Principles of Tendon

Tr ans f er
Danielle Wilbur, MDa, Warren C. Hammert, MDb,*

KEYWORDS
 Tendon transfer  Principles of tendon transfer  Restoration of nerve injury
 Reconstruction following nerve injury

KEY POINTS
 Tendon transfers can be useful for restoration of function following peripheral nerve injuries or other
conditions affecting the muscle/tendon units.
 Essential elements for successful tendon transfer include (A) supple joints: it is easier to prevent
contractures than reverse them, so maintaining passive motion is preferable; (B) tissue equilibrium:
timing of transfer is based on appropriate wound healing and scar maturation; (C) adequate
strength and excursion; (D) one tendon for each function if possible; (E) straight line of pull; (F)
expendable donor; (G) synergistic transfer (preferred, but not mandatory).
 If identified early, consider nerve reconstruction before embarking on tendon transfers.

INTRODUCTION (see Garcia RM, Ruch DS: Free flap functional

muscle transfers, in this issue). The first success-
Tendon transfers provide a substitute, either tem- ful tendon transfers were performed in the foot to
porary or permanent, when function is lost due to treat deformities caused during the polio endemic
peripheral nerve injuries or injuries to the muscu- in Vienna in the 1880s by Carl Nicoladoni. This
lotendinous unit itself or when function is imbal- tendon transfer was expanded on by Codivilla,
anced due to spasticity from neurologic injury in an Italian surgeon who performed a series of
stroke, cerebral palsy, or central nervous system 30 tendon transfers and introduced the concept
lesions. Understanding of the fundamental prin- of muscle balance and need for preoperative
ciples of tendon transfer allows the surgeon to assessment of donor muscle force.1 The mid-
establish, strengthen, or augment motor func- 1900s brought further advances by hand sur-
tion that has been compromised. The tendon geons that bear the names of many transfers
transfer itself is the release of a terminal or prox- that are still performed today, including Mayer,
imal tendon insertion from one functional muscle- Almquist, Steindler, Bunnel, Brand, Boyd, Omer,
tendon unit and reinsertion distally to restore lost and Jones.1–3
or deficient muscle action.1 In contrast, a tendon It is essential to understand the fundamental
graft is transected both proximally and distally elements that allow tendon transfer surgery to
and used as an intercalary segment without pre- restore function. Only with understanding of the
serving its neurovascular supply. A free muscle general principles of tendon transfer can the sur-
transfer involves transecting both the origin and geon formulate an effective treatment algorithm
the insertion of a musculotendinous unit and per- specific to the patient’s needs.
forming a distal revascularization and nerve repair

Disclosures: None.
hand.theclinics.com

a
Department of Orthopaedics and Rehabilitation, University of Rochester Medical Center, 601 Elmwood
Avenue, Box 665, Rochester, NY 14612, USA; b Division of Hand Surgery, Department of Orthopaedics and Reha-
bilitation, University of Rochester Medical Center, 601 Elmwood Avenue, Box 665, Rochester, NY 14612, USA
* Corresponding author.
E-mail address: Warren_Hammert@URMC.Rochester.edu

Hand Clin 32 (2016) 283–289

http://dx.doi.org/10.1016/j.hcl.2016.03.001
0749-0712/16/$ – see front matter Ó 2016 Elsevier Inc. All rights reserved.
284 Wilbur & Hammert

GENERAL PRINCIPLES OF TENDON TRANSFER allow improvement in function. When performed

Preoperative Assessment late or when there is no chance of nerve recovery,
either end-to-end or end-to-side transfers can be
Tendon transfers can be used to restore function
used at the surgeon’s discretion.
to a joint, recover a specific motion across a
joint, improve deformity, or to act as an internal
Tissue Equilibrium
splint to support partial function after distal nerve
injury while awaiting recovery of a peripheral nerve The term tissue equilibrium was coined by Stein-
injury.4 Appropriate expectations of both the patient dler and infers that maturation of the tissue bed
and the family must be ascertained before interven- has occurred before the timing of the tendon trans-
tion with a careful assessment of the needs and fer.3 The motion after a tendon transfer is hindered
goals of the patient and caretakers. A thorough if there is an inadequate gliding surface due to scar
evaluation of the patient is necessary to map out tissue, residual edema, or residual joint stiffness. If
the functional deficits requiring transfer, establish the tissue bed is not pliable, the surgeon may
which muscles are available for transfer, and deter- consider flap coverage or developing a subcutane-
mine the sensibility of the affected limb.5 If the limb ous bed or tunnel through which to place the trans-
is insensate, the reduced sensory feedback will fer in unscarred tissue. A silicone rod may also be
cause the brain to exclude these areas from func- used to create a smooth tunnel for later tendon
tional activity and lead to suboptimal outcomes.2 transfer in a second-stage procedure. If these
The time from injury, the type of injury, and success soft tissue constraints cannot be overcome, then
of previous treatment should be delineated. Electro- another transfer should be chosen. Brand7
diagnostic studies may be helpful in determining the described tunneling the tendon transfer with blunt
extent of motor loss of a limb and predicting further dissection through natural subcutaneous tissue
muscle recovery by the presence of polyphasics. planes in order to find the path of least resistance
Although clinical examination is sufficient in many for the tendon transfer. The use of curved incisions
instances, there are scenarios where electromyog- allows for the tendons to be placed under subcu-
raphy (EMG) can greatly enhance the diagnostic taneous flaps and avoids placement of the
capabilities. Evaluation of 2 muscles with similar rerouted tendons under the incision site.7,8 Contra-
function, such as pronator teres and pronator indications for an elective tendon transfer surgery
quadratus, can determine if one is expendable, include chronic wounds, contractures, or evidence
such as with use of pronator teres transfer for wrist of bony instability/nonunion below the area of the
extension in radial nerve palsy. In addition, elec- transfer.4 The involved muscle will have diminished
trodiagnostic studies can clarify confounding pic- function in scar tissue because of the increased
tures by demonstrating abnormal nerve patterns, fibrotic tension and shorter residual fiber length of
such as Martin-Gruber connection or Riche- the muscle-tendon unit.4
Cannieu connections, which may have implications
on choice of transfer. Omer4 described quantitative Mobile, Intercalary Joints
tests that should be performed during preoperative
The basis upon which tendon transfers work is the
assessment, including voluntary muscle tests and
application of an active motor unit across a passively
measurements of range of motion, measurement
mobile joint that has adequate stability. Dowd and
of 2-point discrimination, gross grip and finger pinch
Bluman6 described the concept that tendon transfer
strength tests, and timed pickup tests in cases
procedures are most effective when they are used
involving median and/or ulnar nerve lesions. Real-
to correct supple deformities caused by dynamic
istic treatment goals must be agreed on by the pa-
muscular imbalances. The preoperative range of
tient, caretakers, surgeon, and rehabilitation team
motion will never be exceeded by active motion
to ensure postoperative success. Unrealistic ex-
following a tendon transfer; thus, maximal preoper-
pectations that are not addressed before surgery
ative passive motion must be obtained. Preopera-
can lead to frustration, lack of trust in the sur-
tive rehabilitation with occupation and/or physical
geon and rehabilitative team, and lack of perceived
therapy is required to re-establish supple passive
improvement in function.6
range of motion of the targeted joints.1,9 Adjuvant
Tendon transfers are not time-sensitive, but
therapies such as casting, stretching, dynamic
when reconstruction for traumatic injuries is un-
orthotic use and/or surgical release of joint con-
dertaken before motor end plate degeneration
tractures before tendon transfer surgery may be
(typically 18 months from injury), nerve grafting or
necessary to achieve mobile joints.
nerve transfers should be considered. If tendon
On the contrary, hypermobile joints or exces-
transfers are performed early, end to side are
sive joint laxity can predispose tendon transfers to
preferable, because regeneration of the nerve will
overcorrection. For example, patients undergoing
 Principles of Tendon Transfer 285

tendon transfers into the lateral bands for intrinsic Table 2

reconstruction may result in swan neck deformities Forearm muscle work capacity from strongest
in the presence of proximal interphalangeal joint to weakest
hyperlaxity. Tendon transfers should be used with
extreme caution in patients with known collagen Donor Muscle m-kg
disorders or hyperlaxity. FDS 4.8
FDP 4.5
Donor Muscle Properties FCU 2
Adequate strength BR 1.9
The donor tendon must have adequate strength to PT 1.2
perform its intended function. Its strength must be FPL 1.2
at least a 41 or 5 out of 5 on the Lovett scale
FCR 0.8
(Table 1), because motor strength has been shown
to lose one grade postoperatively.1,10 An injured PL 0.1
muscle with denervation that subsequently recovers Recipient Muscles m-kg
is not an acceptable donor muscle due to the loss of EDC 1.7
strength that will result when it is transferred.9
ECRL 1.1
When selecting a donor, the surgeon must take
ECU 1.1
into consideration the work capacity of the mus-
cles and choose one that is has sufficient strength ECRB 0.9
to perform the recipient function. The work capac- EIP 0.5
ity of a muscle depends on both its fiber length and EPB 0.1
its cross-sectional area, given that work is a prod- EPL 0.1
uct of both force and distance. The work capacity APL 0.1
of a given muscle is therefore proportional to its
mass or volume. The relative tension capacities Abbreviations: APL, abductor pollicis longus; BR, brachior-
adialis; ECRB, extensor carpi radialis brevis; ECRL, extensor
of the forearm and hand muscles were deter-
carpi radialis longus; ECU, extensor carpi ulnaris; EDC,
mined by dividing the tendon’s fiber length (excur- extensor digitorum communis; EIP, extensor indicis pro-
sion) into the volume of each muscle to determine pius; EPB, extensor pollicis brevis; EPL, extensor pollicis
the cross-sectional area. The work capacities for longus; FCR, flexor carpi radialis; FCU, flexor carpi ulnaris;
hand and forearm muscles are seen in Table 2.11 FDP, flexor digitorum profundus; FDS, flexor digitorium
superficialis; FPL, flexor pollicis longus; PL, palmaris lon-
Multiple factors contribute to the loss of muscle gus; PT, pronator teres.
grade strength after tendon transfers, including the
need for the muscle to pull through postoperative sarcomere’s normal, natural resting length, then
adhesions, differences in the line of pull between the muscle is placed on tension into the passive
the donor and recipient muscle, and the tensioning portion of a Blix curve, which corresponds to an inef-
of the donor muscle.4,12 When muscles are ficient biomechanical starting tension.6 This corre-
tensioned during surgery, the surgeon attempts to sponds to an inefficient starting tension
mimic the natural resting length of the muscle. If biomechanically. Friden and Lieber13,14 showed
the muscle is set at a length that is longer than the that when a donor muscle is overtensioned into the
passive portion of the Blix curve, its potential con-
Table 1 tractile force decreases to 28% of its maximum
Lovett scale force. This paradoxical loss of sarcomeres occurs
when tension is too tight, resulting in a passive
Grade Muscle Effort tenodesis effect of the transfer, causing a decrease
0 No movement in strength and function of the transfer.12 This
1 Contraction visible or palpable, decrease in strength and function of the transfer is
fasciculations due to the inability to develop active tension in a
2 Active movement with gravity stretched sarcomere, with a suboptimal interaction
eliminated between myosin and actin filaments, producing sub-
3 Active movement against gravity optimal muscle force.13 Thus, it is crucial to attempt
to set the muscle length during tendon transfer at the
4 Active muscle contraction against
gravity with some resistance
resting length of the muscle, within its effective range
of motion of the recipient joint that it is acting upon.12
5 Active muscle contraction against
It is also important to determine the amount
full resistance; full strength
of strength needed by the recipient muscle.
286 Wilbur & Hammert

Transfers can broadly be grouped into either power to benefit from the natural tenodesis effect to
or positional transfers. Power transfers would augment the amplitude of muscle excursion.3
include transfers for restoring grasp, pinch, elbow
flexion, and shoulder abduction/flexion and require Expendable Donor
more powerful donor muscles. Positional transfers,
on the contrary, do not require such powerful mus- Boyes, and later Omer,4 outlined the 50 different
cle donors and include the restoration of thumb op- muscles that are used to activate movement in
position and radial nerve function. the hand and forearm and include (Table 4):
Last, the strength of the antagonist muscles
1. 5 muscles that control supination/pronation
needs to be considered to avoid overcorrection;
2. 7 muscles that control movement of the hand at
this is especially true in cases of combined nerve
the wrist
palsy or global neurologic deficits where small alter-
3. 18 muscles that flex and extend the digits
ations in forces can have profound impacts on the
4. 20 small muscles of the hand that contribute to
overall balance of the hand. The tensioning of the
precise motion
same tendon transfer will need to be dramatically
different in a patient with cerebral palsy and sub- The redundancy in the number of muscles acting
stantial spasticity versus a patient with Charcot- together to produce a motion allows one or more to
Marie-Tooth and global neurologic weakness.
Table 4
Excursion Muscles used to activate movement in the
The tendon excursion, or amplitude, must be suf- forearm/hand
ficient to restore the lost function of the recipient
muscle and be similar to the tendon that it is Action Muscles
replacing (Table 3). The amount of excursion that
Supination/pronation Pronator teres
can be expected from tendons is directly related Pronator quadratus
to its resting fiber length and can be estimated Supinator
by the Boyes’ 3,5,7 rule: Bicepsbrachii
Brachioradialis
1. Wrist flexors and extensors: 33 mm
Movement of hand FCU
2. Finger extensors and extensor pollicis longus
at the wrist FCR
(EPL): 50 mm Palmaris longus
3. Finger flexors: 70 mm ECRB
ECRL
Augmentation of excursion can occur by the
ECU
tenodesis effect in muscles that are in-phase or EDC
synergistic to each other. Wrist flexion and exten-
Flexion/extension of FDS  4
sion can add 20 to 30 mm of excursion (effective
digits FDP  4
amplitude) through the tenodesis effect, facilitating FDM
finger extension and flexion, respectively.2,15 Mobi- FPL
lization and release of the fascial attachments of the EDC  4
donor muscle as well as release of the donor mus- EDQ
cle belly can also be used to increase the excursion EIP
of muscle, especially in the brachioradialis.2,3,15 EPL
A muscle can also be converted from a monoar- EPB
ticular unit into a biarticular or multiarticular unit APL
Precise motion Dorsal interossei  4
Palmar interossei  4
Lumbricals  4
Table 3 Thenar muscles
Tendon excursion Hypothenar muscles
Adductor pollicis
Tendon Excursion (mm) Palmaris brevis
Wrist flexors 33 Abbreviations: APL, abductor pollicis longus; ECRB,
Wrist extensors 33 extensor carpi radialis brevis; ECRL, extensor carpi radialis
longus; ECU, extensor carpi ulnaris; EDC, extensor
Finger extensors 50 digitorum communis; EDQ, extensor digiti quinti; EIP,
EPL 50 extensor indicis proprius; EPB, extensor pollicis brevis;
Finger flexors 70 FDM, flexor digiti minimi; FDP, flexor digitorum profun-
dus; FPL, flexor pollicis longus.
 Principles of Tendon Transfer 287

be used to augment function elsewhere. The donor synergistic with finger extension, and wrist exten-
tendon must be expendable, and its use must not sion is synergistic with finger flexion. Synergistic
result in considerable functional impairment after muscle contraction is easier for retraining muscle
transfer, meaning the remaining muscles must function after transfer, especially in children, who
have sufficient strength to account for the loss of have greater cerebral plasticity than adults.1,3
function the donor used to provide. For example, Use of preoperative dynamic EMG may help in
there must be a sufficient wrist flexor remaining to determining appropriate muscles for transfer.
flex the wrist after transfer for reconstruction of a Transfer of a wrist flexor for finger extension in
radial nerve palsy.3,4,9 radial nerve palsy is a common synergistic trans-
fer. If this type of tendon transfer fails, the transfer
One Tendon, One Function can still function as a tenodesis effect if it is trans-
When evaluating functional deficits and planning ferred in phase with the recipient muscle.2
the muscles available for transfer, the surgeon
must adhere to the one donor, one function rule. Surgical Technique
A single tendon cannot be expected to perform
2 functions, for example, to extend and flex the A variety of techniques have been described in the
joint, without a subsequent loss of effectiveness literature for coaptation of the donor and recipient
because of the dissipation of the force and ampli- tendons following tendon transfer. When choosing
tude of the muscle by performing 2 opposite a coaptation style, the surgeon must determine
tasks.2,3 In addition, it is difficult to use one tendon whether an end-to-end type of attachment or an
to perform 2 similar functions, such as extend the end-to-side type of coaptation can be used. This
fingers and thumb. When 2 separate insertions decision depends on multiple factors: length of
are used, the tendon that is set with the greatest tendon available for transfer, site of transfer,
tension will be the active tendon and will overpower amount of soft tissue to cover the bulkiness of
the other function.1 One tendon may, however, be the tendon transfer, tensioning of the graft, and
used to restore one function in multiple digits or caliber of the tendons. The Pulvertaft weave
multiple joints. For example, the flexor carpi radialis can be used in both an end-to-side and a side-
(FCR) or flexor carpi ulnaris (FCU) can be used to-side transfer.16 Pulvertaft tendon weave was
to restore digital extension to all fingers simul- originally described in his paper on flexor tendon
taneously. Similarly, one tendon transfer may be fixation in the hand using the palmaris, plantaris,
used to influence more than one joint in the resto- or extensor digitorum longus to the fourth toe as
ration of intrinsic function in ulnar nerve palsy by tendon grafts. He advocated the use of a fish-
simultaneously improving metacarpophalangeal mouth end-to-end interlacing stitch, first intro-
flexion and interphalangeal extension with the duced by Bunnell, which is useful when the tendon
transfer of a donor through the lumbrical canal and grafts have differing cross-sectional diameter.
and into the lateral bands. The remainder of the donor tendon is then inter-
laced through a series of 90 slits cut through the
Straight Line of Pull recipient tendon with cross stitches to interlock
the tendons together.16,17 The use of 4 to 5 weaves
The vector of motion of the tendon being trans- to increase the overall repair strength of the tendon
ferred is crucial in creating functional motions weave, including in a Pulvertaft weave, was shown
across a joint and prevention of secondary defor- to be the strongest in peak load to failure and peak
mities.12 To maximize the force and efficiency of stress biomechanically in Gabuzda’s study.18 This
a transferred tendon, the line of pull from the donor biomechanical study also evaluated the tensile
motor site to the recipient insertion site should be strength between 2 suturing techniques during
as straight as possible without the use of redirec- an end-to-end tendon repair. A cross stitch was
tional pulleys whenever possible.2,3 compared with horizontal mattress sutures, with
cross-stitch patterns having a notable increase in
Synergism
pullout strength.18
The concept of synergy was advocated by Littler The importance of high ultimate load to failure is
and relates to the concept that a transferred crucial to allow early motion protocols to prevent
tendon’s normal contractile period should be the tendon adhesion, limit postoperative complica-
same as the contractile period of the tendon that tions, and potentiate improved clinical and func-
is being augmented. A synergistic transfer allows tional outcomes.19 Pulvertaft weaves usually fail
the muscle to contract during the expected mo- at the knot site in the repair region due to the knot
tion, in a contraction sequence that is in phase either slipping or pulling out through the tendon
with the recipient muscle.1,6 Wrist flexion is itself, whereas side-to-side techniques can fail via
288 Wilbur & Hammert

shearing through the fibers of the donor tendon, or before surgery to begin learning what will be
via failure at the outer suture site.19,20 involved with the postoperative therapy, such as
The tensile strength of the sutures used in a learning to focus on a specific muscle motion
tendon weave and its effect on tendon vascularity that will be used in the transfer (isolating flexion
have also been scrutinized. Tanaka and col- of flexor digitorum superficialis [FDS] to learn
leagues21 introduced a new corner-stitch construct how to activate for finger extension after transfer).
during the use of tendon weave fixation and The more complicated the transfer, the more time
compared this to the traditional central cross- and effort to learn to use the transfer. In addition,
suture design. This new design does not penetrate the rehabilitation process may vary between static
the full length of the tendon, and by avoiding full and dynamic transfers for specific conditions,
thickness stitches within the central substance of such as correction of clawing associated with
the tendon, theoretically poses less of a risk to ulnar nerve palsy.
the longitudinal intratendinous vasculature that is
important for the revascularization of tendon grafts REFERENCES
without sacrificing tensile strength. Despite the the-
ories behind preservation of intratendinous blood 1. Fitoussi F, Bachy M. Tendon lengthening and
supply by peripheral placement of sutures, Gelber- transfer. Orthop Traumatol Surg Res 2015;101(1):
man and colleagues22,23 showed no effect on the S149–57.
vascularity of the proximal tendon in a tendon graft 2. Solomons M. Disorders of the hand: tendon transfers.
with horizontal mattress suture repair. In: Trail IA, Fleming AN, editors. Disorders of the hand
Side-to-side tendon repairs have been advo- volume 2: hand reconstruction and nerve compres-
cated in children because of the smaller caliber sion. London: Springer; 2015. p. 33–55.
of tendons.20 Alternative coaptation methods in 3. Ingari JV, Green DP. Green’s operative hand surgery.
addition to side-to-side and Pulvertaft weaves In: Wolfe SW, Hotchkiss RN, Pederson WC, et al, ed-
include the lasso and the loop-tendon suture tech- itors. Green’s operative hand surgery. 6th edition.
nique.16,20 The lasso technique of Bidic and col- Philadelphia: Elsevier Inc; 2011. p. 1075–92.
leagues20 is useful for tendon transfers proximal 4. Omer GE. Tendon transfers for traumatic nerve in-
to the carpal tunnel in the volar forearm and prox- juries. J Am Soc Surg Hand 2004;4(3):214–26.
imal to zone 5 on the dorsal hand/forearm. It has 5. Ratner JA, Peljovich A, Kozin SH. Update on tendon
a maximal load to failure similar to that of the Pul- transfers for peripheral nerve injuries. J Hand Surg
vertaft weave, is stronger than the side-to-side Am 2010;35A(8):1371–81.
technique, and requires less tendon length and 6. Dowd T, Bluman EM. Tendon transfers—how do they
half the weave time of the Pulvertaft weave.20 work? Planning and implementation. Foot Ankle Clin
The loop-tendon suture technique described by N Am 2014;19:17–27.
Kim and colleagues16 had a higher ultimate tensile 7. Brand P. Biomechanics of tendon transfer. Orthop
load than end-weave suture techniques and is Clin North Am 1974;5:205–30.
beneficial in its ease of technique, ease of revision 8. Omer G. The technique and timing of tendon trans-
when readjusting tension due to removal of only fers. Orthop Clin North Am 1974;5:243–52.
one loop of tendon, ability to set independent ten- 9. Kozin SH. Tendon transfers for radial and median
sion of the finger cascade, as required in an FCR nerve palsies. J Hand Ther 2005;18:208–15.
to EDC tendon transfer, and the lack of damage 10. Omer GJ. Evaluation and reconstruction of the
to the donor or recipient tendon due to lack of lon- forearm and hand after traumatic peripheral
gitudinal slit incisions for weaving. Disadvantages nerve injuries. J Bone Joint Surg Am 1968;50A:
to all tendon transfer coaptation methods include 1454–78.
bulkiness of the tendon graft site that may increase 11. Brand P, Beach R, Thompson D. Relative tension
friction and smooth gliding of tendons and its abil- and potential excursion of muscles in the forearm
ity to be used in places other than the dorsal hand and hand. J Hand Surg Am 1981;6A:209–19.
or within the forearm.16 12. Peljovich A, Ratner JA, Marino J. Update of the
physiology and biomechanics of tendon transfer sur-
gery. J Hand Surg Am 2010;35A(8):1365–9.
Rehabilitation
13. Lieber RL, Murray WM, Clark DL, et al. Biomechan-
In general, rehabilitation following tendon transfers ical properties of the brachioradialis muscle: impli-
involves a period of immobilization followed by cations for surgical tendon transfer. J Hand Surg
therapy to learn how to use the transfer. Synergis- Am 2005;30A(2):273–82.
tic transfers are easier to learn and are the authors’ 14. Friden J, Lieber RL. Evidence for muscle attachment
preference when possible. There may be some at relatively long lengths in tendon transfer surgery.
benefit to having the patient see the therapists J Hand Surg Am 1998;23A:105–10.
 Principles of Tendon Transfer 289

15. Seiler JG, Desai MJ, Payne SH. Tendon transfers for 20. Bidic SM, Varshney A, Ruff MD, et al. Biomechanical
radial, median, and ulnar nerve palsy. J Am Acad comparison of lasso, Pulvertaft weave, and side-by-
Orthop Surg 2013;21(11):675–84. side tendon repairs. Plast Reconstr Surg 2009;
16. Kim SH, Chung MS, Baek GH, et al. A loop-tendon 124(2):567–71.
suture for tendon transfer or graft surgery. J Hand 21. Tanaka T, Zhao C, Ettema AM, et al. Tensile strength
Surg Am 2007;32A(3):367–72. of a new suture for fixation of tendon grafts when
17. Pulvertaft R. Tendon grafts for flexor tendon injuries using a weave technique. J Hand Surg Am 2006;
in the fingers and thumb: a study of technique and 31A(6):982–6.
results. J Bone Joint Surg Am 1956;38B:175–94. 22. Gelberman R, Khabie V, Cahill C. The revasculariza-
18. Gabuzda G, Lovallo JL, Nowak MD. Tensile strength tion of healing flexor tendons in the digital sheath.
of the end-weave flexor tendon repair. J Hand Surg A vascular injection study in dogs. J Bone Joint
Am 1994;19B(3):397–400. Surg Am 1991;73A:868–81.
19. Brown SHM, Hentzen ER, Kwan A, et al. Mechanical 23. Gelberman R, Chu C, Williams C, et al. Angiogen-
strength of the side-to-side versus Pulvertaft weave esis in healing autogenous flexor tendon grafts.
tendon repair. J Hand Surg Am 2010;35A(4):540–5. J Bone Joint Surg Am 1992;74A:1207–16.