
and biomechanical factors need to be considered in
the decision making process. In the absence of a defin­
itive evidence-based method of rehabilitation after
flexor tendon repairs, the surgeon and the therapist
together must customize the rehabilitation regimen
according to varying patient and surgical factors in
order to achieve favorable outcomes. The chapter also
describes the different types of regimen available,
detailing timing of motion, splintage, and the treat­
ment of associated complications.
In recent years a significant amount of research in the
field of flexor tendon treatment has contributed to
advances in both surgical and rehabilitation tech­
niques.1,2 There is an overwhelming amount of evidence
to show that carefully devised rehabilitation programs
are critical to achieving favorable outcomes following
primary flexor tendon repair in the digits, but very little
attention has been given to other zones of injury in the
hand and forearm. Whatever the level of injury, the
ultimate goal of both the surgeon and the therapist is
to minimize or modify peritendinous adhesions and
promote optimal tendon glide. At the same time the
tendon repair must be protected from excessive stress to
avoid gapping or rupture at the repair site.
Although early mobilization of primary flexor tendon
repairs is now accepted practice, there is no one defini­
tive rehabilitation regimen guaranteed to restore a full
range of active digital motion. Therapists dealing with

416
Section 5:  Rehabilitation of Tendon Surgery
Injuries to Bone and Joint
Concomitant injuries to bone impact postoperative
management as the timing and type of exercises chosen
can be limited by the presence and characteristics of the
bony injury and the stability of fracture fixation. Primary
tendon repair is contraindicated in the presence of
complex unstable or intra-articular fractures.2 In the case
of stable, undisplaced fractures in the hand or digits, it
is possible to initiate an early motion regimen, provid­
ing supplementary splinting support for the fracture
while at rest and additional manual support when per­
forming exercises. Rigid fixation of fractures of the pha­
langes or metacarpals usually provides sufficient stability
to permit the early passive and active motion following
discussion with the surgeon regarding safety. Injuries to
more than one digit presenting with different patterns
of injury in each require individual exercise plans for
each digit. Care must be taken, however, initiating an
active motion protocol in complex injuries to the
middle, ring, and little fingers due to the quadriga effect
of the conjoined flexor digitorum profundus (FDP)
tendons. The most severely injured digit will dictate the
mobilization program and the therapist must devise
an exercise regimen that most effectively preserves the
tendon glide for each digit without placing excessive
stress on the repairs. This will be a combination of
Number and Type of Structures Damaged
Although single-digit injuries are common, the therapist
is often presented with more complex injuries. In the
presence of multidigit injuries, each digit may require
an individual exercise scheme according to the nature
of the injury. Controlled active motion regimens will be
unsuitable for severely injured digits and will therefore
restrict the use of this form of rehabilitation in adjacent
digits that are damaged to a lesser degree. After replanta­
tion surgery, there is a requirement to balance the needs
of both the flexor and extensor tendons. It is necessary
to adopt a regimen which safeguards the most vulner­
able structures.
Injuries to the Skin
The nature of the skin wounds affects the decision
making during rehabilitation. Untidy wounds, includ­
ing a crushing element, will impact the final outcome
and the therapist must prioritize the prevention of joint
and scar contracture. The therapist should differentiate
between incisions made during the surgery and those
that are part of the original injury. Carefully designed
surgical incisions made to access the tendon for explora­
tion and repair of the tendon and other structures will
minimize postoperative contracture. Random longitudi­
nal wounds which cross the metacarpophalangeal
(MCP) and the interphalangeal (IP) joints may cause
joint contracture.4 When the surgical incision is made
across these joints, active digital extension exercises and
prophylactic splinting should be incorporated into the
regimen to prevent loss of full joint extension.2-4
In more severe soft tissue injuries, the presence of
skin grafts or flaps necessary for wound closure delays
the start of postoperative mobilization until vascularity
is ensured. In digital-level injuries, care must be taken
not to impede joint motion or increase resistance to
tendon glide by the use of restrictive dressing. Dressings
should be shaped to reduce the resistance over the IP
joints and should cover the wound area only, avoiding
the use of circumferential adhesive tape (Figure 39-1).
Postoperative complications such as infection, heavily
exuding wounds or skin loss may require more substan­
tive dressings, which will result in delays in initiating
active motion. In cases where there is infection involv­
ing the tendon sheath, all motion should be delayed
until the infection is treated. Where the status of the
wound has the potential to impede active motion, for
example, in large open wounds with extensive edema,
mobilization must be confined to gentle passive exer­
cises to preserve tendon glide and minimize stress on
the repair until the restrictive swelling has reduced. In
the presence of heavily exuding wounds, dressings
should be removed on attendance for hand therapy, and
provided there is minimal edema, a controlled active
motion regimen may be initiated even in the presence
of skin loss.
Figure 39-1  Wound dressings must be kept to a minimum
and should not impede active motion following digital flexor
tendon repairs.

417
Chapter 39:  Customizing Flexor Rehabilitation Based on Zone or Type of Injury
Flexor Pollicis Longus (FPL) Injury
The postoperative care of the repaired FPL tendon
requires special consideration. The FPL tendon appears
to have a greater tendency to rupture when active motion
regimens are applied following a two-strand core suture
with simple peripheral stitches.8,9 Recent studies support
the use of active motion programs following a multi­
strand core, even without an epitendinous suture.10,11 If
there is any doubt as to the ability of the repair to with­
stand active motion the therapist should use passive
motion techniques. When using a controlled active
motion regimen, care must be taken to ensure that
movement is isolated to the IP joint (Figure 39-2). A
neutral position at the wrist benefits both flexion and
extension at the IP joint. Positioning the wrist in exten­
sion can impede active extension at the IP joint and
place stress on the repair site.
OTHER FACTORS
Length of Time From Injury to Repair
The therapist should have knowledge of the impact that
delayed repair may have on expectations and outcomes.
Delaying the surgery for 7 to 10 days or more may result
in formation of adhesions and collapse of the sheath
and pulleys and the tendon becomes difficult to retrieve
because of muscle–tendon unit shortening. In these
cases, the repair may require increased postoperative
passive only exercises for the more seriously injured
digits and careful active for those able to withstand
greater stress. In the presence of joint stiffness, restora­
tion of passive motion of the joint is imperative. Deep
lacerations to the digits may damage the palmar plate
or capsule of the IP joints, resulting in flexion contrac­
ture of the joints. Prophylactic extension splinting is
indicated to prevent the contracture from the early
stages of rehabilitation.
Injuries to Nerve
When nerves are repaired under tension, full range of
digital extension should be restricted in the early stages.
Major nerve injury at the wrist level is common, and
affects the details of the rehabilitation regimen. Con­
comitant median and ulnar nerve injuries at the wrist
or forearm levels, whether complete or partial, result in
some degree of intrinsic muscle paralysis and conse­
quent impairment of active digital motion. Under this
circumstance, special attention must be given to joint
positioning within the protective splint to optimize
tendon motion. Increased tendon excursion is facili­
tated if the MCP joints are positioned in 60° to 70°
flexion with the wrist in neutral. Loss of sensation and
cortical representation impacts the quality of motion
during the early rehabilitation phase and delays func­
tional recovery.
Vascularity of Injured Structures
Injuries requiring revascularization (e.g., digital replan­
tation) require careful management. Early mobilization
regimens should not be initiated until the viability of
the digit is established. Whether damage to the vincula
affects adhesion formation remains uncertain.5
Triggering of the Repaired Tendon
Against the Pulley
Bulky tendon repairs may trigger and are prone to dis­
ruption when gliding through annular pulleys.5,6
TYPE OF TENDON INJURY
Complete or Partial Injuries
Complete divisions of tendon require protected mobili­
zation and a customized program of exercises with no
functional activity of the hand for at least 6 weeks. Less
severe forms of the partial tendon injury (cut through
less than 40% of the cross section) can be managed
conservatively without surgery. The patients should be
permitted to move their injured fingers without restric­
tions.7 Surgically repaired partial lacerations of the
tendon (cut through greater than 40% of the cross
section) are usually capable of withstanding the stresses
of an active motion protocol and are less likely to rupture.
The hand requires some form of dorsal splint protection
but this may be removed as early as 3 to 4 weeks.
Figure 39-2  Following a strong multistrand repair of the
flexor pollicis longus tendon active motion should be
isolated to the IP joint to preserve tendon glide.

418
Section 5:  Rehabilitation of Tendon Surgery
lead to poor outcomes are long travel distances to
therapy and long waiting times. Postoperative pain,
especially that which is neurogenic in origin, should be
effectively controlled as this will significantly affect
compliance and behavior.20,21 Despite the best efforts,
the therapist is often presented with patients who do
not or are unable to adhere strictly to the requirements
of the rehabilitation regimen for a variety of reasons. In
these situations a compromise must be reached to
prevent repair rupture while permitting the patient to
continue with certain aspects of their daily life.
Age
Both the very young and some elderly patients are not
able to comply with treatment. A complex active motion
regimen is not suitable for elderly patients with dimin­
ished cognitive function or physical impairment. In
these cases immobilization is the option of postopera­
tive care.
The choice of immediate aftercare following tendon
repairs in children depends solely on the maturity of the
child. The principles of postoperative management are
similar to those applied to adults but the ability to
comply with splinting and exercises must be carefully
considered. Older children and teenagers will be able to
follow a specific regimen, but younger children may
inadvertently use the hand in functional activity and a
splint guard is thus required to deter use of the injured
hand (Figure 39-3). Although younger children may
cooperate with a thermoplastic splint and a guard, they
will require assistance with mobilizing exercises and the
responsibility for this then falls to the carer. The selec­
tion of the type of exercises then depends on the capa­
bility of the carer and their ability to comply. Babies
and toddlers are at risk of dehiscence through falls and
protection in a flexed position with serial extension as
appropriate. In the author’s own experience long delays
in repairing flexor tendons in zone 2 have been shown
to impact significantly on outcome leading to less favor­
able results than those that are repaired within the first
48 hours.
The Type and Quality of the Repair
There has been a significant amount of research
devoted to improving the tensile strength of flexor
tendon repairs, permitting greater confidence in the
use of active mobilization regimens and resulting in
reduced rates of dehiscence.12-14 Although some have
demonstrated the successful application of active mobi­
lization regimen with two-strand repairs of varying tech­
niques, others have reported unacceptably high rates of
rupture.15-17
Before selecting an appropriate postoperative regimen
the therapist should be acquainted with the caliber and
type of both core and circumferential suture in each
particular case. Four- or six-strand core suture repairs
with 4-0 or 3-0 suture produce a repair strong enough
to accommodate early active motion of the repaired
digital flexor tendon. If the tensile strength of the repair
is judged not to withstand the forces generated during
active digital motion, alternative and safer methods of
maintaining tendon glide should be considered.18,19
Venting the constrictive part of a major pulley (such as
the A2) will decrease the resistance to the tendon.
Repairing the tendon just distal to the A2 pulley, without
venting the constrictive part of the pulley, may subject
the tendon to greater resistance during early active
tendon movement. Particular care should be taken
when instructing the patient to move the tendons that
are repaired just distal to the A2 pulley, as greater inci­
dence of rupture of the primary repair is experienced in
this area.
Patient Characteristics and Compliance
The ability to comply with a rehabilitation regimen,
which involves commitment to a strict exercise program
within a protective splint for several weeks, vary among
individuals. Age, medical and mental health, and socio­
economic factors influence the choice of postoperative
management. Repairs are at risk of rupture if the patient
removes the splint for function and of poor gliding
action if they do not perform the required exercises.
Enhancing compliance requires a confident, knowledge­
able, and enthusiastic therapist who develops a good
working relationship with the patient. Good communi­
cation, the provision of high quality information, and
the enlistment of family support are also necessary. A
regimen that is versatile and tailored to suit the lifestyle,
socioeconomic circumstances, and work commitments
of the patient also promotes cooperation. Negative
factors that are most likely to impact on compliance and
Figure 39-3  The addition of a guard to the protective
splint permits unimpeded active flexion and extension and
deters functional use of the hand in a child following
reinsertion of the FDP tendon of the little finger.

419
Chapter 39:  Customizing Flexor Rehabilitation Based on Zone or Type of Injury
of zones 1 and 2, but there is very little to inform
evidence-based practice with regard of zones 3 to 5. The
therapist has to rely on a combination of the available
evidence for digital injuries, clinical reasoning skills,
and experience in dealing with zone 3 to 5 injuries.
Zone 1
The postoperative regimen is dictated by the type and
anatomical location of the surgical repair. Within zone
1, the FDP tendon lacerations are subdivided into three
areas (Box 39-1)23:
1. Under A4 pulley—there is an increased likelihood
of the formation of peritendinous adhesions if the
repaired tendon does not move smoothly through
the A4 pulley. Each 10° of the distal interphalan­
geal (DIP) joint motion produces 1 to 2 mm of
tendon glide up to a total of 7 mm in the unin­
jured finger.24 Following repair, edema and friction
to glide will reduce this excursion significantly and
therefore exercises should focus on achieving
active DIP joint motion where possible.25 Optimal
passive movement of the digit must be achieved
before careful active flexion is initiated at the DIP
joint (Figure 39-5).
Figure 39-4  An occlusive boxing glove–type dressing will
protect tendon repairs in very young children and babies.
functional activity and cannot follow mobilization regi­
mens. For this reason the hand should be immobilized
in a substantial, occlusive dressing for a period of 4
weeks. An effective way to protect the tendon repairs is
the application of an above elbow plaster of Paris or
resin cast. Nevertheless, both are prone to slippage and
may require removal under anesthesia in some children.
In our experience, a well-padded, occlusive, boxing
glove–type dressing using generous amounts of adhe­
sive tape to prevent removal is equally effective (Figure
39-4). This type of dressing can be easily reapplied in
an outpatient setting during the 4-week period of immo­
bilization, permitting inspection of the wound and
integrity of the repair along with passive exercises. The
paucity of literature on the subject leaves the therapist
with little guidance in this field. Much depends on the
ability of the child to cooperate with a regimen likely
to provide the most optimal outcome. Immobilization
for 4 weeks in the very young child has no detrimental
effects on the result.22
ZONE OF INJURY
Postoperative management should vary according to the
zone of injury. Therapists must possess an intimate
knowledge of the relevant anatomy and biomechanics
of each particular zone to customize the regimen accord­
ingly. Particular attention has been given in the litera­
ture to both surgery and rehabilitation in tendon repairs
*Subdivision of Moiemen and Elliot.
Close to the Insertion (Zone 1A)*
 Requires reinsertion to the distal phalanx
 Risk of muscle-tendon unit shortening
 Risk of DIP joint flexion deformity
 Prioritize passive digital flexion
 Initiate careful active flexion from the DIP joint
 Kleinert regimen not appropriate
 In presence of delicate repair, Evans protocol affords
greater safety28
Distal to the A4 Pulley (Zone 1B)*
 Risk of snagging on the A4 pulley
 May include injury to the palmar plate
 Increased risk DIP joint fixed flexion deformity
 Emphasize active digital extension exercises
 Apply prophylactic digital extension splints
 Initiate careful active motion at the DIP joint to
ensure tendon glide through the pulley
Under the A4 Pulley (Zone 1C)*
 Difficult to repair
 Increased risk of tendon adherence
 Prioritize passive digital flexion
 Encourage careful active motion at the DIP joint to
ensure glide through the pulley
 Careful blocking of the PIP joint by the therapist will
facilitate this if necessary
 No blocking of the PIP joint by the patient to facili­
tate motion until after 6 weeks
Box 39-1 Zone 1 FDP Tendon Lacerations

420
Section 5:  Rehabilitation of Tendon Surgery
Figure 39-5  Following repairs in zone 1 and zone 2 careful
active digital motion should be initiated from the DIP joint
to minimize stress on the repair site and promote differential
glide.
Active DIP joint flexion
2. Distal to A4 pulley—Lacerations at the level of the
DIP joint may include concomitant damage to the
palmar plate resulting in increased potential for
flexion contracture. Particular attention should be
given to active extension exercises and prophylac­
tic extension splinting. At this level there is a risk
that the tendon may snag on the distal edge of the
A4 pulley, placing excessive stress on the repair site
and resulting in loss of DIP joint flexion. Close
attention to the intraoperative details are required
in defining rehabilitation parameters to minimize
the risk of rupture while maximizing tendon
motion.
3. At the FDP tendon insertion—Tendon divisions at
close proximity to the FDP tendon insertion
require reinsertion of the tendon to the distal
phalanx. The consequences of surgery at this level
are some degree of tendon shortening and
increased risk of DIP joint flexion deformity. Par­
ticular attention should be paid to motion of the
DIP joint with emphasis on maintaining passive
flexion. Active flexion of the DIP joint is applied
only following a strong repair which has no
reported impediment to gliding. The Kleinert-
type regimen, even with modification, is unsuit­
able because this regimen can not sufficiently
move the DIP joint.26 In the presence of a repair
which requires greater protection, the Evans pro­
tocol, using a dorsal block splint to restrict exten­
sion and limiting the arc of active motion to
between 45° and 75°, affords greater safety and
permits tendon glide of up to 3 mm in the early
phase.27
In closed avulsion injuries at the insertion of the FDP
tendon, selection of an appropriate rehabilitation
regimen is decided by the type of the injury,30 as detailed
in Box 39-2. In all types of avulsion injury controlled
active motion is usually desirable.
Type 1—Tendon Retracted into Palm
 Muscle tendon unit may be shortened especially
after delay in treatment
 Loss of blood supply to the tendon
 Neutral position of wrist in protective splint
 Commence early action digital flexion initiated from
the DIP joint
 Prioritize active digital extension exercises
 Prioritize prevention of fixed flexion deformity at IP
joints
 Commence early passive finger extension stretching
without tension on the repair
 Remove protective splint and stretch IP joints into
extension with wrist and MCP joints in maximum
flexion
 Use digital extension splinting at night and between
exercise periods
Types II and III—Tendon Retracted to the PIP Joint
Level, With Small Bony Fragment (Type II) or
Remaining Distal to the A4 Pulley With Large Bony
Fragment (Type III)
 Less shortening of the muscle–tendon unit
 Delayed reinsertion has potential for favorable
outcome
 Tendon retains good blood supply
 Wrist position in protective splint 10° to 30°
extension
 Commence early action digital flexion initiated from
the DIP joint
 Prioritize active digital extension exercises
 Use prophylactic digital extension splinting at night
to prevent fixed flexion deformity
Rehabilitation Regimen for All 3 Types of Closed
FDP Tendon Avulsion Injuries
 Protective dorsal splint for 6 weeks
 Wrist position dependant on type of injuries
 MCP joints 30° flexion
 Restore optimal passive digital flexion
 Initiate early controlled active motion regimen
emphasizing DIP joint motion
 Hourly exercises; 10 repetitions of digital flexion
 Encourage active digital extension exercises
 Apply prophylactic digital extension splints
 Watch for complications arising from button reinser­
tion technique—Infection, edema catching/breaking
of exposed suture, and pain if this technique is used
 Excessive pain will impede compliance and motion
 Commence composite extension stretching at 6
weeks to treat muscle–tendon unit shortening
 Consider serial plaster of Paris casts at 6 weeks to
treat residual IP joint flexion deformities (see Figure
39-6)
 Return to light function at 6 weeks and contact
sports at 12 weeks
Box 39-2
Closed FDP Avulsion Injuries

421
Chapter 39:  Customizing Flexor Rehabilitation Based on Zone or Type of Injury
the tendon suture through the nail and tying over a
button. Infection, edema, catching, and breaking of the
exposed suture and pain impeding motion are some
common problems. The source of pain should be inves­
tigated as persistent discomfort will deter compliance
with exercise regimen. The therapist should also be
careful to instruct the patient that the suture should
remain in situ for 6 weeks as early inadvertent removal
may result in separation of the reinserted tendon espe­
cially in the presence of an early motion regimen.
Zone 2
Rupture of the repairs, cross adherence, triggering, or
snagging of the repair on the A2 pulley and finger joint
contracture are common complications in this area. The
essential aim of rehabilitation is to restore sufficient
tendon gliding but avoid rupture of the repair during
tendon motion. Another objective is the preservation of
differential glide of the FDS and FDP tendons if both
tendons are repaired. The exercise program should be
designed to promote differential tendon movement
(Box 39-3). Once passive flexion is regained, active
digital flexion exercises will promote optimal glide and
prevent cross adherence (Figure 39-7).
If the surgical repair is judged strong enough, an
active motion program can be applied. The therapist
should have a clear knowledge of the resistance the
repaired tendon may encounter during early active
finger motion. The presence of digital edema and posi­
tion of the MCP joint are important considerations. In
an edematous finger, where there is resistance to active
motion and tendon glide, placing the MCP joints in
excessive flexion within the dorsal splint will bias active
motion to the PIP joints and can impede the initiation
of flexion from the DIP joint. Confining motion to the
PIP joint only will encourage cross adherence, prevent­
ing differential glide. Therefore the patient should be
encouraged to attempt active DIP joint flexion at the
start of active finger flexion as in the early stages post
repair they may be unable to perform this action in the
inner range of motion30,31 Tang has demonstrated that
the strength of the repair decreases as the angle of
tension increases and patients should therefore be dis­
couraged from both performing composite flexion and
using maximal effort to flex the DIP joints at the end of
the range of flexion in the early postoperative stages.32
If optimal DIP joint range of motion is to be encour­
aged, active digital flexion should be initiated at this
joint. This can be facilitated in most patients by posi­
tioning the wrist joint in 10° to 30°extension and the
MCP joints in 30° of flexion.1,35
Flexion contractures of the IP joints are a common
complication and can occur even after simple clean
tendon injuries. Use of rubber band traction regimens
has been criticized for encouraging the development of
flexion contracture of the repaired fingers; thus, this
In type I injuries, a careful passive stretching element
should be incorporated into the regimen taking care not
to place any stress on the repair. To facilitate this, the
hand should be removed from the protective splint and
all tension removed from the flexor tendon by position­
ing the wrist and MCP joints in maximum flexion while
carefully passively extending each IP joint individually.
Provision of a palmar splint should be considered for
use at night and between exercise periods and this may
be serially extended as rehabilitation progresses. At the
end of the protective phase residual flexion deformities
or shortening may be addressed by progressive stretch­
ing of the joints and composite stretching of the muscle
tendon unit with all joints in full extension. Despite
concerted effort flexion deformities may be a persistent
problem and serial plaster casting should be considered
as an early option (Figure 39-6).29
Type II and III injuries are less problematic as there
is little or no retraction and shortening of the muscle–
tendon unit. The surgical exposure required for reinser­
tion necessitates a lesser incision and the tendon
maintains good condition via an intact blood supply.
Delayed reinsertion often produces favorable outcomes;
however, there is still a tendency to develop flexion
deformity especially at the DIP joint and the therapist
must work within the rehabilitation regimen to prevent
or correct this. The technique of insertion should not
affect the choice of regimen but the therapist should be
aware of potential complications in the case of stitching
Figure 39-6  Serial plaster of Paris casts may be applied to
treat persistent fixed flexion deformities of the IP joints in
the later stages of rehabilitation.

422
Section 5:  Rehabilitation of Tendon Surgery
method of rehabilitation has become less popular and
some surgeons advocate that they should be abandoned
altogether.33 Effective means of maintaining digital
extension should be incorporated from the early stages.
There are various reported methods from thermoplastic
palmar splints to strapping the digits into extension at
night and between exercise periods.34-36 Flexion contrac­
ture at the IP joints prolongs the rehabilitation, and
more importantly, these contractures tend to be persis­
tent, and are a major cause of poor outcomes in zone 2
tendon repair. They should be treated after 6 weeks by
stretching and palmar based extension splinting at night
(Figure 39-8).
Zone 3
There is little published information relating to zone 3
rehabilitation and therapists have been obliged to adapt
those rehabilitation regimens designed for zone 2 inju­
ries. Anatomical differences and injury characteristics,
however, necessitate modification to splinting and exer­
cise protocols. Lacerations in the mid-palm may include
damage or denervation of the lumbrical and interosse­
ous muscles with resultant loss of function and scarring.
Tendons are likely to adhere to surrounding structures
and the patient may experience difficulty in fully extend­
ing the IP joints and may even develop flexion deformi­
ties. Manual positioning of the MCP joints in greater
*Subdivision by Tang.
Rehabilitation Regimens: Common to All Zones 2 to 5
 Protective dorsal splint for 6 weeks: joint positions may be changed within this period
 Prioritize restoration of passive digital flexion in all patients
 Initiate early controlled active motion regimen in cooperative patients if surgical repairs are robust and tendon gliding
resistance is judged not particularly high
 Promote differential glide if multiple tendons are repaired
 Encourage active digital extension exercises—vital to prevent extension deficits of joints
 Apply digital extension splints if there is early loss of active extension or significant joint injury
 Incorporate early active wrist motion as appropriate
 At 6 weeks commence stretching of residual flexion deformity in conjunction with serial splinting/casting
 Commence light functional activity at 6 weeks
 Return to normal activities at 10 to 12 weeks
Special Considerations
Zone 2
 Set wrist position at 10° to 30° extension and MCP joint at 30° flexion
 Commence early active motion regimen at 4 to 5 days if appropriate
 Restore optimal passive digital flexion prior to active motion. Passive digital flexion stretches repeated until free motion
achieved especially in an edematous finger
 Encourage hourly exercises—10 repetitions of active digital flexion. No forced active flexion. Slowly increase range of
flexion over the first 3 weeks
 Prevent cross adherence of FDP to FDS by encouraging exercises, which promote differential glide (see Figure 39-7)
 Do not initiate place and hold maneuver until full active tendon glide is ensured
 Permit early active wrist motion in trust worthy patients
 Ruptures of the repairs are most frequent when the repair sites are distal or under the A2 pulley (zones 2B and 2C).*
Majority of ruptures are in the first 1 or 2 weeks post surgery. Therapist and patient must take care to avoid repair
rupture during motion
Zone 3
 Wrist position 10° to 30° extension
 MCP joints position 30° flexion in absence of nerve injury and intrinsic damage
 MCP joints position 60° to 70° flexion with intrinsic muscle injury or denervation
 Beware risk of fixed flexion deformity at MCP joints due to palmar scar
Zones 4 and 5
 With major nerve injury: wrist position neutral for 2 weeks
 Position MCP joints at 60° to 70° flexion to prevent claw finger deformity and maximize tendon excursion
 Delay wrist motion for 2 weeks
 At 6 weeks begin composite extension by stretching and serial splinting
 Beware risk of and prevent development of intrinsic muscle tightness. Avoid prolonged splinting; use prophylactic
passive stretching techniques. Incorporate early sensory re-education to preserve cortical representation
Box 39-3
Author’s Rehabilitation Regimens for Flexor Tendon Injuries Zones 2 to 5

423
Chapter 39:  Customizing Flexor Rehabilitation Based on Zone or Type of Injury
joints should be positioned in 60° to 70° of flexion.
Care must be taken to ensure that an optimal range of
digital flexion can be achieved with the MCP joints in
this position to preserve tendon glide. Careful active
wrist motion with the digits in a relaxed semi flexed
position is permitted out of the splint in trustworthy
patients. Frequent monitoring of patients is required if
there has been intrinsic muscle damage or denervation.
This potent combination of flexor tendon and intrinsic
muscle injury can lead to claw deformity, which can
rapidly become fixed.
Flexion deformities of the MCP joints may also arise
as a result of palmar scar contracture but these are
usually easy to treat with serial extension splinting once
the dorsal protective splint has been removed (see
Figure 39-8).
Zone 4
Although tendon lacerations in this region are not
common, concomitant damage to the median nerve
and, less commonly, the ulnar nerve in Guyon’s canal
will result in intrinsic muscle paralysis and loss of sen­
sation. At this level, adhesions to the synovial sheath
and between the tendons in the restricted space under
the retinaculum very quickly become established. The
postoperative exercise regimen should promote the
maximum available tendon glide. An active motion
regimen is desirable, but the required amount of tendon
flexion may be necessary during exercise periods to
facilitate IP joint extension and prevent the develop­
ment of joint contractures. Edema of the digits is rarely
a problem and free passive and active motion of both
the DIP and PIP joints should be encouraged. The wrist
should be positioned in 10° to 30° extension. If there
have been significant intrinsic muscle injuries, the MCP
Figure 39-7  To promote differential glide and prevent the development of fixed flexion deformity the components of an
active motion regimen should include four separate exercises. A, First, gentle active motion initiated at the DIP joint in a hook
type action. B, Second, active flexion at the PIP joints with the FDP tendons at rest. C, Third, once edema has subsided careful
composite flexion. D, In each exercise period, full active IP joint extension should be encouraged within the confines of the
splint.
A
B
C
D
Figure 39-8  Residual fixed flexion deformities and loss of
composite extension should be treated with palmar based
serial extension splinting 6 weeks following repairs.
Palmar extension splinting
Zone 2
Tendon repair
Zone 3
Tendon repair
Palmar extension splinting

424
Section 5:  Rehabilitation of Tendon Surgery
flexion should be emphasized to prevent the develop­
ment of IP joint contracture, especially in the presence
of intrinsic muscle paralysis.
Active protocols are indicated as repairs are unlikely
to rupture. Exercises that focus on differential glide
should be incorporated to avoid adhesions between the
tendons, although this may not affect the outcomes.37
Another commonly seen consequence is adhesions
between the skin and the tendons. These do not usually
restrict tendon glide but can be of cosmetic concern to
the patient.
The protective splint should be removed at 6 weeks
after surgery. At this stage common complications are
loss of composite extension and fixed flexion deformi­
ties of the joints. Forearm based splints should be pro­
vided to achieve gradual restoration of extension. In the
presence of nerve injuries, anticlaw devices may be
required throughout the later stages of rehabilitation.
Sensory re-education techniques should also be
incorporated into the rehabilitation regimen from the
start of treatment. These are designed to assist in the
preservation of cortical representation of the hand by
modifying the functional reorganization, which will
inevitably occur within the somatosensory cortex.38
SELECTING A REGIMEN FOLLOWING
RE-REPAIR OF RUPTURES OF THE
SURGICALLY REPAIRED FLEXOR TENDON
At any stage during rehabilitation the patient may
report a loss of motion. This indicates rupture of the
surgically repaired tendon. The occurrence may vary
from an obvious sudden event during exercise or
unprotected use of the hand to a slower decrease in the
amount of movement as a gapping repair slowly gives
way altogether. The exact cause or time of rupture can
be unclear and may even be masked in cases of infec­
tion or significant edema. The rupture of the primarily
repaired flexor tendons in the hand can occur in a
period from several days to 6 weeks after surgery.15 The
repaired tendon is at most risk of rupture during the
first 2 weeks after repair with the second week being
the most dangerous. Diagnosis of a ruptured tendon is
made by examination of posture or manual tendon
testing, but recent advances in the technique of diag­
nostic ultrasound have made this process considerably
more efficient. If a re-repair is to be undertaken, it is
vital that the surgeon, therapist, and patient are involved
in any decision making. These decisions must be based
on the reason for the rupture, if known, and the likeli­
hood of a favorable outcome of the re-repair. In 1982
Leddy suggested that the best outcomes are likely if a
re-repair is undertaken soon after rupture and most
subsequent studies appear to support this although
careful consideration should be given to the little
finger.39-41 Some patients choose not to opt for further
surgery because of inability to commit to an extended
motion can be hard to achieve for a number of reasons.
For example, wrist extension may increase tension on a
median nerve repair. A neutral position or slight flexion
of the wrist is generally the best compromise. After 2
weeks, the wrist can be gradually extended by altering
the splint to promote greater tendon glide.
Loss of intrinsic muscle action and muscle–tendon
unit shortening can result in a change in the resting
position of the digits. To facilitate tendon motion, the
MCP joints require positioning in a greater degree of
flexion than in digital injuries. Prior to application of
the protective splint active digital flexion should be
observed in varying positions of the MCP joints and the
position from which the greatest tendon excursion can
be achieved selected. In patients with severe clawing of
the digits, maintaining the MCP joint in 60° to 70°
flexion is recommended, to achieve active IP joint exten­
sion and a greater range of IP joint flexion. Active digital
flexion and differential glide exercises should be encour­
aged from the start of the rehabilitation. Active wrist
extension exercises in conjunction with active digital
flexion taking advantage of the tenodesis effect should
be initiated around 3 weeks, out of the splint in trust­
worthy patients.
Zone 5
Injuries at the level of the wrist and forearm may range
from an isolated tendon to a complex “spaghetti wrist”
involving neurovascular structures. In cases of injury to
isolated wrist tendons, the dorsal protective splint
should extend to the MCP joints only and the hand is
mobilized freely. The wrist should be positioned in
neutral or slight flexion depending on the tightness of
the repair and active wrist motion may commence at 3
weeks postoperatively. Patients can be permitted to
perform light activity but should be deterred from
strong gripping. When the injury involves multiple
tendon and nerve injuries, the splint position should
be dictated by (1) protection of the repaired structures,
(2) patient comfort, and (3) facilitation of active motion
regimens. In the presence of nerve injuries the wrist
should not be positioned in extension beyond neutral;
minimal wrist flexion is more appropriate in the early
postoperative period. The position of the wrist can be
gradually altered to achieve greater extension during the
first few weeks of the protective phase usually to a
maximum of 10° to 20°. Within the protective splint,
the MCP joints should be positioned to facilitate digital
flexion, inhibit the development of a claw deformity
and allow maximal IP joint extension. As in zones 3 and
4, 60° to 70° of MCP joint flexion is required depen­
dent on the degree of clawing and the presence of a
nerve injury. Digital edema is not usually present in the
early stages in injuries at this level and consequently
there is no restriction to passive IP joint motion. Passive
extension exercises with the MCP joints in maximal

425
Chapter 39:  Customizing Flexor Rehabilitation Based on Zone or Type of Injury
active motion or combined early passive-active motion
with resultant improvement in outcomes.1,2,8-11,17
SUMMARY
A skilled surgeon, experienced therapist, and compliant
patient are desirable requirements but not a guarantee
of an excellent result. Active motion regimens are best
in all zones of injury as they provide the optimal require­
ments for motion and tension following flexor tendon
repair. There are no standard protocols, however, and
therapists managing these injuries must assimilate a
considerable amount of information before selecting
the most appropriate treatment for each individual
patient. The pursuit of optimal tendon glide must not
take precedence over safety; however, there are a variety
of alternative options available with the potential to
achieve favorable outcomes.
Acknowledgments
I would like to thank Professor Gus McGrouther and
Mr. Stewart Watson, consultant hand surgeons at
Wythenshawe Hospital, Burns and Plastic Surgery Unit,
who read through my manuscript. The rehabilitation
regimens described in this chapter are the result of expe­
rience and close collaboration between the surgeons
and the therapists in South Manchester. I would also
like to thank my specialist hand therapy colleagues
Sarah Turner, Alison Roe, and Emma Kelly for their
contributions.
period of rehabilitation. The therapist should be aware
of the intraoperative findings of the re-repair before
embarking on rehabilitation, as the status of the
re-repair should differ from that of the original. An
understanding of the mechanics of the rupture is neces­
sary whether due to technical failure or excessive force
placed on the repair. Tendon ends may have softened
necessitating careful handling and making re-repair dif­
ficult to perform. The tendon may also be shortened
due to excision of the previous repair site requiring
careful postoperative positioning to ease the tension
across the repair site. Following a re-repair the therapist
should consider modifying the protocol and exercising
caution in the use of active motion until safety is
ensured. Generally, a lesser aggressive form of exercise
is prescribed for the repair of the ruptured primary
tendon repair.
Over the past 20 years, the materials and methods
used in the postoperative management of flexor tendon
injuries have changed considerably. Although plaster of
Paris was originally used for the entire 6-week period of
protection, it is now used only in the immediate post­
operative period. It has been superseded by thermoplas­
tic material, which provides greater comfort and
longevity. In the late 1970s, when mobilization of flexor
tendon repairs became popular, the passive flexion regi­
mens of Duran and Houser and Kleinert predomi­
nated.18,19 Although these passive regimens still have
their place, there has been a general transition to early
References
1. Elliot D: Primary flexor tendon repair-operative repair, pulley
management and rehabilitation: invited personal review,
J Hand Surg (Br) 27:507–513, 2002.
2. Tang JB: Indications, methods, postoperative motion and
outcome evaluation of primary flexor tendon repairs in Zone
2, J Hand Surg (Eur) 32:118–129, 2007.
3. Pettengill KM: The evolution of early mobilization of the
repaired flexor tendon, J Hand Ther 18:157–168, 2005.
4. Bruner JM: The zig-zag volar-digital incision for flexor-tendon
surgery, Plast Reconstr Surg 40:571–574, 1967.
5. Lindsay WK, Thomson HG: Digital flexor tendons: an experi­
mental study. Part I: The significance of each component of
the flexor mechanism in tendon healing, Br J Plast Surg
12:289–316, 1960.
6. Tang JB, Wang YH, Gu YT, et al: Effect of pulley integrity on
excursions and work of flexion in healing flexor tendons,
J Hand Surg (Am) 26:347–353, 2001.
7. Al-Qattan MM: Conservative management of zone II partial
flexor tendon lacerations greater than half the width of the
tendon, J Hand Surg (Am) 25:1118–1121, 2000.
8. Sirotakova M, Elliot D: Early active mobilization of primary
repairs of the flexor pollicis longus tendon, J Hand Surg (Br)
24:647–653, 1999.
9. Sirotakova M, Elliot D: Early active mobilization of primary
repairs of the flexor pollicis longus tendon with two Kessler
two strand core sutures and a strengthened circumferential
suture, J Hand Surg (Br) 29:531–535, 2004.
10. Giesen T, Sirotakova M, Copsey AJ, et al: Flexor pollicis
longus primary repair: further experience with the Tang
technique and controlled active mobilization, J Hand Surg
(Eur) 34:758–761, 2009.
11. Elliot D, Southgate CM: New concepts in managing the long
tendons of the thumb after primary repair, J Hand Ther
18:141–156, 2005.
12. Cao Y, Tang JB: Biomechanical evaluation of a four-strand
modification of the Tang method of tendon repair, J Hand
Surg (Br) 30:374–378, 2005.
13. Dona E, Gianoutsos MP, Walsh WR: Optimizing biomechani­
cal performance of the 4-strand cruciate flexor tendon repair,
J Hand Surg (Am) 29:571–580, 2004.
14. Birks M, Peck F, Lees V, et al: The influence of multistrand
repairs on the rupture rates of flexor tendon injuries in zone
II, BBSH Autumn Meeting 2008.
15. Elliot D, Moiemen NS, Flemming AF, et al: The rupture
rate of acute flexor tendon repairs mobilised by the con­
trolled active motion regimen, J Hand Surg (Br) 19:607–612,
1994.
16. Peck FH, Bücher CA, Watson JS, et al: A comparative study of
two methods of controlled mobilization of flexor tendon
repairs in zone 2, J Hand Surg (Br) 23:41–45, 1998.
17. Silfverskiöld KL, May EJ: Flexor tendon repair in zone II with
a new suture technique and an early mobilization program
combining passive and active flexion, J Hand Surg (Am)
19:53–60, 1994.
18. Duran RJ, Houser RG: Controlled passive motion following
flexor tendon repair in zones 2 and 3. In AAOS Symposium
on Tendon Surgery in the Hand, St Louis, 1975, Mosby,
pp 105–114.

426
Section 5:  Rehabilitation of Tendon Surgery
31. Cao Y, Tang JB: Resistance to motion of flexor tendons and
digital edema: An in vivo study in a chicken model, J Hand
Surg (Am) 31:1645–1651, 2006.
32. Tang JB, Xu Y, Wang B: Repair strength of tendons of varying
gliding curvature: A study in a curvilinear model, J Hand Surg
(Am) 28:243–249, 2003.
33. Tang JB: Clinical outcomes associated with flexor tendon
repair, Hand Clin 21:199–210, 2005.
34. Savage R: The influence of wrist position on the minimum
force required for active movement of the interphalangeal
joints, J Hand Surg (Br) 13:262–268 1988.
35. May EJ, Silfverskiöld KL, Sollerman CJ: Controlled mobili­
zation after flexor tendon repair in zone II: a prospective
comparison of three methods, J Hand Surg (Am) 17:942–952,
1992.
36. Peck FH, Bucher CA, Watson SJ, et al: An audit of flexor
tendon injuries in zone II and its influence on management,
J Hand Ther 9:306–308, 1996.
37. Yii NW, Urban M, Elliot MD: A prospective study of flexor
tendon repair in zone 5, J Hand Surg (Br) 23:642–648, 1998.
38. Rosén B, Balkenius C, Lundborg G: Sensory re-education
today and tomorrow: A review of evolving concepts, Hand
Ther 8:48–56, 2003.
39. Allen BN, Frykman GK, Unsell RS, et al: Ruptured flexor
tendon tenorraphies in zone II: Repair and rehabilitation,
J Hand Surg (Am) 12:18–21, 1987.
40. Leddy JP: Flexor tendons–acute injuries. In Green DP,
editor: Operative Hand Surgery, New York, 1982, Churchill
Livingstone, pp 1347–1350.
41. Dowd MB, Figus A, Harris SB, et al: The results of immediate
re-repair of zone 1 and 2 primary flexor tendon repairs which
rupture, J Hand Surg (Br) 31:507–513, 2006.
19. Lister GD, Kleinert HE, Kutz JE, et al: Primary flexor tendon
repair followed by immediate controlled mobilization,
J Hand Surg (Am) 2:441–451, 1977.
20. Dobbe JG, van Trommel NE, Ritt MJ: Patient compliance with
a rehabilitation program after flexor tendon repair in zone II
of the hand, J Hand Ther 15:16–21, 2002.
21. Sandford F, Barlow N, Lewis J: A study to examine patient
adherence to wearing 24-hour forearm thermoplastic splints
after tendon repairs, J Hand Ther 21:44–53, 2008.
22. Elhassan B, Moran SL, Bravo C, et al: Factors that influence
the outcome of zone I and zone II flexor tendon repairs in
children, J Hand Surg (Am) 31:1661–1666, 2006.
23. Moiemen NS, Elliot D: Primary flexor tendon repair in zone
1, J Hand Surg (Br) 25:78–84, 2000.
24. McGrouther DA, Ahmed MR: Flexor tendon excursions in “no
man’s land,” Hand 13:129–141, 1981.
25. Silverskiöld KL, May EJ, Törnvall AH: Flexor digitorum pro­
fundus tendon excursions during controlled motion after
flexor tendon repair in zone II: a prospective clinical study,
J Hand Surg (Am) 17:122–131, 1992.
26. Slattery PG, McGrouther DA: A modified Kleinert controlled
mobilization splint following flexor tendon repair, J Hand
Surg (Br) 9: 217–218, 1984.
27. Evans RB: Zone I flexor tendon rehabilitation with limited
extension and active flexion, J Hand Ther 18:128–140, 2005.
28. Leddy JP, Packer JW: Avulsion of the profundus tendon inser­
tion in athletes, J Hand Surg (Am) 2:66–69, 1977.
29. Colditz JC: Plaster of Paris: the forgotten hand splinting
material, J Hand Ther 15:144–157, 2002.
30. Cao Y, Tang JB: Investigation of resistance of digital subcuta­
neous edema to gliding of the flexor tendon: An in vitro study,
J Hand Surg (Am) 30:1248–1254, 2005.

427
CHAPTER
40 
STATE OF THE ART OF
EXTENSOR TENDON
REHABILITATION
Ton A.R. Schreuders, PT, PhD, and
Gwendolyn Van Strien, lPT, MSc
smaller excursion than the flexors and the forces gener-
ated by extensor muscles are weaker. The extensor mech-
anism (e.g., the structures of the extensor tendons on
the dorsum of the digits) consists of multiple tendon
slips intertwined and intimately covering the dorsum of
the phalanges.1
The extensor tendons do not have a synovial sheath
system, but at the wrist level (zone 7) (Figure 40-1),
the extensors are restricted by the extensor retinaculum
that forms six fibro-osseous compartments within
which 12 extensor tendons pass, which also help to
prevent bowstringing. Adhesion formation after exten-
sor tendon injuries is not uncommon, but because the
requirement of tendon excursion is low and adhesions
form under largely moveable skin, adhesions often do
not pose an important problem for function of the
extensor tendons. The extensor retinaculum at the wrist
functions as a pulley, keeping the wrist and finger exten-
sor tendons close to the axis of the carpus during
motion. At this location, dense adhesions may occur
between extensor retinaculum and the tendons, that
hinder tendon movement.
The architecture of the extensor tendon is complex,
especially at the proximal phalanx and middle phalanx
level (zones 3, 4, and 5). Here the extensor tendon gives
off the central slip and lateral bands and connecting
fibers, forming a broad expansion over the dorsum and
lateral aspects of the proximal phalanges with complex
intertwined tendon fibers.2 The fibers of this extensor
apparatus connect to the tendons of both the interosse-
ous and lumbrical muscles. The central slip is the direct
distal extension of the extensor digitorum communis
(EDC) tendon. Extensor tendons on the dorsal side of
the hand (zone 6) are separated tendons, but they have
connections through the juncturae tendinum.3 The
juncturae tendinum coordinate the opening of the hand
via the extension of thumb, digits and wrist, as well as
assist in force redistribution through the extensors.4
Because of the interconnections and multiple inser-
tions, the extensor tendons are less likely to retract when
OUTLINE
The extensor tendons have complex anatomical struc-
tures and mechanically they are dependent upon each
other. A large variety of protocols are available to
accommodate therapy for the varied extent and regions
of injuries. Generally, the protocols are classified into
three types: immobilization, early passive controlled
motion by means of dynamic splinting, and early
active motion. After digital extensor tendon laceration
and repair protective positioning of the finger is
important to release tension across the repair and to
prevent contracture of the joints of the finger. Main-
taining the joint(s) in an extended position will ease
tension on the tendon repair, and early motion can
prevent the joints from developing stiffness, a balance
that should be carefully planned. In zones 5 and 6,
most therapists and surgeons advocate early motion
(including dynamic splinting and active motion pro-
tocols). Though these early motion protocols produce
earlier recovery of hand motion, the late outcomes are
similar to immobilization protocols. Adhesion forma-
tion is less of a problem compared to the flexor tendon;
however, but in zone 7, where the extensor retinacu-
lum covers the extensor, occurrence of adhesions can
drastically hamper tendon gliding. After injuries in
this area, early movement of the tendon is critical to
ensure recovery of function. After tendon repair in
zone 8, early movement of the tendon may not be a
necessity. Early motion is usually implemented from
3 to 5 days postsurgery and therapy may need to con-
tinue for 2 to 3 months. In the initial days, edema
formation should be properly dealt with; information
about the severity of injuries, patient compliance, and
strength of surgical repairs should be taken into serious
consideration to decide rigor of the early motion.
The extensor tendons lie directly under the skin with
very thin subcutaneous tissue above. In most parts, the
tendons are thin and flat. The extensor tendons have

428
Section 5:  Rehabilitation of Tendon Surgery
should be taken not to create gapping or rupturing of
the suture site if early motion is initiated. The extensor
tendon can be moved proximally through joint exten-
sion or distally through joint flexion. Proximal move-
ment of the tendon requires active contraction of the
extensor muscles (which are generally weaker compared
to the flexors). Active or passive flexion of the joints
produces distal glide of the extensor tendons. Similar to
the rehabilitation for the flexors, extensor tendon
therapy focuses on tendon gliding with a gradual
increase of load to the repaired tendon.
The development of protocols in flexor tendons was
spurred on by new suture techniques. Suture techniques
for extensors have not seen similar advances. Neverthe-
less, new extensor protocols have mostly followed the
same trends. Similarly, there are three types of protocols
for the extensors: (1) immobilization, (2) passive mobi-
lization with a dynamic extension splint, and (3) early
active mobilization.
The Initiation of Therapy: Timing
No studies have discussed specifically the optimal timing
to initiate motion of the extensor tendons. Following
flexor tendon repair, investigators have proposed initiat-
ing active flexion at postsurgical day 3 to 5, based on
experimental evidence.7 Though it is unsure whether we
can extrapolate the conclusions from the flexor tendon
research to the extensors, we generally adopt these con-
clusions, because of a paucity of investigation on exten-
sors in this regard.
Ideally, postoperative mobilization should be initi-
ated when the therapist has a wide margin of safety
between the tendon gapping strength and the load
required to initiate tendon gliding.5 Active tendon
gliding has to overcome resistance caused by (1) friction
between tendons and their surrounding structures;
(2) resistance resulting from joint stiffness, edema, and
adherence to surrounding tissues; and (3) resistance of
antagonist muscles.5 In zone 7, the tendon encounters
additional resistance from the restrictive extensor reti-
naculum, that covers the tendons tightly.
When the patient is immobilized for the first 3 or 5
days after surgery, extra care must be taken to ensure
correct positioning of the hand, which should be
checked at the first or second day with assessment of
pain, excessive swelling, or bleeding. An earlier dressing
change may be necessary.
Patient Referral Information
Detailed and specific information about the patient, the
injury, and operative procedures performed are helpful
in deciding protocols for individual patients. A copy
of the operative report is invaluable in this regard.
The therapist needs to know the mechanism of injury,
level and severity of the tendon injuries, and structures
repaired. It is also important to know the type of tendon
injured than are the flexor tendons. Loss of length in the
extensor tendon in the digits disturbs function of the
extensors, as structures in each segment are extremely
different. Shortening of a phalanx greatly impairs the
working capacity of the extensors.
GENERAL GUIDELINES FOR CARE AFTER
EXTENSOR REPAIRS
There have been fewer clinical investigations devoted to
surgery and rehabilitation of the extensor tendons than
the flexor tendons. Conceptually, many principles and
methods used in the postoperative care were derived or
modified after those for flexor tendons. Essentially, ade-
quate edema control and protection of the tendon
repair from excessive stress by eliminating inadvertent
hand use are important after surgery. Protective splint-
ing and instructions for home care should be given from
the first postoperative days.
As with care after flexor tendon repairs, therapy for
extensor tendons should start early after surgery, but care
Figure 40-1  Zones of extensor tendon injuries. (Copyright
of Judy Colditz.)
I
II
T-I
T-II
T-III
T-IV
T-V
VII
VIII
VI
V
III
IV

429
Chapter 40:  State of the Art of Extensor Tendon Rehabilitation
returning to their daily activities directly after splint
removal may risk a tendon re-rupture or develop a DIP
joint extension lag. The entire course of therapy can be
divided into the following phases:
First Phase (6 to 8 Weeks): Therapy During
Splinting Phase
After a mallet finger injury, a splint is applied to hold
the distal interphalangeal (DIP) joint in extension.13
It is preferable to apply a custom made splint to posi-
tion the DIP joint in submaximal extension, that is,
maximum extension that does not produce skin blanch-
ing. However, circulation is more often compromised
by too much pressure of the splint rather than too much
DIP joint hyperextension.
The splint should not include the proximal interpha-
langeal (PIP) joint and should allow full PIP joint
flexion. Nonelastic adhesive tape should securely fix the
splint to the volar skin of the middle part of the finger.
In the supple and hypermobile fingers, a mallet finger
injury encourages development of a swan neck defor-
mity. This can be prevented by adding a 30° extension
block for the PIP joint as part of the splint (Figure 40-2).
When the patient is less compliant or needs more
secure protection, a plaster of Paris or Quickcast cylindri-
cal cast is a good choice to immobilize the DIP joint.
Therapists often have difficulty keeping the DIP joint
extended while fabricating a plaster cast. This difficulty
may be overcome if an adhesive nonelastic tape is first
applied starting on the volar aspect of the distal end of
the finger and ending on the dorsum, crossing the DIP
joint crease and extending to the middle phalanx. This
tape holds the DIP joint in extension and the plaster of
Paris is easily applied over it. The tape also prevents the
plaster from slipping off as the plaster adheres to the tape.
With the splint/cast in place, the patient is instructed
to perform active PIP flexion exercises with the adjacent
fingers held in extension. With this exercise, the flexor
digitorum profundus (FDP) cannot be activated and the
flexor digitorum superficialis (FDS) will flex the PIP
joint only, eliminating flexion forces at the DIP joint.
repair and the strength of the repair. The surgeon should
inform the therapist if the tendon is frayed, if there is
the possibility of a postoperative infection, or if tissues
were seriously damaged. All these factors greatly influ-
ence the extent and amount of scar and the probability
of tendon adherence or repair rupture. Associated inju-
ries of bones, nerves, blood vessels, and skin should also
be taken into account in deciding protocols that best
suit the patient. Information on the use of skin grafts
and the location of repaired blood vessels is of great
importance since pressure over these areas should be
avoided by careful design of a splint. Finally, therapists
should obtain information regarding pertinent medical
conditions that may influence wound healing or the
normal progression of tendon healing.7
Record Keeping
Active and passive range of motion of the hand should
be regularly measured and recorded. Improvement of
range of motion is an indication for the effectiveness of
therapy and helps to decide if a protocol needs modifi-
cations. Function is assessed by measuring passive and
active range of motion of the joints. Extensor pollicis
longus (EPL) function, after a transfer of the extensor
indices to the EPL tendon, can be measured by the dis-
tance of thumb elevation from the surface of a table on
which the hand is placed flat.8 In addition, the Kapandji
thumb rule by giving a score (0-10 score; 0: to the meta-
carpophalangeal joint level of the index finger, and 10:
to the distal palmar crease of the little finger) to how far
the thumb can reach different parts of the fingers can
be used to score thumb opposition.9
The timing of initiating or modifying therapy and
frequency of exercises, such as active joint extension,
should be clearly recorded. The strength of finger and
wrist extension, pain, and potential signs and symptoms
of complex regional pain syndrome should also be
recorded.
REHABILITATION METHODS IN EACH ZONE:
FINGER EXTENSORS
Zones 1 and 2 (Mallet Finger)
Zones 1 and 2 extensor tendon injuries may involve
avulsion fracture from the base of the distal phalanx or
present as tendon disruption alone, leading to mallet
finger deformity. A full-time splint is usually applied
for 6 to 8 weeks with the DIP joint held in extension
for both types of mallet finger.10 After the initial 6 to 8
week splinting period, the splint should still be worn at
night and during strenuous activities for 2 to 6 weeks.
Several types of splint are available: foam-padded alu-
minum splints, molded plastic splints, and thermoplas-
tic splints. With splinting, most acute injuries achieve
a success rate of around 80%.11 Therapy after this
splinting period is not always continued.10,12 Patients
Figure 40-2  A mallet finger splint often used by these
authors to immobilize the DIP joint and prevent PIP joint
hyperextension with a PIP joint extension block.

430
Section 5:  Rehabilitation of Tendon Surgery
Third Phase (10 to 12 Weeks): Increase Loading
The range of active DIP joint flexion during exercises can
be increased and active extension of the DIP joint is
performed. Light daily activities can be started without
the splint and the time without splint can be gradually
increased. Night-time splinting continues. If at any time
a marked DIP joint extension lag occurs, joint flexion
exercises should be stopped and the splint reapplied for
4 extra weeks. The patient can return to all activities
once the splint can be discontinued during the day.
Nighttime splinting may be continued for another
month or two for some patients.
When therapy starts late or joint stiffness or other
complications develop during the course of therapy,
full DIP joint flexion may not be reached for 6 or 8
months. If a DIP flexion contracture has developed,
therapy should focus on correcting the flexion contrac-
ture first.
Zone 3: Central Slip Lesion
(Boutonnière Deformity)
Closed injuries are often missed in the acute stage
because the triangular ligament initially maintains
the dorsal positions of the lateral bands, helping to
maintain extension of the PIP joint. Only when the
triangular ligament yields and allows the lateral bands
to slip volarly will the lack of PIP joint extension
become apparent due to the developing boutonniere
deformity.14-16 Chronic bou­tonnière deformity leads to
flexion contracture of the PIP joint.
For a boutonnière deformity after closed injuries, the
underlying pathology is attenuation or partial disrup-
tion of the central slip and nonsurgical treatment is
attempted first. The goal is to immobilize the PIP joint
in extension to allow the attenuated tendons to heal. A
thermoplastic splint can be used to maintain the PIP
joint in full extension for 6 weeks on a full-time basis.
Others prefer the PIP joint to be pinned in extension for
the first 3 weeks, or even 5 to 6 weeks. The DIP joint
should be free and is allowed to move actively. The
immobilization by splinting should be extended beyond
6 weeks if extension lag continues to exist after initia-
tion of PIP joint flexion. After the splint is discontinued,
gradual active flexion of the PIP joint is initiated.
After surgical treatment of the persistent deformity or
surgical repair of an open, complete disruption of the
central slip, the goal of therapy is to protect the central
slip, prevent adhesions, prevent DIP joint stiffness and
extension lag, and maintain the length of the oblique
retinacular ligament (ORL).
Treatment protocols include immobilization, passive
early mobilization with a dynamic splint and immedi-
ate controlled active mobilization of the PIP joint.17-19
Maddy and Meyerdiercks advocate immobilizing
the PIP at 0° of extension with a static finger-based
splint for 3 to 312 weeks followed by another 3 weeks
The patient is instructed not to carry heavy objects
with the injured hand and to avoid playing contact
sports, etc. We prefer not to have the patient remove the
splint but have the therapist remove the splint carefully
every week. If the skin has signs of pressure or macera-
tion, the splint needs to be adjusted or an alternative
splint design considered.
Second Phase (6/8 to 10 Weeks): Initial Exercises
Without Splint
Only when there is no extension lag at the DIP joint can
the patient be allowed to start actively flexing the DIP
joint and gently increasing the flexion force. The active
exercises are done 3 to 5 times a day out of the splint.
The amount of DIP joint flexion can be controlled by
using a large diameter cylinder (Figure 40-3). With the
hand on top of the cylinder rolling it forward allows
some DIP joint flexion and rolling it backward provides
a passive assistance for DIP joint extension. The patient
then lifts the hand and holds this position for 5 seconds,
similar to place-hold active exercise. General instruc-
tions are to do the exercises slowly and stay within pain
limits. The splint is worn between exercises and at night.
Only when patients understand the exercises should
they do them independently at home.
Therapists should measure the active flexion and
extension of the DIP joint, and if there is adequate active
DIP extension, both the frequency of exercises and the
range of joint motion allowed can be increased gradually.
The range of flexion exercises can be increased by using
a smaller cylinder. Weaning off the splint gradually and
carefully increasing flexion range prevent a recurrence
of the mallet finger. Most patients require close monitor-
ing of splint wear and exercises during this phase.
Figure 40-3  In the first phase of exercises after a mallet
finger injury, the amount of DIP joint flexion is controlled by
using a large diameter cylinder. When the cylinder with the
hand on top is rolled forward it allows some DIP joint
flexion; rolling back the DIP joint is pushed into extension
similar to a place-hold exercise.

431
Chapter 40:  State of the Art of Extensor Tendon Rehabilitation
patient grasping a large cylinder. PIP joint flexion is
increased by 15° each week.
From week 3, place-and-hold exercises for finger
extension can be incorporated by passively extending
the PIP joint and actively maintaining the joint at this
position. The exercises are done one to two hours with
10 repetitions each session.
At week 4, active PIP joint extension is started. The
patient continues to use a small finger extension splint,
such as a neoprene splint, during the day to support
extension. At week 5, achieving 90° of active PIP joint
flexion is the goal. Starting at week 6, light activities of
daily living are allowed. Resisted finger motion is started
at week 8. If the PIP joint has less than 60° of passive
flexion, corrective flexion splinting can be applied by
bandaging the fingers in flexion for periods of 10 to 15
minutes, several times a day. We emphasize that the
above program is progressed only when the patient
maintains full active PIP joint extension. If the patient
does not have full extension, the period of splinting in
extension should be prolonged, and more emphasis
should be given to extension than to flexion. Walsh and
colleagues22 compared the dynamic extension splint
with static splinting and the results suggested an earlier
return to work with less frequent therapy visits for the
dynamic extension splint group.
Early active digital extension is also used after surgical
repair of the extensor tendon in this zone. This method
advocated by Evans is known as Short Arc Motion.17 In
the active motion protocol, for the first 4 to 6 weeks, the
digit is immobilized in a finger based static splint in full
extension with the MCP joint excluded. Every hour the
patient is allowed to perform intermittent limited active
flexion and extension exercises using a template splint
(Figure 40-6). The patient actively flexes to the limits of
the template splint and actively extends to full extension.
The template splint supports only the PIP and DIP joints
of active flexion and assisted extension with a finger-
based dynamic extension splint.20
The dynamic extension splint in this study was a
custom-made spring coil finger-based splint, much like
a Capener splint (Figure 40-4). At 3 weeks, when the
dynamic splint was fabricated, a structured program
with three exercises was given to the patient: active DIP
flexion with the PIP blocked manually in extension,
active PIP flexion with the MP blocked in extension, and
composite active MP, PIP, and DIP flexion, with the
splint extending the PIP joint between repetitions of
each exercise.
Some authors start the early controlled mobilization
using a prefabricated spring coil dynamic PIP extension
splint (Capener) as early as 10 to 14 days postopera-
tively, following immobilization of the PIP joint in
extension with a static splint.21 Both groups initiated
active DIP flexion exercises with the PIP held in exten-
sion within the first week. However, with extensive
dorsal tissue damage or severed lateral bands the DIP,
exercises were postponed until 4 weeks post-operation.
Because the intent is to passively extend not against
the resistance of a flexion contracture but as an assist to
healing extensors, the spring coil does not need to be
very strong, and a lower force type can be chosen or
fabricated. The goal of the splint is to achieve maximum
extension at rest with the lowest possible force and at
the same time make active flexion easy to perform. The
position of full PIP extension (possibly 10° of hyper­
extension) in both the static and the dynamic splints is
critical for a successful outcome.23
The dynamic splinting can also be hand based,22,23 or
extending across the wrist. The wrist is usually placed in
30° extension, metacarpophalangeal (MCP) joints in
slight flexion, and interphalangeal (IP) joints fully
extended24,25 (Figure 40-5). The DIP joint is allowed to
actively flex to release the tension on the central slip
and to ensure some gliding of the intact lateral bands.
During the first week of active PIP joint flexion, the PIP
joint is allowed to flex up to 30° either by putting a stop
on the outrigger line, by adding a volar splint or by the
Figure 40-4  A Capener splint for correction of PIP joint
deformity.
Figure 40-5  Dynamic extension splinting after surgical
repair of zone 3 extensor tendons. Here active flexion and
passive extension is achieved by the dynamic traction of the
extension splint.

432
Section 5:  Rehabilitation of Tendon Surgery
adherence. which restricts gliding of the extensor hood.
Therapists need to be aware of this extra risk and start
active gliding exercises as soon as possible. Protective
tendon gliding can be provided with a dynamic exten-
sion splint as in zone 3 injuries, or using gentle active
exercises similar to the short arc motion method advo-
cated by Evans.17
Zones 5 and 6: MCP Joint and Dorsum
of the Hand
In zone 5, the sagittal bands are responsible for main-
taining the extensor tendon central over the MCP joint.
Injuries may lead to subluxation of the extensor tendon.
The sagittal bands are tensed during MCP joint flexion.
Following sagittal band repair, MCP joint flexion should
be gentle. Flexing the IP joints while the MCP joint is
kept in extension reduces tension on the sagittal bands
and ensures some gliding of the tendon.
The zone 5 extensor tendon repair can adhere to the
MCP joint capsule. Immobilization of the joint in exten-
sion would lead to contracture of the collateral liga-
ments. To prevent these complications, use of a protocol
with early MCP joint flexion is beneficial.
In zone 6, the extensor tendon is more rounded than
distal extensor tendons and allows stronger surgical
repairs. Because of the juncturae tendinum connections
between the EDC tendons on the dorsum of the hand,
movements of adjacent finger(s) affect tension on the
repaired tendon. When the repair site is distal to the
interconnections, flexion of the adjacent digits may pull
through the interconnections on the proximal stump,
moving it distally, toward the distal stump, bringing the
tendon ends together. Therefore, splinting the adjacent
digits in slightly greater flexion or excluding the adjacent
digits from immobilization and allowing them to flex,
may reduce tension across the repair site. In contrast,
when the repair site is proximal to the interconnections,
flexion of the adjacent fingers may increase tension on
the repair site so the uninjured fingers should be held
in the same position in the splint. Zone 6 repairs benefit
from early motion protocols as postoperative edema on
the dorsum of the hand limits MCP joint flexion; early
motion facilitates reducing edema.
Lacerations of the extensor tendons in zones 5 and 6
are often treated with an early motion protocol using a
dynamic splint.27 Within 2 to 5 days postoperatively, a
dorsally based dynamic splint is applied holding the
wrist in 30° to 40° extension and dynamically extend-
ing the MCP joints to full extension via an outrigger
(loops around the proximal phalanx). At this time active
flexion of the MCP joints is allowed to 20° to 30° for
index and middle fingers, and 35° to 40° for the ring
and little fingers while the IP joints are actively extended.
The second exercise is to flex the IP joints while main-
taining extension of the MCP joints (active hook fist).
Exercises are done 5 times every 2 hours in the first 2
and allows for 30° active PIP and DIP joint flexion in
the first 2 weeks, increasing 10° flexion each week.26
Exercises are performed 4 to 6 times a day. At 6 weeks,
both splints are discontinued, and exercises of full range
of active motion are started.
Evans and Thompson17,26 calculated that the amount
of force added on the repair was 3 N during early active
motion, which is well within the strength of most suture
techniques (about 20 N in the first week)6 and the 30°
flexion limit keeps the force within a safe zone.
Focusing on zones 2 and 3, the studies of Hung,27
Saldana,28 and Newport29 do not support the idea that
dynamic splinting has benefits over static splinting.
Therefore some regard static splinting to be as effective
and should be the standard care for simple extensor
tendon lacerations distal to the MCP joint, given its
simplicity and lower cost.
For chronic boutonniere deformity of PIP joint with
a flexion contracture of over 30°, serial casting is a
useful way to correct the deformity. The cast is applied
to the PIP joint and is changed approximately every 2
weeks; PIP joint extension is progressively increased by
5° to 10° after each change of the cast.
A frequent complication in treating closed bouton-
niere deformity is incomplete correction of the defor-
mity. However, an extension lag of about 20°, in the
presence of full PIP and DIP active flexion leads to little
functional disturbance of a finger, and therapy does not
necessarily need to continue if these degrees of digital
motion have been achieved.
Zone 4: Proximal Phalanx
In this zone, injuries to the tendon often cause no
functional loss because the wide extensor hood covers
more than half of the proximal phalanx. However, when
the tendon is repaired, due to the large contact area
to the underlying bone, it creates a substantial risk of
Figure 40-6  Limited active flexion and full extension
exercises of the PIP and DIP joints using a template splint.
Template
splint
Full ext.
Limited
flexion

433
Chapter 40:  State of the Art of Extensor Tendon Rehabilitation
(Figure 40-7). When extension lag persists, the splint
can be worn at night or part of the day, with a V-shaped
insert holding the MCP and IP joints at 0° (Figure
40-8). During weeks 4 to 6, the splint can be removed.
If an extension lag of the MCP joint is persistent and is
greater than 30°, splinting is resumed for 2 more weeks.
If no lag is present, active finger flexion exercises are
initiated at week 4 without wrist flexion.
A study by Bulstrode and colleagues32 compared the
outcomes of the Norwich regimen to immobilization of
all the joints of the hand and static extension splinting
of the wrist and MCP joints while leaving the IP joints
free. The total active range of digital motion in the
immobilization group was less at week 4. However, by
12 weeks, there was no statistically significant difference
between the three regimens, and all patients achieved
good or excellent results. The only significant finding
was reduced grip strength compared to the uninjured
hand at 12 weeks for the immobilization group.32 Other
investigators found better functional results for the
dynamic splint method over early active mobilization
at 4 weeks but failed to identify significant differences
in functional outcomes at 12 weeks.33 A study compar-
ing a dynamic outrigger splint with a static volar MCP
blocking splint also did not show statistically significant
differences in functional outcomes.34 There is no
weeks, and after that 10 times every hour. For patients
needing more support because they experience unrest
during sleep or exhibit a tendency toward development
of an extensor lag, a volar splint may be used to keep
the fingers in extension at night. At week 2, active flexion
of MCP joints is allowed to 45°, week 3 to about 60°
and week 4 to 90°. At week 4, place-and-hold exercises
for full finger extension are started and at week 6 the
splint is discontinued. Strengthening exercises are
started around week 7 and at 12 weeks the hand can be
used normally. Patients are progressed on this timetable
only if no extensor lag is observed.
For isolated EDC injuries in zones 5 and 6, the imme-
diate controlled active motion (ICAM) technique can
be used, as described by Howell and colleagues.30 This
method allows immediate active controlled motion by
using a yoke splint which places the injured finger in
MCP joint extension slightly greater than the adjacent
fingers, and a second splint is applied to hold the wrist
in 20° to 25° extension. The ICAM protocol has the
following three phases: (1) Days 0 to 21—Two splints
are worn full time and active composite finger flexion
exercises are started. (2) Days 22 to 35—The patient
continues to wear the yoke splint at all times but removes
the wrist splint only for wrist exercises. (3) Days 36 to
49—The wrist splint is discontinued but the patient
continues wearing the yoke splint while using the hand
but removes it for active finger motion exercises. In
Howell’s report 81% patients achieved excellent results
for both digital extension and flexion.30 Only five patients
developed a lag and there were no tendon ruptures.
The Norwich regimen31 designed for zone 5 and 6
extensor tendon rehabilitation is an active protocol
using a volar immobilization splint, which hold the
wrist in 45° extension, the MCP joints in 50° flexion,
and the IP joints in full extension. Two exercises are
performed for 4 weeks, 4 times a day, with 4 repetitions
each time (“the rule of four”). After loosening the digital
straps the following exercises are performed: combined
active MCP and IP joint extension (lifting the extended
digits off the splint) and active MCP joint extension
with IP flexion (making a hook fist within the splint)
Figure 40-7  Norwich exercise regimen. Combined active MCP and IP joint extension (lifting the extended digits off the
splint) and active MCP joint extension with IP flexion (making a hook fist within the splint).
Figure 40-8  The Norwich splint can be modified with a
V-shaped insert holding the MCP and IP joints at 0° when
extension lag persists.

434
Section 5:  Rehabilitation of Tendon Surgery
finger extensors. A static splint or a dynamic extension
splint is used for 4 to 6 weeks, followed by progressive
exercises of active thumb extension for a further 2 to 4
weeks. Between 70% and 90% of repairs of the EPL
tendons achieved excellent or good results, with these
protocols.46 Loss of thumb movement is often caused
by tendon adhesions to the bone and skin, with thicken-
ing of the dorsal joint capsules with scar formation.46
Zone T1
Although not seen as frequently as mallet finger, a
closed mallet thumb can be treated similarly to the
closed zone 1 digit injury with 6 to 8 weeks of immo-
bilization of the thumb with splinting. Afterward,
splinting is applied between exercises and at night for
an additional 2 to 4 weeks, and thumb extension is
allowed. However, a zone T1 laceration can also be
surgically repaired by virtue of the large size of the EPL
tendon as compared to the finger extensor.
After surgical repair, Elliot and Southgate46 place the
hand in a palmar plaster of Paris splint with the wrist in
30° of extension and both the MCP joint and the IP
joints in the neutral position. The tendon is mobilized
early after surgery in a dynamic extension splint, which
controls MCP joint flexion only, the loop being under
the proximal phalanx. The IP joint moves freely into
flexion and extension from the commencement of mobi-
lization. For the first 5 days, the hand rests in the splint.
From day 5, the MCP joint alone is actively flexed and
extended, with the sling under the distal phalanx and
the base of the thumb supported with the other hand.
In the second week, the patient actively flexes and extends
both MCP and IP joint together, aiming for the base of
the ring finger, 10 repetitions hourly. Between days 12
and 28, the patient performs active combined MCP and
IP joint flexion and opposition, aiming for the base of
the little finger. After 28 days, the same exercise are con-
tinued without the splint, active combined MCP and IP
joint flexion and opposition are continued and thumb
active abduction, adduction, and opposition exercise are
added. These exercises continue from week 8 onward.
Zone T2
Injuries to the extensor at the proximal phalanx of the
thumb can be immobilized in a hand-based splint,
holding the MCP and IP joint at 0° and the thumb radi-
ally extended. Gentle active exercises can be initiated at
3 weeks, progressing flexion of the thumb slowly while
splinting between exercise sessions.
Due to a high incidence of tendon-to-bone adher-
ence in this zone, early protective motion protocols
should be considered. Crosby and Wehbe18 use a
forearm-based splint, with the wrist in 20° to 30° of
extension, MCP joint at slight flexion of 10° to 20°,
with a rubber band holding the IP joint extended. Active
flexion of the IP joint is allowed, with the IP joint
evidence of superior results in the long term for any of
the different treatment methods in these publications.
Still, all the authors recommended early mobilization
on the basis of lower cost and the simplicity of the early
active mobilization regimens with less time spent in
therapy and especially with simpler splint fabrication
and fewer adjustments. However, early mobilization has
achieved earlier return of grip strength, which may be a
reason to choose early mobilization over static treat-
ment.32,45 One group, however, states that with similar
outcomes the static regimen should be the treatment of
choice with poorly compliant patients, as it is both a
simple and an effective method.35
Ip and Chow36 and Kerr and Burczak37 used a dynamic
splinting regimen after extensor tendon repairs in zones
4 to 8, with good outcomes and no rupture of the repairs.
Newport and colleagues38 used static splinting for all
zones of injury and achieved similar good results. Static
splinting has been described by Blair and colleagues,39
Slater and colleagues,40 and much earlier by Stuart and
colleagues41 and Dargan.42 Soni and colelagues43 recom-
mended early active motion and consider it especially
beneficial in complex injuries distal to the MCP joint.
Zone 7: Extensor Retinaculum at the Wrist
In zone 7, the extensor tendons run through tight com-
partments formed by the extensor retinaculum and
distal radius. This restraining retinaculum increases the
risk for adhesions and limits gliding of the healing
tendon. For these reasons, early mobilization of the
tendon in this zone is preferred.
A practical treatment method is to splint the wrist in
30° to 40° of extension, and the MCP joints in 0° to
15° flexion for 3 to 4 weeks, allowing the PIP and DIP
joints full active range of motion. After 3 to 4 weeks the
MCP joints are left free but the wrist is still held in
extension in a splint. Wrist flexion is started and weaning
from the splint occurs while slowly increasing activities
over the next 2 weeks.44
Zone 8: Forearm
Injuries in the forearm involve tendons, the muscle–
tendon junction, or the muscle belly. Following surgery,
the wrist is immobilized in mild extension with a pro-
tective splint, but thumb and fingers are left out of the
splint. Although active digital motion can be allowed,
lifting of heavy objects should be avoided. Adhesions
occurring in this zone are unlikely to restrict tendon
motion.
REHABILITATION METHODS IN EACH ZONE:
THUMB EXTENSORS
The three thumb extensors, extensor pollicis longus
(EPL), extensor pollicis brevis (EPB) and abductor pol-
licis longus (APL), can be injured, mostly through open
wounds. Rehabilitation after repair is similar to that for

435
Chapter 40:  State of the Art of Extensor Tendon Rehabilitation
passively extended in the first 3 weeks. At weeks 4 to 5,
gentle active extension is initiated with the addition of
gentle active flexion of the IP joint. At week 8 and
beyond, graded resistive exercises are initiated. Elliot
and Southgate46 use the same regimen after zone T2 to
T4 EPL tendon repair as in zone T1.
Zones T3 to 5
The proximal zones of the thumb include the EPL,
extensor pollicis brevis (EPB), and abductor pollicis
longus (APL). After repair of the APL tendon at zones
T3 and 4, the wrist is positioned at 30° extension, with
slight radial deviation and metacarpal extension. Exten-
sion is increased to 40° for zone T5 APL repair.
After EPL repair, immobilization by a static splint
worn full-time for 4 weeks, and starting active exercises
at 4 weeks is common practice. The splint can be dis-
carded at 6 weeks. However, if there is an extension lag
at 6 weeks, extension splinting is continued for another
2 weeks. If stiffness of the joints prevents full thumb
flexion and extension is normal or close to normal,
passive flexion exercises and flexion splinting should
start after 7 to 8 weeks.
Dynamic splinting protocols vary; for postoperative
management of a zone T4 to 5 laceration, Evans17 recom-
mended a dorsal forearm–based splint that positions
the wrist at 30 to 40° of extension, the carpometacarpal
joint at neutral, and the MCP joint at 0°. The IP joint is
held in neutral by a sling with dynamic traction. The
splint allows 60° of active IP joint flexion, which is pas-
sively brought back to extension via dynamic traction.
Evans now adds more motion when the patient is at
therapy. She does controlled passive motion to the MCP
joint of approximately 30° while the wrist is held in
maximum extension and the IP joint is held at 0°. To
ensure gliding, an “active hold” exercise is also done.
Crosby and Wehbe18 use a forearm-based splint and
rubber band traction to the proximal part of the thumb.
The MCP joint motion is not restricted. The rubber band
traction produces MCP joint extension after active MCP
joint flexion (Figure 40-9).
In the Manchester regimen (personal communica-
tion, with Fiona Peck, 2011), a 2 part forearm based
thermoplastic splint is applied with wrist at 30° and
the MCP and IP joints at 0°. The thumb is positioned
comfortably in abduction and extension. The remov-
able plate allows active 0° to 60° IP joint flexion and
extension exercises (Figures 40-10 and 40-11). In week
1, active IP joint flexion and extension exercises are
initiated, progressing to 60° flexion during the first 3
weeks. No passive flexion is permitted and the splint is
worn full time. At week 3, the splint is removed to
perform careful wrist flexion and extension with thumb
relaxed to encourage tenodesis action and promote
tendon glide, especially in zone T5 where there is risk
Figure 40-9  Drawing showing the protective active thumb
motion with a forearm-based dynamic splint for the extensor
tendons injured proximal to the MCP joint level of the
thumb. The rubber band permits active MCP joint flexion
and passive extension of the joint.
Rubber band
Figure 40-10  A two-part forearm based thermoplastic
splint with a removable plate allows flexion and extension of
the IP joint of the thumb following repair of the EPL tendon
in zones T 3 to 5. (Courtesy of Fiona Peck.)
of adhesions under retinaculum. At week 5 the splint is
removed. Active IP joint flexion and extension exercises
are continued and composite active MCP and IP joint
motion is encouraged to restore opposition. In the pres-
ence of extensor lag, the splint continues to be worn at
night. Corrective passive flexion is initiated at 6 weeks
if flexion is slow to progress. The patient returns to
normal activity by weeks 10 to 12.
Hung and colleagues27 described a dynamic exten-
sion splint for all EPL repairs that places the wrist at
40° extension and the thumb in mid-abduction/mid-
extension. The IP joint is allowed to flex to 45° in the
first week, and full flexion is allowed in the second

436
Section 5:  Rehabilitation of Tendon Surgery
motion 3 to 5 days after repair and continued exercise
in the splint for 5 weeks. Chinchalkar48 suggested active
wrist and thumb motion to start at 4 to 6 weeks for
injuries at the first extensor compartment.
CLINICAL EVIDENCE AND OUTCOMES
Currently few randomized comparative clinical studies
are available for extensor tendon injuries; rehabilita­
tion methods for individual patients are largely
decided according to personal preference and judg-
ment. Talsma and colleagues49 reviewed findings from
five studies (Table 40-1), about one of the three regi-
mens: (1) immobilization, (2) early controlled passive
mobilization (dynamic splinting), or (3) early active
mobilization. There is strong evidence that early con-
trolled mobilization regimens used after surgical repairs
in zones 5 and 6 lead to better recovery of both range
of motion and grip strength of the hand compared
with immobilization in the short term. Nevertheless,
no conclusive evidence can be uncovered regarding the
long-term (6 months) effectiveness of these regimens.
The methods used to assess the outcomes include Buck-
Gramcko, White, Miller11 and Dargan.42
Surgical repairs of open extensor tendon injuries fol-
lowed by appropriate postoperative care have generally
favorable outcomes.27-29,36 Ip and Chow36 reported 101
primary end-to-end extensor tendon repairs due to
clean-cuts or crush open injuries in zones 4 to 8 of all
fingers or zones 1 to 6 of thumbs in 84 patients treated
with dynamic splinting and passive mobilization.
Among 37 thumb extensor repairs, there were 67% (25
tendons) with excellent results, 30% (11 tendons) good,
and 3% (one tendon) fair by Buck-Gramcko criteria.
week. In the third week, the MCP joint is allowed to flex
to 45°, increasing to full composite flexion by week 4.
The dynamic splint is discontinued in week 5 and is
replaced by a wrist cock-up for one additional week. For
T4 and proximal zone injuries, full active flexion of the
thumb is allowed within the dynamic splint, with the
wrist held in 20° of extension.
Browne and Ribik47 also allowed full flexion across
the palm in a dynamic outrigger splint. They began
Table 40-1  Summary of Studies Comparing Outcomes of Extensor Rehabilitation Protocols49
Authors
Interventions/Groups
Zones
Outcomes
Mowlawi et al45
N = 34*
Start 3 to 5 postoperative days
A.  Dynamic extension splint 4 weeks
B.  Immobilization for 3 weeks, active motion at  
4 weeks
5, 6
ROM in group A was significantly
better than group B at 4 to 8 weeks
and grip strength was greater at 8
weeks, but not at 6 months
Bulstrode et al32
N = 42
A.  Immobilization in plaster, 4 weeks
B.  Extension splint: wrist in 30° and MCP joints
extended; IP joints free, 4 weeks
C.  Extension splint: wrist in 45° and MCP in 50°
flexion, IP joints free, 4 weeks
5, 6
ROM in group B and C was better than
A at 4 and 6 weeks not at 8 and 12
weeks with a compared with B
Chester et al33
N = 54
A.  Early active motion. with a static splint,  
4 weeks
B.  Dynamic extension splint, 4 weeks
4 to 8
ROM was better with a compared with
B at 4 weeks, but not at 12 weeks
No ruptures
Khandwala et al30
N = 100
A.  Dynamic extension splint 4 weeks
B.  Early active with palmar blocking splint 4 weeks
4 to 8
ROM had no differences in two groups;
1 rupture in group 1 and 2 in B
Russell et al35
N = 65
A.  Immobilization 2 to 3 weeks and dynamic
splint 4 weeks.
B.  Immobilization, 4 weeks followed by static
splint, 4 weeks
5 to 8
ROM had no differences between A
and B.
No ruptures
*N is the total number of patients.
Figure 40-11  Active flexion and extension of the IP joint of
the thumb within this splint following repair of the EPL
tendon. (Courtesy of Fiona Peck.)
Active IP joint flexion
Active IP joint extension