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

138
Jin Bo Tang, MD, Jun Tan, MD, and Ren Guo Xie, MD
D
The Nantong Experience
OUTLINE
In this chapter, we review our experience and the
current methods of flexor tendon repair (in zone 2 in
particular) in our hand surgery service. We discuss
methods of tendon repair and postsurgical rehabilita­
tion, particularly procedures that may be critical to
optimal tendon function.
Our hand center is a part of the university hospital
system located in Nantong, serving about 7 million
people in the Yangtze River Delta area and neighboring
industrial and metropolitan cities, Suzhou and Shang­
hai. Injuries involving tendons that are seen at our clinic
vary from simple clean-cut injuries to major limb or
digital amputations. Three kinds of injuries—tendon
divisions in the digits and palm, tendon lacerations
in the wrist or forearm, and tendon injuries in digital
amputations—undergo primary, delayed primary, or
secondary repairs, respectively. Tendon transfer opera­
tions are performed for late functional reconstruction
following nerve injuries or neurological disorders. There
are 150 to 200 operations for these tendon injuries per­
formed annually, which constitutes 7% to 10% of the
total number of operations in our hand service. During
2011, the total number of operations for hospitalized
patients in our hand center was 2132 (day surgery cases
not included), of which tendon injuries in the digits and
palm, or dorsal hand, accounted for 68 cases (3.2%);
tendon injuries in the wrist, 44 cases (2.1%); and digital
(or limb) replantations, 84 cases (3.9%). Established as
an independent department with 100 beds devoted
solely to surgery of the hand and upper extremity, our
unit has the capacity to hospitalize patients for 10 days
to 2 weeks after tendon surgery. This system allows the
patients to be cared for and supervised within the hospi­
tal as they initiate early active digital motion exercises.
Our treatment of tendon injuries in wrist and distal
forearm and those associated with digital replantation
surgeries are discussed in other chapters. We presented
here our methods of flexor tendon repairs in the digits
and palm.
CLINICAL METHODS
Table 13(D)-1 summarizes the methods that we have
used over past decades, as well as the conceptual changes
underlying these methods. Our current methods are
illustrated in more detail next.
Anesthesia and Skin Incisions
We mostly use axillary regional block anesthesia when
performing flexor tendon repairs in the digits and palm.
General anesthesia is used in some patients with exten­
sive wounds or with injuries involving multiple sites. A
tourniquet is applied to the upper arm in all patients to
ensure a bloodless operation field. We prefer a Bruner’s
zigzag incision to expose the tendons (Figure 13[D]-1).
A midlateral incision is rarely used in primary or delayed
primary tendon repair. For a clean-cut injury, the inci­
sion usually does not exceed 3 cm longitudinally, a
length of a field just sufficient for surgical tendon repair.
We include the laceration in the skin as a part of this
incision in primary tendon repair. In delayed primary
repair, the skin wound is closed primarily and is healed
in 1 or 2 weeks, and we do not usually include the
original wound into the incision.
Lacerations through the skin and sheath are either
transverse or oblique and, in most cases, do not overlie
the site of tendon transection. The tendon ends can
mostly be brought into vision by passive flexion of the
digits. For zone 1 tendon injuries, the proximal flexor
digitorum profundus (FDP) tendon end does not
retracted far proximally and flexion of the metaphalan­
geal (MP) and proximal interphalangeal (PIP) joints
usually can bring the tendon end into the wound. In
some cases, a separate incision in the proximal part of
the digit or palm is necessary to find the retracted
tendon. For zone 2 tendon injuries, those in the proxi­
mal part of the fingers, in particular, often retract into
the palm (see Figure 13[D]-1). During surgery for
delayed primary repair, about half of our zone 2 repairs
require an additional incision to find the retracted prox­
imal tendon end; an incision at the distal palmar crease
usually reveals the tendon end (see Figure 13[D]-1).
Zone 2 Flexor Tendon Repairs
Incisions in the Sheath and Pulleys
By pulling the skin and subcutaneous flaps created by
the Bruner incision with retractors, the flexor sheath is
exposed. The major annular pulleys (the A1, A2, and A4
pulleys) are easily recognizable and are distinguished
from the remainder of the sheath by their thickness and

139
Chapter 13D:  The Nantong Experience
pulley. In other words, some part of the sheath is pre­
served between the A2 and A3 pulleys.
In many other cases, the tendons are cut at the loca­
tion of one, or more, of the strong annular pulleys (e.g.,
a little distal to the A2 pulley, through the A2 and/or A1
pulleys and in the vicinity of the A4 pulley). We cut the
entire A4 pulley and a major part of the A2 pulley
around tendon repair sites, while leaving the synovial
part of the sheath and the other pulley structures intact.
When the tendons are cut slightly distal to the A2 pulley,
within zone 2B, we open the sheath longitudinally for
1 cm distal to the A2 pulley and also open the distal
half of the A2 pulley (see Figure 13[D]-3). When repair­
ing the tendon at the distal edge, or in the distal part,
of the A2 pulley, we cut open a part of the sheath distal
to the A2 pulley together with the distal two-thirds of
the A2 pulley.
When repairing a tendon under the middle or proxi­
mal part of the A2 pulley, we cut open the proximal
two-thirds of the A2 pulley. Because the excursion of the
FDP tendon within this part of zone 2 is usually about
2 cm, the lengths of release of the A2 pulley and the
appearance as condensed, white fibrous structures. The
thin annular pulleys (the A3 and A5 pulleys) are not
seen clearly and, sometimes, appear indistinguishable
from the adjacent sheath. The cruciform pulleys are
usually not identifiable. However, the location of the A3
pulley has little variation and is almost always at the
level of the PIP joint. It is not necessary to identify the
A5 pulley or any of the cruciform pulleys.
The injured tendons are approached by incisions in
the sheath-pulleys at different sites according to the
level of the tendon laceration. When the level of tendon
laceration is judged to be in the vicinity of the PIP joint,
we incise the synovial sheath longitudinally, or even
excise a portion of it, to create a window in the sheath.
When the tendon is cut at the level of the A3 pulley, we
open the sheath between the A2 and the A3 pulleys. The
A3 pulley can be incised as well, but the incision should
not extend distally to the A4 pulley. A part of the syno­
vial sheath is kept intact distal to the PIP joint. When
the cut is between the PIP joint and the A4 pulley, we
frequently have to include the A3 pulley in the sheath
incision, but preserve the sheath proximal to the A3
Table 13(D)-1  Clinical Methods Used Over the Past Two Decades in Our Unit
1989–1994
The First Period
Tendon: repaired with two-strand modified Kessler, double, or six-strand looped suture methods.1,2
Sheath: closure, leave it open without repair, or interposing sheath graft reconstruction.2
New methods/concepts used: (1) Zone 2 subdivisions for recording sites of tendon lacerations and
analysis of treatment results; (2) stronger surgical repair methods.1-3
Investigations: (1) clinical results of double or multiple looped suture repairs;1 (2) analysis of clinical
outcomes in zone 2C;2,3 (3) results of sheath enlargement plasty.
Results: Zone 2C had the worst outcomes compared with other parts of zone 2, and 4% repair ruptures
occurred even with strong (six-strand) surgical repairs.1,3
1995–2002
The Second Period
Tendon: repaired with double or six-strand looped suture methods.
Sheath: closure, leave it open without repair, or interposing graft reconstruction.
New methods/concepts used: partial A2 pulley release or release of the entire A4 pulley.
Investigations: examination of finger movement after partial release of the A2 or A4 pulley.4
Results: release of a part of the A2 pulley or the A4 pulley, does not lead to clinically noticeable tendon
bowstringing, when other annular pulleys are intact.4
2003–2011
The Third Period
Tendon: repaired with a modified six-strand looped suture method (M-Tang repair).5
Sheath: leave it open without repair in most cases, but avoid opening sheath-pulley >2 cm.
Pulley: in cases when tendons are cut in or just distal to major pulleys, the A2 pulley is vented partially,
or the A4 pulley is vented entirely.5,6
New methods/concepts used: (1) a modified six-strand suture repair,5,6 (2) an early passive-active
motion regimen.6 and (3) definitions of length and limit of judicious pulley-sheath release.6
Subjects of interest: pulley-release may be a key to reducing resistance to tendon.
Loosely supervised early active motion may be feasible and safe clinically.
Investigations: examination of the cases treated with pulley-venting, strong core suture, and the new
passive-active motion regimen.
Results: Combination of stronger surgical repair, proper treatment (release) of the major pulley, and a
combined passive-active motion regimen favors near-ideal recovery.
Proper release of the constrictive part of the pulleys probably is most critical.

140
Section 2:  Primary Flexor Tendon Surgery
ends are temporarily fixed using a needle proximal to
the laceration site, but the distal tendon ends do not
usually need temporary fixation because passive flexion
of the DIP joint during surgery by an assistant easily
brings the ends into the center of the operation field.
If the tendon ends are untidy, the tendon ends are
trimmed using a pair of scissors or a scalpel. In delayed
primary repairs, the tendon ends are usually covered by
granulation tissues or collagen clots. These are removed
to refresh the tendon ends. Then, the core sutures are
placed into the tendon when the two tendon ends pulled
together tightly, with the distal joints of the finger held
in flexion by an assistant. Looped sutures carried on a
single needle are used for the core suture. In repairing
the FDP tendons, we currently use a modified six-strand
repair, performed by passing two looped 4-0 sutures
through the tendons. A U-shaped repair is created first
to connect the tendon ends using one looped suture,
and the second looped suture is then placed at the
center of the tendon, making a six-strand M-shaped
core suture repair. For the flexor digitorum superficialis
adjacent sheath referred to above are, in most cases, suf­
ficient to free the tendon repair from restriction by the
pulley, or catching on rims of the remaining parts of the
pulleys, during movement of the finger joints through
a full range of motion.
Tendons lacerated proximal to the A2 pulley are the
least difficult to access. They can be approached through
an incision in, or excision of, the sheath proximal to the
A1 pulley, or an incision over the A1 pulley. Extending
the skin incision proximally may help expose the
retracted tendon end.
Tendon Repair Techniques
After making the incisions in (or excisions of) the sheath
and pulleys, the tendon lacerations are exposed. The
distal tendon ends are brought into the operation field
by flexion of the distal interphalangeal (DIP) joint and
the proximal tendon ends are brought into the field
by, either flexion of the metacarpophalangeal and PIP
joints, or through a separate proximal incision (see
Figure 13[D]-1), as illustrated. The proximal tendon
Figure 13(D)-1  Skin incisions for primary or delayed primary repairs: A, skin incisions used in the hand to exposed the
injured tendons; B, Bruner’s zigzag incision is used to expose the wound in delayed primary repair; C, additional small
transverse incision proximal to the tendon cut level may be required if the proximal end of the tendon retracted proximally.
The retracted tendon end is lead through the synovial sheath tunnel by insertion of a silicone tube (or a catheter) to
approximate the distal tendon stump.
B
C
A

141
Chapter 13D:  The Nantong Experience
We complement the core repair with simple periph­
eral suturing, with a simple running peripheral suture
or a few interrupted sutures most often being used.
Peripheral sutures serve to smooth the junction site of
the two tendon ends. We are able to put six to eight
loops of the running stitch in tendons of larger diameter
using 6-0 monofilament nylon. However, adding even
the simplest peripheral suture to the dorsal aspect of the
tendon is difficult in some cases. Instead, we sometimes
add a few interrupted sutures on the palmar and dorso­
lateral aspects of the tendon. Nevertheless it also seems
reasonable not to supplement the repair with peripheral
sutures when a strong six-strand core suture is used.
Although we have no experience with “core-suture only”
(FDS) tendons, and some parts of the FDP tendon (such
as in zone 1 or in the little fingers), we use only a four-
strand U-shaped repair using one looped suture.
Figure 13(D)-2 summarizes the methods we use in
repairing tendons in zones 2 to 5. A two-strand Kessler
repair is used only sometimes in repairing tendons cut
in zone 5. We currently do not use two-strand repairs
in zone 2 or 3. Figure 13(D)-3 details our method of
making a six-strand M-shaped repair. The method of
placing three Tsuge looped sutures is illustrated in Figure
13(D)-4. This technique was simplified to the M-shaped
repair (see Figure 13[D]-3) that we currently use.
Figure 13(D)-2  Summary of repair methods we use
in repairing flexor tendons: A, modified Kessler repair;
B, cruciate repair; C, U-shaped repair; D, cross-lock repair
(a four-strand repair made with one needle carrying two
separate double suture strands); E, Tang technique of three
Tsuge suture repairs; F, M-Tang repair; the two-strand
Kessler repair is now used in repairing tendons in zone 5. In
repairing the tendons, in zones 1 to 4, four- or six-strand
repairs are used. C, D, Two four-strand repairs developed in
our unit.
A
B
C
D
F
E
Figure 13(D)-3  Illustrations of how to make a six-strand
core suture using a looped suture repair (M-Tang method),
which forms an M-shaped configuration within the tendon.
A
B
C
D
E

142
Section 2:  Primary Flexor Tendon Surgery
tendon repairs, this simplifies the surgery and does not
appear to decrease tendon strength.
In performing core sutures, we are careful to (1) use
multistrand (four- or six-strand), rather than conven­
tional two-strand, repairs; (2) ensure the length of core
suture purchase is greater than 0.7 cm (0.7 to 1.0 cm in
most cases); and (3) add some tension to the repair site
by tightening the core suture, producing a roughly 10%
shortening of the tendon segment encompassed by the
core suture (Box 13[D]-1). We avoid making a loose, or
tension-free, tendon repair. We no longer use conven­
tional two-strand repairs in repairing tendons in the
fingers or palms.
We believe that the above three points are crucial to
the strength of a surgical flexor tendon repair. These
three measures may be particularly necessary when one
considers that the tendon ends usually become softened
after injuries and tensioned during active tendon
motion, before biological healing is accomplished.
Increasing the number of strands across the repair site
Figure 13(D)-4  Two operative views showing the methods
of placing three looped sutures into the tendon (the Tang
method). The pictures show a case repaired by the authors
in 1989. This repair method was replaced by the M-shaped
repair in the early 2000s.
A
B
is the easiest and most effective way to provide the
tendon repair with greater strength. Maintaining suffi­
cient length of the core suture purchase ensures that the
suture is anchored reliably in sufficient amount of the
tendon substance that will not soften during the early
period after repair. Otherwise, the core suture may slip
or be pulled out, particularly when the tendon ends
soften after repair.
Adding slight tension to the repair site can greatly
decrease the chances of gapping of the tendon repair.
The tendon repair site may look a little bulky when
proximal pull of the muscles is eased by temporary
needle fixation of the proximal tendon during surgery.
However, this is not the case after release of the tempo­
rary fixation. Adding such a baseline tension to the
repair sites produces appropriate tension after surgery,
as this counteracts the tension of the muscles proxi­
mally during active motion. If tension has not been
added to the repair site, the repair may overstretch and
gap after surgery during active digital mobilization. We
consider that adding a little tension is as important as
adding a peripheral suture to resisting gaps between
tendon ends. With appropriate pretension of the repair
site, the chance of gapping, or the size of the gap, can
be decreased, thereby lowering the risk of catching of
the repair sites on any edge of the sheath.
Treatment of the FDS Tendon
The FDS tendon bifurcates under the middle part of the
A2 pulley and only injury in the proximal part of the
finger can produce a complete FDS tendon laceration.
In Approaching the Tendon:
1. Small sheath incision and preservation of majority of the
sheath
2. A sheath incision less than 2 cm is allowed, which may
include a part of the A2 pulley (up to two-thirds) or
entire A4 pulley.
3. At least a portion of the A2 pulley should be maintained,
and loss of more than two annular pulleys should be
particularly avoided.
4. Additional small incisions in the sheath may be neces­
sary to find the retracted tendon ends. Avoid a lengthy
incision of the sheath, which will damage the pulley
function.
In Suturing the Tendon:
1. Use strong repairs: a four- or six-strand core suture
repair, or a strong suture material.
2. Ensure sufficient length of core suture—about 0.7 to
1.0 cm in either tendon stump.
3. Add some tension to the repair site by tightening the
core suture.
4. Only simple (or no) peripheral sutures are needed when
a strong core suture is used.
Box 13(D)-1
Critical Technical Points

143
Chapter 13D:  The Nantong Experience
The FDS tendon does not always need repair. A partial
laceration of the FDS tendon does not need repair at all.
Complete laceration of the FDS tendon does not require
surgical repair if the FDP tendon is intact. For complete
FDS and FDP tendon lacerations, least problematic is
repair of the FDS tendon cut proximal to the bifurca­
tion. Injuries to this part of the FDS tendon can be
treated almost identically to those of the FDP tendon,
except that the FDS tendon is flatter and does not
accommodate more than four suture strands. The bifur­
cating part of the FDS tendon (the segment within zone
2C) is much more difficult to treat. We use two sepa­
rated Tsuge sutures (a two-strand repair in each FDS
slip) to repair FDS injuries in zone 2C. In repairing the
FDS tendon within zone 2B, we use a variety of tech­
niques, including repair with a tendon-to-bone junction
(as for reattachment of the FDP tendon to the distal
phalanx) if the residual distal stump is very short, or
repair with a two-strand core suture for each slip if the
distal stump is long enough. When one slip is com­
pletely cut, but the other is uninjured, repair may not
be necessary. Quite often, we find that one, or both, slips
of the bifurcated FDS tendon are partially severed. In
this situation, we make a judgment according to the
length of the distal stump and the extent of the tendon
division. The partial cut is left untreated, or one or two
stitches are placed, to prevent triggering of the FDS
tendon during tendon movement.
Repair of the FDS tendon together with the FDP
tendon may favor gain of the strength of finger flexion
and prevent hyperextension of the PIP joint. Neverthe­
less, observations indicate that these benefits may not
be substantial. Repair of only the FDP tendons in our
patients have not led to a significant loss of power of
finger flexion, and daily use of the hand has not been
affected after repair of only the FDP tendon. Develop­
ment of hyperextension of the finger joints usually is
not notable. During surgery, after FDS tendon repairs,
we noted that the repaired FDS tendon is sometimes
entrapped by the narrow pulleys and the repair decreases
motion of the FDP tendon. Therefore, in primary repairs,
we only repair the completely lacerated FDS tendon
when such repairs appear not to decrease or impair FDP
tendon gliding. Repair of both FDS and FDP tendons is
more difficult when the surgery is delayed by 1, or more,
weeks after injury. When the cut is at, or just distal to, the
A2 pulley (zones 2C and 2B), we find it almost impos­
sible to repair both tendons after such a delay. The FDS
tendon retracts far proximally, and it is hard to pass both
tendons under the A2 pulley, or even a residual part of
this pulley. We usually leave the FDS tendon unrepaired,
or excise it locally, in delayed primary tendon repair.
Figures 13(D)-5 and 13(D)-6 illustrate two cases of
delayed primary repair of the FDP tendon with the six-
strand repair method. The partially injured FDS tendon
was left unrepaired in one case (see Figure 13[D]-5)
and locally excised to accommodate tendon movement
in the other (see Figure 13[D]-6), which was an end-
to-end repair of the FDP tendon 3 weeks after rupture
of the primary repair of the tendon. In both cases, the
A2 pulley was either vented or partially excised.
Treatment of the FDP and FDS Tendons
in the Palm (Zone 3)
Injuries in zone 3 involve the flexor tendons to several
fingers in most cases, and are often complicated by
nerve and vascular injuries. However, the treatment of
the flexor tendons in the palm is not as difficult as in
zone 2, because no sheath or pulleys are present in this
region and the proximal tendon ends are not retracted
too proximally. In making core and epitendinous sutures
in the FDS and FDP tendons, we follow the principles
and methods that are described for zone 2 injuries.
Repair of both the FDP and FDS tendons is usually
possible and less difficult than in zone 2. We use either
a four- or a six-strand core suture, using looped sutures
together with a simple running peripheral suture. Both
the superficialis and profundus tendons are repaired in
almost all cases, when no soft tissue loss is present.
Nevertheless, it should be particularly noted that com­
pound injuries with nerve and vascular injuries, or soft
tissue loss are quite frequent, and repair of the nerves
and, in some cases, vascular anastomoses, are integral
to these surgeries.
Treatment of the FDP Tendon in Zone 1
In our clinic, we see patients with an injury to the FDP
tendon in the distal part of the digits (zone 1 injury)
much less frequently than injuries in zones 2 and 3.
Treatment of FDP tendon lacerated about 1 cm proxi­
mal to the insertion is the same as for the FDP tendon
in zones 2 or 3. For the tendon laceration in which end-
to-end repair is not possible, we advance the tendon to
anchor it to the distal phalanx with mini-anchors
(Figure 13[D]-7) or reinforced suture repairs (such as
placement of two cross-stitch repairs, one on each of the
two FDP tendon bundles, using Fiberwire, or, rarely,
passing sutures through a transverse drill hole through
the distal phalanx).
The conventional pull-out repair method was used in
our clinic for many years and had problems of trauma­
tizing the nail in a significant proportion of the cases.
The pull-out repair method is arguably not an ideal
method. Reattachment of the FDP tendon to the phalanx
using mini-anchors or to the residual distal stump with
reinforcement with two cross-stitches to create a strong
tendon junction appears as efficacious as the pull-out
repair. Because the distal end of the FDP tendon does
not glide, reinforcing suture repairs that might lead to
tendon bulkiness elsewhere cause no problem of tendon
function here. With currently available repair materials
(such as Fiberwire), we can easily achieve a strong repair

144
Section 2:  Primary Flexor Tendon Surgery
that allows protected early active finger motion after
repairing the distal FDP tendon. The opened sheath is
not repaired in zone 1.
Treatment of the FPL Tendon
A number of facts should be considered in flexor pollicis
longus (FPL) tendon repairs within the thumb: (1) only
one flexor tendon is present; (2) the FPL tendon is
among the largest in diameter in the digits; (3) the
pulleys in the thumb are not as broad, narrow, and rigid
as in the fingers; and (4) the flexor sheath is relatively
short. Surgical repairs of this large flexor tendon are
perhaps the easiest. Nevertheless, its gliding path in the
hand is unique, and flexion force transmitted along this
Figure 13(D)-5  A case of delayed primary repair in zone 2C. A–D, The partially cut FDS tendon was left unrepaired. The FDP
tendon was repaired with the M-Tang method. The distal two-thirds of the A2 pulley was vented. E and F, Postoperative
digital extension and flexion.
A
B
C
D
E
F

145
Chapter 13D:  The Nantong Experience
access the tendon. Keeping two pulleys amply maintains
tendon function.
It is common to find the proximal tendon end
retracted, under the thenar muscles or in the carpal
tunnel, particularly with delayed repair. Because there is
no sheath over the FPL tendon in the thenar muscles,
we usually attempt to find the retracted FPL end without
incising the muscles, but by retracting the thenar muscles
tendon is perhaps the highest, subjecting the repaired
FPL tendon to greater danger of rupture when moved
actively. Consequently, a strong surgical repair is par­
ticularly necessary. This tendon can easily accommodate
a six-strand repair, which we use regularly to repair it,
even in children (Figure 13[D]-8).
We preserve at least two pulleys in the thumb, and
most often we have to incise one annular pulley to
Figure 13(D)-6  A to D, A case of repair of a ruptured tendon 3 weeks after rupture of the primary repair of this tendon. The
FDS tendon was excised locally during the repair and the A2 pulley was shortened. E and F, Postoperative digital extension
and flexion.
A
B
C
D
E
F

146
Section 2:  Primary Flexor Tendon Surgery
Figure 13(D)-7  A mini-anchor was used to anchor the
tendon to the distal phalanx in distal zone 1. This method
has recently replaced the pull-out suture method in our
practice.
Figure 13(D)-8  Repair of a lacerated FPL tendon in a 9-year-old child with a six-strand M-shaped repair.
A
C
B
D
to expose the FPL tendon underneath. When the FPL
tendon retracts more proximally, sometimes a small
incision is made just distal to the transverse carpal liga­
ment to locate the FPL tendon.
Flexor Tendon Repairs in Children
Flexor tendon injury in children (i.e., patients under 12
years of age) is less common than in adults. Among our
cases, flexor tendon injuries in children account for
about 15% of the acute tendon injuries treated. Func­
tional recovery of children with flexor tendon injuries is
usually satisfactory. In approaching the tendons, we take
particular care not to violate the annular pulleys, which
are usually not as rigid as in adults, and pulley venting
is usually not necessary. However, we do not usually
repair FDS tendons in children, as such repairs (particu­
larly in the bifurcated part of the FDS) are difficult, or
even impossible, in many cases. The FDP tendon is
repaired with either a four-strand repair using one
looped suture to make a U-shape repair, or two looped
sutures to make a double Tsuge repair. In patients over

147
Chapter 13D:  The Nantong Experience
to disrupt; (4) we start motion exercise 3 days (or as late
as 5 days in cases with remarkable edema in the hand
or with soft tissue repair as well) after surgery, and we
perform only four or five sessions of exercise each day,
rather than require patients to move hourly; and (5) we
emphasize full extension of the fingers in the early
weeks after surgery when tendon healing is still weak,
and then shift our emphasis to active flexion in later
weeks, when tendon healing is stronger.
At postoperative 212 weeks, we change the protective
position of the hand and shift the emphasis to active
flexion. We encourage partial active digital flexion first,
and then proceed to full active flexion of the fingers
when no resistance is perceptible. Regarding the timing
of change of the hand position, we consider that tendon
healing strength has only started to increase at that time
and that immobilization of a joint for 212 weeks would
not cause contracture of soft tissues of the joint. Never­
theless, we also believe that such a change in protective
hand position may be made any time between 212 weeks
and 3 weeks.
The details of our method is as follows: the hand
is protected in a dorsal thermoplastic splint, with the
wrist in slight flexion (20° to 30°), the MP joints in
slight flexion, and the interphalangeal joints in exten­
sion (or minimal flexion), for the first 212 weeks (Figure
13[D]-9). We do not encourage patients to move the
7 or 8 years old, we sometimes use a six-strand repair;
4-0 looped nylon suture is used in all cases.
We perform primary repair of the tendon in children
whenever possible, or try to carry out the repairs within
2 or 3 days of injury, without a long delay, because their
wounds heal quickly. After a long delay, the tendon ends
may retract quite far proximally, due to inability to
prevent contraction of the flexors. We treat flexor tendon
injuries in young patients older than 12 years identically
to those in adults.
POSTOPERATIVE REHABILITATION
We currently adopt a combined active and passive finger
flexion regimen in postoperative care, except in (1) the
patients younger than 12 years old; (2) the patients with
flexor division(s) associated with fractures; and (3) in
rare instances, in which the patients are judged to be
unable to cooperate.
A few changes can be noted between our current
practice and those described by other surgeons: (1) we
do not use rubber band traction because we believe that
it contributes to extension deficits of the fingers; (2) we
emphasize that active motion should proceed after mul­
tiple cycles of passive motion in each session of exercise,
to reduce the resistance to the active movement; (3) we
do not encourage full active flexion in the early weeks
after surgery, as tendons with marked flexion are easiest
Figure 13(D)-9  The protective position of the hand and methods of rehabilitation in the first 2 12 weeks after surgery. Full
finger extension is emphasized, but only partial active finger flexion is encouraged. Inclusion of thumb in the splint is not a
necessity, but it prevents unintentional pinch action of the hand.
First 2.5 weeks
Exercise of full
finger extension
Wrist flexion
Full passive flexion
Active finger flexion-limited range
Active finger flexion-limited range

148
Section 2:  Primary Flexor Tendon Surgery
finger during the first few postoperative days, because
the hand is painful, edema is more prominent and,
more importantly, adhesions have not yet developed.
Decreasing the number of days of motion decreases the
chance of repair rupture.
The First Period (Surgery to
2 12 Weeks Postsurgery)
Exercise starts at 3 to 5 days after surgery. Before each
episode of active digital flexion, the finger is passively
flexed 10, or more, times to lessen the overall resistance
of the finger joints and soft tissues—a “warming up”
process—after which active flexion should encounter
lower resistance. The patient is then instructed to flex the
fingers actively with gentle force 20 to 30 times during
each morning, noon, evening and before sleep session,
up to the range with which the patients feels comfort­
able. The motion range is usually from full extension to
one-third, or half, of the full flexion range, although this
may even increase to two-thirds of the full range if this
can be achieved with ease. Active flexion over the full
range is not encouraged, unless it can be achieved very
easily. Patients may increase the number of motion epi­
sodes up to five or six per day, but we do not require
patients to move hourly. In this first period, full active
extension is particularly encouraged and the fingers are
passively stretched against the splint if full extension is
not achieved. Prevention of extension deficits rather
than full active flexion is a major goal during this period.
The Second Period (2 12 to
5 Weeks Postsurgery)
At 212 weeks, a new thermoplastic splint is made, with
the wrist splinted at 30 degrees of extension (Figure
13[D]-10). Finger flexion, both passive and active,
is emphasized during this period. The patients are
instructed to actively flex after a passive warm-up, as
earlier. Active flexion up to the mid-range is required as
a minimum and is encouraged further, up to two-thirds,
or the full, range of flexion, depending on the patient’s
ability to perform resistance-free motion. Digital flexion
from the mid-range to the full range, in particular over
the final one-third of the flexion range, is usually carried
out passively if the fingers encounter resistance. Our
studies show that finger flexion over the final one-third
of the full range of motion range encounters 5 to 10
times more resistance than in the previous two-thirds of
the range of motion. Thus, ruptures are much more
likely in this final part of the flexion range. Ensuring full
passive flexion, to prevent dorsal ligament tightening
and extensor tethering, and encouraging finger flexion
actively, while avoiding flexing the finger forcefully over
the final flexion range, are guidelines during this second
period.
If both FDS and FDP tendons are repaired at the
same level, differential FDS and FDP motion exercise is
Figure 13(D)-10  The protective position of the hand and
methods of rehabilitation in the second 2 12 weeks after
surgery. Full active digital flexion is encouraged and
emphasized.
Second 2.5 weeks
encouraged through the first 5 weeks, with separated
active flexion of the two interphalangeal joints.
After 5 weeks, full active finger flexion is encouraged.
This can be started earlier if flexion in the final part of
the flexion range is judged to have less resistance. After
5 to 6 weeks, the splint is discarded or used only at
night. Patients can return to normal use of the finger
from 8 weeks.
The mechanical basis of the protocol design is synergy
between wrist and finger actions: with the wrist flexed,
full finger extension is achieved with less tension on the
flexor tendons, while full finger flexion can be achieved
with less tension on the repairs with the wrist extended.7,8
Active finger flexion to full flexion encounters much less
resistance when the wrist is extended than when the
wrist is flexed. This, effectively, avoids overload of the
repaired tendons. However, we do not encourage
maximal active flexion of the finger when the exercise
meets remarkable resistance, but instead suggest incor­
porating active finger flexion up to the mid-range and
passive motion from mid- to maximal flexion into indi­
vidual motion cycles.
Our experience indicates that repair rupture can be
minimized using this exercise regimen, which is aimed
at avoiding placing high levels of tension on repairs

149
Chapter 13D:  The Nantong Experience
of the A2 pulley. The zone 2C tendon repairs were asso­
ciated with the worst recovery of function.
In a series of later cases, repaired with a six-strand
repair by experienced surgeons with venting of either a
part of the A2 or the entire A4 pulley when necessary,
91% of repairs in 36 fingers were given a Strickland
rating of good or excellent,5 and the results of zone 2C
repairs were no different from those in other parts of
zone 2. No tendon ruptures were noted. Since 2005, we
have used a modified six-strand repair (referred to as the
“M-Tang” method in our department because of its
M-shaped configuration inside the tendon) using looped
suture to repair the FDP tendon.6 The restricting part of
any annular pulley was released through either a midline
incision or partial excision, when necessary. For clean-
cut tendon injuries in the thumb, index, middle, and
ring finger, or those wounds that could be converted to
clean-cut wounds, the current authors—all experienced
in primary tendon surgery—have achieved good or
excellent results in almost all cases. We had less satisfac­
tory recovery of function (graded as “fair”) in some
tendon repairs in the little finger, and in fingers with
more extensive loss of soft tissues and untidy tendon
lacerations (occurring in the palm area). When injuries
in the palm involve tendon injuries to three or more
digits, not all digits are likely to achieve good or excel­
lent recovery. Good or excellent recovery of function was
achieved in 84% of primary and delayed primary repairs
in the thumb, fingers, and palm, including those with
borderline indications. We note that the expertise level
of the operator exerts a significant influence on the out­
comes. When treated primarily by junior surgeons (resi­
dents) who fail to consult with the senior attending
surgeons, the outcomes are hardly able to reach “excel­
lent.” However, when treated by experienced surgeons,
adhering to established principles of treatment, and
given a clean-cut wound (or a wound that can be con­
verted into a clean-cut wound after débridement and
trimming), good or excellent recovery is fairly predict­
able, and almost all the cases recover good or excellent
function, except in some little fingers. No repair rupture
was found in this period. Careful examination also
revealed no perceivable tendon bowstringing after the
venting or shortening of the A2 or the A4 pulley (Figure
13[D]-11).
Factors that worsen the results roughly by “one grade”
include (1) tendon repairs in a finger with a fracture;
(2) tendon repairs in the little finger; (3) crush injuries
in fingers or the palm causing extensive tissue damage;
and (4) tendon repairs by less experienced surgeons.
DISCUSSION
Mastery of anatomy and careful dissection. Repair of
flexor tendon injuries is a perfect test of the technical
and intellectual competence of a hand surgeon. To
achieve satisfactory functional recovery is a challenge for
while achieving sufficient active motion, together with
release of the sheath and pulleys and an increase in
surgical repair strength.
For children, we immobilize both the wrist and the
MP joints in slight flexion after surgery for 3 weeks, and
a protective splint to immobilize the patient for an addi­
tional 1 week or 10 days is usually prescribed. Active
flexion exercise of the fingers starts in the fourth week.
For an incomplete tendon cut and repair, we immo­
bilize the hand in a protective position for 2 weeks, and
early active finger motion is initiated immediately after
surgery. Patients return to normal use of the hand 4 to
5 weeks after surgery.
For FPL tendon repair, we immobilize the thumb
with the MP joint flexed at about 45°, the interphalan­
geal joint slightly flexed, and the wrist in a resting posi­
tion (extension of about 20°). Active thumb flexion
starts 2 or 3 days after surgery. Full active flexion is
encouraged if no noticeable resistance is encountered.
The hand is brought out of the splint to assume wrist
flexion when active extension is ordered to ease the
tension. The splint is removed 5 weeks after surgery, and
normal thumb motion is initiated in week 6 or 7.
For tendon repairs with fixation of a fracture, we
prescribe early motion exercise of the fingers only when
the fixation is stable, and motion is initiated 10 days to
2 weeks after surgery. The protective splint is applied for
4 or 5 weeks. With screw fixation of bones, or placement
of mini-plates on the lateral aspect of the shafts of pha­
langes or metacarpus, early active finger flexion is pos­
sible after edema subsides and preliminary fracture
healing is achieved. In cases of simultaneous repair of
the vascular structures, we delay motion exercise for 7
to 10 days after surgery. In cases with tension-free nerve
repairs, we do not usually modify our motion regimen
or timing of exercise initiation.
OUTCOMES
Our results have been separately presented in several
reports.1-5 Of particular note, our first series of cases with
use of four- or six-strand repairs had a rupture rate of
4%. In 1994, we reported the results of using double
or multiple looped suture tendon repairs in 51 fingers
(46 patients).1 By the White criteria, 76.5% of cases had
the good or excellent results, with 4% of the finger
repairs rupturing. In an earlier report, we analyzed the
results of a series of 72 tendon injuries in 54 fingers in
43 patients by subdivisions. Tendon injuries in zone 2C
(16 fingers), which constituted one-third of the injuries
in zone 2, had only 69% good or excellent results by
Strickland criteria, compared with 84% good or excel­
lent results in the other three subzones of zone 2.2 The
tendons in this series of cases were repaired with one of
the four methods: two-strand modified Kessler, two-
strand Tsuge, four-strand double Tsuge, or six-strand
triple loop-suture method, without purposeful release

150
Section 2:  Primary Flexor Tendon Surgery
six-strand core suture repairs provide strong surgical
repair. Although a number of factors affect the strength
of surgical repairs, not all exert a clinically significant
influence. Among the factors that most notably influ­
ence strength are (1) adequate strength of the suture
materials, (2) increased number of repair strands, or
alternatively larger caliber sutures, and (3) secure anchor
of sutures to the tendon. Increasing the number of
stands across the repair site is the most straightforward
way to achieve stronger and safer repairs. Maintaining a
sufficient number of sutures passing through the repair
sites, ensuring adequate purchase length of all sutures,
and secure tendon-suture junction sites ensure a strong
repair. We recommend a strong surgical repair (four-
strand or beyond) be always used. Suture configuration
within the tendon does not alter the strength very drasti­
cally, given an equal number of suture strands passing
across the repair site. For this reason, given the available
four-, six-, or even eight-strand repairs described, sur­
geons actually have ample freedom to choose any one
when implementing a strong repair.
Currently, we use four- or six-strand repairs in repair­
ing the tendon, and believe that an eight-strand repair
is probably not necessary, although it is a valid option.
The FPL tendon and the FDP tendon in the adult index,
middle, and ring fingers can easily accommodate four-
or six-strands repairs. The FDP tendon in the little finger
and the FDS tendon in all fingers may easily accom­
modate a four-strand repair. All repairs using four-strand
(or greater) core sutures provide sufficient strength for
early active digital motion.
Release of the pulleys as a means to reduce “inter­
nal” resistance to tendon motion. Free tendon gliding
without repair rupture and restrictive adhesion remains
the ultimate goal of flexor tendon repair. Such free
even an experienced hand surgeon. Detailed anatomical
knowledge and meticulous dissection and repairs are
essential to both the success of the surgery and later
functional recovery. Optimal surgical outcomes (rather
than just completed surgeries—“success of surgery”)
require us to master the anatomy of the flexor tendon
system with great precision. When we begin a flexor
tendon repair, we should have a clear mental picture of
the location of each of major pulleys and its length and
diameter, and which parts of the sheath and pulley are
the narrowest. Knowing the locations of the A2 and A4
pulleys is only the minimal requirement.
Most surgeons do know the nomenclature, number,
and rough distribution of the pulleys in the fingers, but
frequently do not master the anatomical details of these
pulleys to the extent of knowing their lengths and exact
locations. These details are imperative to appropriate
exposure of the tendon, release of the resistance to
tendon motion after repair and prevention of tendon
bowstringing. The two most essential anatomical points
are (1) the A2 pulley, of length about 1.5 to 1.7 cm in
the adult middle finger, spans the proximal two-thirds
of the proximal phalanx and the middle and distal part
of this pulley is the narrowest; and (2) the A4 pulley is
much shorter, of a length about 0.5 to 0.7 cm, and is
located in the middle part of the middle phalanx. It is
also narrow. We advise that surgeons, junior surgeons
in particular, take time to review relevant diagrams, or
textbooks, before surgery. During surgery, the A1, A2,
and A4 pulleys appear as easily recognizable white
dense bands.
Stronger surgical repair is always recommended. A
great number of repair techniques are available cur­
rently. Two-strand core suture repairs (using either 4-0
or 3-0 sutures) provide weak tendon repairs. Four- or
Figure 13(D)-11  Examination to confirm the absence of perceivable tendon bowstringing during active finger flexion (A and
B) and resisted finger flexion (C) in a patient following venting of the A2 pulley up to two-thirds of its entire length.
A
B
C

151
Chapter 13D:  The Nantong Experience
distinct from those in the other fingers, the anatomical
features and biomechanics of the tendon system in the
little finger merit more thorough study, and different
approaches to treatment may be necessary. The anatomy
should be further investigated and treatment be further
documented, to provide repairs specifically designed for
the little finger.
Not repairing the FDS tendon facilitates return of
function in cases of delay or cases of unfavorable
conditions. Repairing both the FDS and FDP tendons
under favorable wound conditions may produce the
most ideal results—in terms of digital motion, strength,
and prevention of PIP joint hyperextension. However,
repairing all structures is not always possible and may
not be a necessity for ordinary use of the hand. In most
delayed repair cases, or with wounds that include some
degree of soft tissue loss, we suggest repairing the FDP
tendons alone. For injuries to both the FDS and FDP
tendons in the A2 pulley area, we recommend repair of
only one slip of the FDS, or removal of both slips of the
FDS tendon.
Repair of only the FDP tendon usually produces ade­
quate function for patients, restoring normal daily use
of the hand. A more complicated surgery sometimes
risks finger motion; adhesions formed between the
tendons and pulleys can render the tendons totally
immovable under unfavorable wound conditions.9,10
Loosely supervised hand rehabilitation—a future
direction? This is a topic we would like to raise, but,
thus far, have insufficient clinical data to recommend it
for widespread use. In recent years, in a number of cases,
we were confident that we had reduced resistance to
tendon gliding by venting critical parts of the pulleys,
and early active motion after surgery was only loosely
supervised. We prevented the patients from active finger
flexion against external resistance but did not instruct
patients to follow the number of exercise sessions and
sequence of passive-active motion in each session.
Rather, the patients were allowed rather large or nearly
full range of active finger flexion with greater freedom
if they could move their fingers actively with ease. These
patients experienced no repair rupture and returned to
normal function.
We cannot yet generally recommend this “loosely
supervised regimen,” but we do bring it to the attention
of surgeons and therapists with regard to an eventual
goal of simplifying rehabilitation regimens and allow­
ing more freely active motion in selected individuals in
whom surgeons have confidence. We urge surgeons and
therapists to expand our experience and explore the
usefulness of simplified motion regimens. Of course,
such a loosely supervised exercise program can only
be safely established when relatively resistance-free
tendon motion is ensured internally, no resistance to
finger flexion is applied externally, and the tendons have
undergone a strong surgical repair.
tendon motion may be achievable through implemen­
tation of strong surgical repair, and decrease of tension
over the repaired tendon by decreasing the resistance to
tendon motion. Release of the narrow, constricting
pulleys that may present a “constricting band” during
tendon motion may be an effective way to release resis­
tance to tendon motion internally within the finger. A
pulley release takes only a few minutes, and multistrand
surgical repairs add only 10 to 15 minutes to the total
time of surgery, as compared to a conventional surgical
repair. Proper release of the pulleys and use of stronger
surgical repairs are much easier and more economical
than achieving increase in delicacy and expertise of
hand therapy. A pulley-release procedure or a multi­
strand repair cost less than 20 minutes of surgery, but
elaborate rehabilitation costs much greater efforts over
weeks.
We suggest consideration of such simple procedure
as releasing the critical parts of the pulley to decrease
the resistance to tendon gliding internally. We believe
that pulley release procedures and increasing repair
strength are two effective measures for achieving safe
postoperative active tendon motion, and releasing the
critical part of pulleys eventually allows tendon motion
free from major internal resistance during active tendon
motion.
Tendon repairs in the little fingers are worth par­
ticular attention. The little finger is shorter and smaller
than the other digits. Its tendons and sheath have a
much smaller diameter than those of the other fingers.
We have noted poorer outcomes of flexor tendon repairs
in the little finger. Not surprisingly, injuries in the little
finger are more difficult to treat, given that (1) the
cross-sectional area of the tendon is only 60% to 70%
that of the other fingers, (2) the diameters of the pulleys
are quite small, and (3) the tendons are bent in multi­
ple directions during finger flexion, gliding over paths
not only bent palmarly, but also bent medially. The
tendons in the little finger are subject to major bending
forces in at least two directions; the tendons in other
fingers are also subject to bending in two directions, but
not as extensively as the little finger. An additional
difficulty in carrying out repairs in the little finger is
the narrow space available in which to perform the
surgery. The distance between each pair of annular
pulleys is much shorter, and accessing the lacerated
tendon through an incision in the sheath is much more
demanding technically and requires precise dissection.
Openings of the sheath are frequently insufficient to
allow surgery, requiring extension of these windows in
the sheath.
In the little finger, we have adopted repair methods
similar to those used for other fingers thus far, but these
methods are often more difficult to use and are not
always adequate for the little finger. As we begin to
consider injuries in the little finger as qualitatively

152
Section 2:  Primary Flexor Tendon Surgery
References
1. Tang JB, Shi D, Gu YQ, et al: Double and multiple looped
suture tendon repair, J Hand Surg (Br) 19:699–703, 1994.
2. Tang JB, Shi D: Subdivision of flexor tendon “no man’s land”
and different treatment methods in each sub-zone. A prelimi­
nary report, Chin Med J (Engl) 105:60–68, 1992.
3. Tang JB: Flexor tendon repair in zone 2C, J Hand Surg (Br)
19:72–75, 1994.
4. Tang JB, Shi D, Shen SQ, et al: An investigation of morphol­
ogy and function of flexor tendons in zone IIC in the hand
and treatment of flexor tendons, Chin J Surg (in Chinese)
37:639, 1999.
5. Tang JB: Clinical outcomes associated with flexor tendon
repair, Hand Clin 21:199–210, 2005.
6. 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.
7. Tanaka T, Amadio PC, Zhao C, et al: Flexor digitorum pro­
fundus tendon tension during finger manipulation, J Hand
Ther 18:330–338, 2005.
8. 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.
9. Amadio P, An KN, Ejeskar A, et al: IFSSH Flexor Tendon Com­
mittee report, J Hand Surg (Br) 30:100–116, 2005.
10. Elliot D, Barbieri CH, Evans RB, et al: IFSSH Flexor Tendon
Committee Report 2007, J Hand Surg (Eur) 32:346–356, 2007.

153
Alphonsus K.S. Chong, MBBS, MRCS (Ed), MMed (Orth), FAMS,
Beng-Hai Lim, MD, and Yeong-Pin Peng, FRCS
E
The Singapore Experience
1978 showing its efficacy in our setting.3 At that time,
the practice was to perform the Bunnell technique repair
using a 4-0 wire suture with a 6-0 epitendinous suture.
Rehabilitation consisted of a dynamic protocol using a
dorsal splint and rubber band traction. Currently, our
practice involves the use of a strong surgical repair fol­
lowed by early protective motion protocol. With special­
ist hand surgeons performing primary tendon repairs
and dedicated hand therapists supervising rehabilita­
tion, our clinical results are similar to results obtained
elsewhere.
Despite this, our results mirror that of other centers
in that there are still a significant proportion of patients
with poor clinical results. Beyond the ultimate result,
there is also a large socioeconomic burden in terms of
time off work and other opportune costs.4 This is due
to the long rehabilitation time required before the
patient returns to having function of the injured hand.
Improvement of the results following flexor tendon
injury has long been an important research focus in
Singapore. In 1974, Dr. Chacha described his results
with free composite autologous tendon grafts in pri­
mates and humans5 as a method to combat the problem
of adhesions following flexor tendon injuries and
primary repair. We have been trying to improve the
outcome of primary repair of the injured flexor tendon
and to reduce adhesion formations after primary or
secondary tendon surgery since then.
Our research strategy to address the limitations with
current repair and rehabilitation techniques has been
on two fronts: (1) improving the strength of the repair
to enable early active motion and shorter rehabilitation
and (2) research into biological methods to improve
the results of flexor tendon healing. One of our aims in
the primary repair is to enable a splint-free flexor tendon
repair and rehabilitation.6 A more robust repair will
allow splint free, unprotected use of the hand during
the rehabilitation phase. This will reduce the costs of
such an injury and possibly improve the ultimate
outcome. We collaborated with our Finnish colleagues
to evaluate different materials, such as nickel-titanium
alloy, to assess its suitability for use in flexor tendon
repairs.7
OUTLINE
We used looped sutures to make a six-strand core
suture repair supplemented with running epitendi­
nous suture in repairing the lacerated flexor digitorum
profundus tendon. After surgery, we performed passive
flexion and active extension exercise. In our initial
experience, in 32 fingers that underwent this tendon
repair, there were 41% excellent, 41% good, and 16%
fair results according to the revised Strickland criteria.
The rupture rate was 3% (1 of 32 digits). One case with
a fair result underwent tenolysis, with no improve­
ment after the procedure. Our results in more recent
case series with this technique are similar to our initial
results.
Flexor tendon injuries are common work injuries in
Singapore because there is a large manufacturing and
construction sector in our economy. Home accidents
and assault with sharp objects are other common mech­
anisms of injury. Singapore’s small physical size means
that our population has easy access to specialist care by
hand surgeons following the initial injury. In addition,
compliance with postoperative rehabilitation by hand
therapists is high because of this accessibility. As a result,
acute flexor tendon injuries that require primary flexor
tendon repairs form the bulk of our clinical experience
with flexor tendon injuries. Secondary or late repairs
requiring staged procedures are uncommon in our
practice.
Flexor tendon injuries occur in isolation or in asso­
ciation with other injuries like fractures, nerve, and
vessel injuries. In more serious injuries, such as amputa­
tion, a strong tendon repair is also critical to allow early
mobilization and return of digital function.1 Great
improvements in the results following primary flexor
tendon repairs have been obtained with better repair
techniques and rehabilitation protocols.2
Our community’s practice pattern for flexor tendon
repairs has closely paralleled the experience of estab­
lished hand centers worldwide. The practice of the
primary repairs started taking off in the 1970s, and Pro­
fessor Robert W.H. Pho published his experience in

154
Section 2:  Primary Flexor Tendon Surgery
A variety of approaches have been taken to address
the biologic aspects of primary tendon healing. The
ability to increase the rate of healing or to improve
the ultimate quality of the healed tendon would be
extremely beneficial to patients. Our research in this
area has focused largely on the use of cells, primary
bone marrow–derived mesenchymal stem cells (bMSCs),
to enhance primary flexor tendon healing. Mesenchy­
mal stem cell marrow–based therapies show promise in
improving outcomes in a tendon injuries.4 Our interest
in this area began with the finding from our collabora­
tors that these bMSCs could be used to enhance the
histological and biomechanical properties of a tissue-
engineered tendon graft.8 We applied this strategy to a
primary tendon healing model in the rabbit Achilles
tendon with encouraging results.9 We are currently per­
forming the experiments using bMSCs to promote the
healing in a digital flexor tendon model.
The strength of a primarily repaired flexor tendon is
correlated with the number of core sutures. We favor six
core sutures in digital flexor tendon repairs. While this
may increase the bulk at the repair site, the gliding
characteristics are not compromised.10 Biomechanical
testing of six-strand repair techniques has also been
favorable. A human cadaveric tendon study demon­
strated that the six-strand repair technique using a loop
suture had a higher tensile strength compared to tech­
niques such as the modified Kessler (two-strand), the
Tsuge technique (two-strand), and the four-strand loop
suture technique.11
METHODS AND OUTCOMES
Operative Techniques
Our procedure for primary flexor tendon repairs is to
perform surgery with the patient under general or
regional anesthesia. The final choice depends on the
patient and the anesthetist.
We favor a six-strand core repair technique
co-developed by our senior author.12 This repair tech­
nique uses a looped suture made of nylon (Supramid
4/0; S. Jackson Inc., Alexandria, VA, USA) for the core
sutures. During the core suture repair, we add an epiten­
dinous repair using polypropylene (Prolene 5-0 or 6-0,
depending on tendon size).
The use of a loop suture brings three advantages: it
doubles the number of core sutures with each pass,
simplifies suture locking to the tendon, and reduces the
number of passes and tendon handling. Our clinical
and published experience involves using the commer­
cially available Supramid sutures. However, we realize
that these looped sutures may not be universally avail­
able, or the cost may be prohibitive in some countries.
We have described a technique to create a looped suture
using a straight 23-gauge hypodermic needle, pliers, and
available sutures such as polypropylene 4-0.13 Using this
technique, a 90-cm length of polypropylene can be used
to make three 15-cm strands of looped suture. This
allows the surgeon to capitalize on the advantages of a
loop suture with minimal effort.
In the original description of the technique,11 the
repair used two strands of looped suture for the repair.
We have since modified the technique to use only one
looped suture per repair. This results in a repair with
six-core sutures to readily allow for dynamic postopera­
tive rehabilitation. For larger flexor tendons in the wrist,
an 8- to 10-strand repair may be possible.
The technique needs all the usual requisites of a good
tension-free repair. The dorsal half of the epitendinous
repair can be done first to allow good approximation of
the tendon ends. The core suture repair itself is then
performed as shown (Figure 13[E]-1).
The flexor digitorum superficialis (FDS) tendon is
repaired with a simple Tsuge loop suture technique if it
is near the insertion or in zone 2C with the knot on the
dorsal side. If the FDS tendon is cut more proximally
(i.e., it lies on superficial to the deep flexor tendon), a
six-strand repair is made.
Postoperative Care
The patient is often discharged the same or next day if
the injury was uncomplicated. For more complicated
cases, such as microvascular repair or mutilating inju­
ries, the patient will remain as an inpatient as required.
The protective backslab and dressing is replaced with a
removable splint within 5 days of surgery by the hand
therapist. Protected dynamic rehabilitation is started
then. The splint is removed at about 4 weeks postinjury
by the therapist. Return to full activities occurs at about
12 weeks.
In the first 4 weeks, a dorsal blocking splint is applied
for the wrist and digits for continuous wear. The splint
places the wrist in a neutral position, with metacarpo­
phalangeal joints flexed 45°, and interphalangeal (IP)
joints in full extension (Figure 13[E]-2). The patient is
advised to strap the fingers in flexion during the day and
loosely in extension within the splint at night.
With the splint in place, the patient performs the fol­
lowing exercises hourly: (1) 20 repetitions of passive
flexion and extension of the distal IP (DIP) joint, (2) 20
repetitions of passive flexion and extension of the proxi­
mal IP (PIP) joint, and (3) passively bringing all fingers
into hook position (full IP joint flexion) and actively
holding down the finger lightly for 5 seconds, followed
by active extension of the fingers to fully straighten the
fingers against the splint for 5 repetitions. This last exer­
cise is done only when edema is resolved. Edema control
is achieved using Coban wrapping (a self-adherent tape)
or fingerstalls, in addition to limb elevation.
Therapy sessions are given three times a week. In
addition to supervision of the patient’s therapy, the
therapist manages the wound and scar and edema and

155
Chapter 13E:  The Singapore Experience
Figure 13(E)-1  A, Application of a superficial locking
suture in the proximal tendon 1.25 cm from the tendon
end. B, Placement of the first pair of core sutures, with the
needle inserted close to the locking suture and proceeding
parallel to the tendon fibers, from proximal to distal,
surfacing 1 cm from the distal tendon end. C, Application
of a superficial locking suture in the distal tendon 1.25 cm
from the tendon end. D, Placement of second pair of core
sutures across repair site in a similar fashion and locking
suture applied. E, Application of final pair of core sutures.
Note that one strand is divided at the repair site, while the
other strand crosses into the proximal end and is locked
before being passed into the repair site. The dorsal
epitendinous running suture is applied. F, The loop suture
repair is knotted in the repair site.
A
B
C
D
E
F
performs passive tenodesis exercises to prevent joint
stiffness.
At 4 to 6 weeks postsurgery, the dorsal block splint is
removed during the daytime. The patient wears the
splint at night and while going into crowded places. A
night gutter splint is worn if there is flexion contracture
of the PIP joints. The exercise program includes (1)
active flexion and extension of fingers (to initiate
blocked active hook, perform straight fist and full fist
exercises) and (2) blocking exercises to improve gliding
of the FDP and FDS tendons.
At 6 to 8 weeks, the following activities are performed:
(1) light activities of daily living (e.g., writing, eating,
and combing hair are encouraged), (2) progressive
strengthening (e.g., putty, hand helper), (3) light resis­
tive exercises (e.g., light putty, light Velcro), (4) gentle
stretching of flexors to reduce tightness and adhesion,
and (5) gliding of flexor tendons and positioning at the
end of elastic limit.
At 8 to 12 weeks, work conditioning and hardening
may be initiated. Normal use of hand is allowed. Heavy
lifting is allowed only after 12 weeks.
OUTCOMES
In our initial experience, we found that in 32 fingers that
underwent this tendon repair, there were 41% excellent,
41% good, and 16% fair results using the revised Strick­
land criteria.12 Twenty-one cases were repaired primarily
(within 24 hours), while 11 cases underwent delayed
primary repair. The rupture rate was 3% (1 of 32 digits).
One case with a fair result underwent tenolysis, with no
improvement after the procedure. Our results in more
recent case series with this technique are similar to our
initial results.
Figure 13(E)-2  The operated hand in a splint during the
first 4 weeks following tendon repair. The digital strap is
removed during exercises. Passive flexion of the digits is
done, followed by an active hold for 5 seconds.

156
Section 2:  Primary Flexor Tendon Surgery
suture. The rehabilitation technique we use affords suf­
ficient protection of the repair, with the advantages of
early mobilization. Effective management of edema
and good communication between patient, surgeon,
and therapist enhance the rehabilitation process. There
remains much to be done to improve the clinical
outcome following these injuries, which typically occur
in the young and socioeconomically active. Efforts to
improve repair materials, techniques, rehabilitation,
and biologic methods to modulate healing will all con­
tribute to the improvement of flexor tendon healing
results in the future.
ACKNOWLEDGMENTS
Ms. Tan Lay Lay, principal hand therapist, and the Hand
Occupational Therapy Workgroup of National Univer­
sity Hospital, Singapore, are acknowledged for the flexor
tendon rehabilitation protocol.
SUMMARY
In an independent study, Hoffman and his colleagues
compared this technique with the modified Kessler
repair, combined with a Kleinert/Duran mobilization
regimen.14 After a follow-up of 8 to 17 weeks, the loop
suture technique group had a better grip strength and a
significantly better total active motion of 141° com­
pared with 123°. In addition, the complication rate
was significantly lower and the average time of treat­
ment was significantly shorter using the loop suture
technique. The rupture rate was lower in the loop suture
technique (1 of 51 versus 3 of 26), although this did not
reach statistical significance. The results of Hoffman and
his colleagues are similar to results of our case series
using similar repair techniques.
The repair technique we describe using the looped
Supramid suture adheres well to well-established prin­
ciples and brings with it the advantages of using a loop
References
1. Lim BH, Tan BK, Peng YP: Digital replantations including
fingertip and ring avulsion, Hand Clin 17:419–431, 2001.
2. Boyer MI, Strickland JW, Engles D, et al: Flexor tendon repair
and rehabilitation: state of the art in 2002, Instr Course Lect
52:137–161, 2003.
3. Pho RWH, Sanguin R, Chacha PB: Primary repair of flexor
tendons within the digital theca of the hand, Hand 10:154–
160, 1978.
4. Chong AK, Chang J, Go JC: Mesenchymal stem cells and
tendon healing, Front Biosci 14:4598–4605, 2009.
5. Chacha P: Free autologous composite tendon grafts for divi­
sion of both flexor tendons within the digital theca of the
hand, J Bone Joint Surg (Am) 56:960–978, 1974.
6. Peng YP, Lim BH, Chou SM: Towards a splint-free repair for
flexor tendon injuries, Ann Acad Med Singapore 31:593–597,
2002.
7. Karjalainen T, He M, Chong AK, et al: Nickel-titanium wire
in circumferential suture of a flexor tendon repair: A compari­
son to polypropylene, J Hand Surg (Am) 35:1160–1164, 2010.
8. Ouyang HW, Goh JC, Thambyah A, et al: Knitted poly-lactide-
co-glycolide scaffold loaded with bone marrow stromal cells
in repair and regeneration of rabbit Achilles tendon, Tissue
Eng 9:431–439, 2003.
9. Chong AK, Ang AD, Goh JC, et al: Bone marrow-derived
mesenchymal stem cells influence early tendon-healing in a
rabbit Achilles tendon model, J Bone Joint Surg (Am) 89:74–
81, 2007.
10. Sanders DW, Milne AD, Johnson JA, et al: The effect of flexor
tendon repair bulk on tendon gliding during simulated active
motion: An in vitro comparison of two-strand and six-strand
techniques, J Hand Surg (Am) 26:833–840, 2001.
11. Gill RS, Lim BH, Shatford RA, et al: A comparative analysis
of the six-strand double-loop flexor tendon repair and three
other techniques: A human cadaveric study, J Hand Surg (Am)
24:1315–1322, 1999.
12. Lim BH, Tsai TM: The six-strand technique for flexor tendon
repair, Atlas Hand Clin 65–76, 1996.
13. Wong M, Sebastin SJ, Lim BH: A simple technique of making
a looped suture for flexor tendon repair, J Hand Surg (Eur)
34:409–410, 2009.
14. Hoffmann GL, Büchler U, Vögelin E: Clinical results of flexor
tendon repair in zone II using a six-strand double-loop tech­
nique compared with a two-strand technique, J Hand Surg
(Eur) 33:418–423, 2008.

157
Arash Momeni, MD, Emily Grauel, MS, and
James Chang, MD
F
The Stanford Experience
Several studies have demonstrated the effect of the
following intraoperative factors on repair strength and
functional outcome:7
1. Number of core suture strands
An increase in the number of core sutures (at least
four strands) has been demonstrated to result in
an increased resistance to gap formation at the
repair site when compared with conventional two-
strand repairs.8-10 As a result of a stronger repair,
early active motion protocols may be instituted
postoperatively.
2. Caliber of core suture
Several authors have demonstrated that the use of
a larger caliber core suture (greater than 4-0) is
paralleled by an increased strength and stiffness of
the repair.11,12
3. Tendon purchase
Adequate tendon purchase when placing the core
suture is critical for a successful repair. The optimal
length of purchase has been shown to be between
0.7 and 1.0 cm.13 Similarly, placing the epitendi­
nous suture 2 mm from the cut end results in
increased repair strength compared with a 1-mm
purchase.
4. Locking of core suture
The superiority of a locking core suture with resul­
tant increased tensile strength and reduced risk of
suture pull-out was initially reported by Penning­
ton and has since been confirmed in numerous
reports.14,15
5. Epitendinous suture
Although it was initially assumed that the epiten­
dinous suture would merely contour the repair site,
it has been demonstrated that it contributes to the
strength of the repair as it bolsters repair strength
and prevents gap formation.16 Implementing an
epitenon-first technique further enhances repair
strength when compared with a modified Kessler
and epitendinous running suture.17
The complexity of flexor tendon repairs along with a
“no-error tolerance” environment in the operating room
represent significant challenges to resident training. As
such, surgical training is increasingly making use of
lessons learned from the aviation industry and has
OUTLINE
This chapter summarizes some of the important con­
siderations and current methods of primary repairs of
the injured flexor tendons in the authors’ unit. The
preferred postoperative rehabilitation protocols are
presented and discussed.
Management of flexor tendon injuries represents one of
the most demanding challenges in hand surgery. The
complexity of the task is reflected by the fact that despite
excellent technical execution numerous complications
may be experienced postoperatively.1 Particularly, repair
of zone 2 injuries has been an area of great controversy
since Bunnell’s proposal that “it is better to remove the
tendons entirely from the finger and graft in new tendon
throughout its length.”2
Flexor tendon repair confronts the surgeon with a task
that is two-fold, namely reestablishment of not only
tendon continuity but also the gliding mechanism of the
tendon and its surrounding structures. Achieving both
goals is equally important for a satisfactory functional
outcome.3 The various factors that affect functional
outcome can be classified into preoperative, intraopera­
tive, and postoperative factors (Table 13[F]-1), of which
the latter two can be influenced by the surgeon.
Particularly with regard to surgical technique, the
scientific literature is replete with technical modifica­
tions proposed for improved functional outcome.4
The multitude of available approaches along with the
nuances of flexor tendon repair make it particularly dif­
ficult for the trainee to identify and learn a reliable
technique, which allows for reproducible results with a
satisfactory functional outcome. In particular, zone 2
injuries are associated with poorer functional outcomes
and a higher complication rate compared with injuries
in other zones.5
Despite the myriad of reported techniques, several
principles of repair should be regarded as established.
These principles include early primary repair whenever
possible, performance of a multistrand repair (at least
four strands) combined with an epitendinous suture,
use of locking stitches, and adequate tendon purchase.
Furthermore, an early motion protocol should be used
in compliant patients to minimize extrinsic scarring and
adhesion formation and to promote intrinsic healing.6

158
Section 2:  Primary Flexor Tendon Surgery
an epitendinous running suture using a 6-0 nonabsorb­
able monofilament suture on a tapered needle. The
tendon is purchased 2 mm from the cut end for place­
ment of this suture. This results in optimal tendon align­
ment and prevents the excessive bulging observed when
placing the core suture first. Next, a longitudinal slit is
made with a No. 11 blade at least 1 cm away from the
repaired edge. Using this slit as the entry point, a 3-0
nonabsorbable, braided suture is placed in a modified
Kessler technique with a locking loop. In the end, the
knot should lie in the depth of the previously created
slit, which will limit irritation of the tendon sheath.
Finally, a four-strand repair is obtained by placement of
a horizontal mattress stitch with a 3-0 nonabsorbable,
braided suture. The knot of this suture should be placed
away from the A2 and A4 pulleys. The repaired digit
should then be placed under passive range of motion
(ROM) intraoperatively to assess for triggering. If trig­
gering at one of the pulleys is observed, partial release
of the affected pulley is indicated. Postoperatively, early
active motion protocols are initiated.
endorsed simulator-based training models. Organiza­
tions such as the American College of Surgeons (ACS)
as well as the Accreditation Council for Graduate Medical
Education (ACGME) have underlined the importance of
such training models in surgical education.
Surprisingly, the challenges of flexor tendon repair
have not been met by appropriate simulation models.
Recently, colleagues in our unit introduced a curriculum-
based simulation model for teaching zone 2 flexor
tendon repairs. Combining a tutorial addressing con­
temporary literature along with providing a standard­
ized approach to flexor tendon repair incorporating
best evidence principles with hands-on experience
using human cadaver tendons resulted in significantly
improved flexor tendon repairs. Key aspects of this sim­
ulation model include inexpensive hands-on experi­
ence, review of the primary literature, and standardization
of the technique. This model is amenable to incorpora­
tion into training programs as part of a hand surgery
simulation curriculum.18
OPERATIVE TECHNIQUE
The authors’ preferred approach to flexor tendon inju­
ries, particularly those in zone 2, is early exploration and
repair via Bruner zigzag incisions. Every effort should be
made to repair each injured tendon, even in the setting
of combined flexor digitorum superficialis (FDS) and
flexor digitorum profundus (FDP) injuries. Although
some authors recommend repair of only the FDP tendon,
the authors believe that excessive scar formation origi­
nating from the free ends of an unrepaired FDS tendon
may result in functional compromise of the digit. Fur­
thermore, repair of both tendons will still allow proxi­
mal interphalangeal (PIP) joint flexion should the
repaired FDP tendon rupture during postoperative reha­
bilitation. During the repair, care should be taken to
preserve the A2 and A4 pulleys to prevent postoperative
bowstringing, which would result in loss of function.
The authors’ preferred method of flexor tendon repair
is illustrated in Figure 13(F)-1. The repair begins with
Figure 13(F)-1  Authors’ preferred flexor tendon repair
using epitenon-first suture (A) followed by a modified
two-strand Kessler core suture (B–E) in locking configuration
and buried knot (F). Subsequently, two-strand horizontal
mattress suture is added (G), to complete a four-strand core
suture.
C
E
G
F
D
B
A
Epitendinous stitch
Slit with #11 blade
Modified Kessler
Horizontal mattress
Locking stitch
2 mm
0.7 to 1 cm
Table 13(F)-1  Factors Affecting Outcome After
Primary Flexor Tendon Repair
Preoperative
Factors
Intraoperative
Factors
Postoperative
Factors
 Type/
mechanism of
trauma
 Degree of
wound
contamination
 Associated
injuries (e.g.,
concomitant
fractures, etc.)
 Atraumatic
tissue handling
 Multistrand
core suture
 Locking stitch
 Epitendinous
suture
 Tendon
purchase
 Suture material
 Postoperative
rehabilitation
(i.e., early
active
motion)

159
Chapter 13F:  The Stanford Experience
at all times until its removal is recommended by the
therapist. Patient education about the proper way to
undergo active motion exercises is addressed in detail at
the first therapy visit.
Five therapeutic exercises are begun between days 3
and 7 postoperatively.
1. Passive range of motion composite wrist flexion.
2. Passive MCP flexion to 90° and active extension
of IP joints to neutral; Passive PIP/DIP extension
with MCP flexed.
3. Place and hold flexion of digits in palm with active
extension to DBS.
4. Passive wrist flexion with active extension to
neutral.
5. Active flexion and extension of unaffected digits
within the DBS while maintaining the affected
digit in passive flexion.
In clinic, the therapist performs retrograde massage, scar
massage, and skin care as needed. Precautions during
this stage of therapy include: no active flexion of the
affected digit, no active extension of MCP beyond 60°,
no active PIP extension unless MCP is flexed at 90°, and
no active wrist extension past neutral.
Weeks 4 to 5
The patient should continue to wear the DBS at all times
and is instructed in new therapeutic exercises:
1. Begin active digital flexion of affected digit within
the DBS with active extension to hood of splint;
the therapist may consider buddy taping affected
to adjacent digit within the splint for active flexion
if the affected digit is stiff.
2. Continue passive range of motion digital flexion.
3. During week 5 include full digital extension to
neutral.
4. Tenodesis of wrist out of DBS.
In clinic, the therapist performs edema management
techniques (retrograde massage, 3M Coban Self-
Adherent Wrap application), scar management tech­
niques (use of silicone gel sheet if needed), and skin
care. The main precaution during this stage of therapy
is avoidance of forceful flexion.
Week 6
At this point, the DBS can be removed except at night
and with lifting or higher risk activities. Therapeutic
exercises for week 6 include:
1. Begin active isolated joint blocking.
2. Active digital extension to neutral.
3. If lacking full PIP extension, a static PIP extension
splint can be fabricated.
4. Active wrist flexion and extension.
POSTOPERATIVE REHABILITATION
Early mobilization protocols are now established com­
ponents of the comprehensive care provided to patients
following flexor tendon repair as they have been dem­
onstrated to result in superior functional outcomes
compared with prolonged immobilization. Among the
proposed benefits are improved tendon gliding in the
postoperative period with resultant decreased adhesion
formation and risk of joint stiffness.18 Early mobiliza­
tion also improves healing and revascularization leading
to an increased strength of the repair.19 Numerous
studies indicate that initiation of mobilization is most
successful in the first week; however, recommendations
for the specific day within that week vary.20,21 In vivo
animal studies suggest days 3 to 5 as the best choice to
begin a postoperative mobilization regimen based on
the strength of the repair and the resistance to digital
motion.21,22 Although the concept of early postoperative
mobilization represents a central dogma, a myriad of
rehabilitation protocols can be found in the literature.24
Standardization of protocols based on best available
evidence will lead to greater success of flexor tendon
repairs and allow comparative analyses of functional
outcomes.
There are two categories of early mobilization: early
passive motion and early active motion. Early passive
motion includes passive flexion (e.g., with rubber band
traction) with either active or passive extension. Early
active motion includes active flexion and extension of
the affected digit. A subtype of early active mobilization
includes “place and hold” techniques where the digit is
passively flexed and then held in place by the patient.23
Early controlled active finger flexion with place and
hold exercises is becoming a more popular treatment
compared to early passive flexion by rubber band
traction.24
Recently, Trumble et al conducted the first random­
ized prospective trial comparing active place-and-hold
therapy with passive motion therapy.6 A total of 103
patients (119 digits) were included in the study. The
authors demonstrated that active motion therapy pro­
vided significantly better range of motion, smaller
flexion contractures, and greater patient satisfaction.
This study provides strong scientific support for the
combination of multistrand tendon repairs and early
active motion protocols postoperatively.6
AUTHORS’ PREFERRED POSTOPERATIVE
REHABILITATION PROGRAM
Weeks 0 to 3
Postoperatively, the patient’s affected hand is placed in
a dorsal blocking splint (DBS) with the wrist neutral,
metacarpophalangeal (MCP) joints at 70° to 90°
flexion, and IP joints neutral. The DBS should be worn

160
Section 2:  Primary Flexor Tendon Surgery
SUMMARY
A growing body of evidence supports the combination
of multistrand (at least four strands) repairs and early
active motion rehabilitation protocols for management
of flexor tendon injuries. Incorporating simulation
modules into the hand surgery curriculum is an impor­
tant step in allowing the trainee to familiarize with the
technical complexity of flexor tendon repairs prior to
execution in the operating room. Standardization of
treatment protocols based on best available evidence
will lead to greater success of flexor tendon repairs and
also allow comparative analyses of functional outcomes
across various institutions.
5. May begin ultrasound/neuromuscular electrical
stimulation as needed for scar management if not
obtaining tendon glide.
6. Encourage use of hand for light functional
activities.
7. Continue edema management techniques as
needed.
Week 8
The DBS can be fully discharged at this point. Strength­
ening exercises should be begun at this point and should
gradually increase as the patient can tolerate. Return to
sports activities should be discussed with the physician,
as full return is sports specific. Generally, full release for
sports is at 12 weeks.
References
1. Momeni A, Grauel E, Chang J: Complications after flexor
tendon injuries, Hand Clin 26:179–189, 2010.
2. Bunnell S: Repair of tendons in the fingers and description
of two new instruments, Surg Gyencol Obstet 26:103–110,
1918.
3. Beredjiklian PK: Biologic aspects of flexor tendon laceration
and repair, J Bone Joint Surg (Am) 85:539–550, 2003.
4. Strickland JW: Development of flexor tendon surgery: twenty-
five years of progress, J Hand Surg (Am) 25:214–235, 2000.
5. Karlander LE, Berggren M, Larsson M, et al: Improved results
in zone 2 flexor tendon injuries with a modified technique
of immediate controlled mobilization, J Hand Surg (Br)
18:26–30, 1993.
6. 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.
7. Kim HM, Nelson G, Thomopoulos S, et al: Technical and
biological modifications for enhanced flexor tendon repair,
J Hand Surg (Am) 35:1031–1037, 2010.
8. Savage R, Risitano G: Flexor tendon repair using a “six strand”
method of repair and early active mobilisation, J Hand Surg
(Br) 14:396–399, 1989.
9. Viinikainen A, Goransson H, Huovinen K, et al: A compara­
tive analysis of the biomechanical behaviour of five flexor
tendon core sutures, J Hand Surg (Br) 29:536–543, 2004.
10. Thurman RT, Trumble TE, Hanel DP, et al: Two-, four-, and
six-strand zone II flexor tendon repairs: An in situ biome­
chanical comparison using a cadaver model, J Hand Surg
(Am) 23:261–265, 1998.
11. Barrie KA, Tomak SL, Cholewicki J, et al: Effect of suture
locking and suture caliber on fatigue strength of flexor tendon
repairs, J Hand Surg (Am) 26:340–346, 2001.
12. Taras JS, Raphael JS, Marczyk SC, et al: Evaluation of suture
caliber in flexor tendon repair, J Hand Surg (Am) 26:1100–
1104, 2001.
13. Tang JB, Zhang Y, Cao Y, et al: Core suture purchase affects
strength of tendon repairs, J Hand Surg (Am) 30:1262–1266,
2005.
14. Pennington DG: The locking loop tendon suture, Plast Recon­
str Surg 63:648–652, 1979.
15. Hotokezaka S, Manske PR: Differences between locking loops
and grasping loops: effects on two-strand core suture, J Hand
Surg (Am) 22:995–1003, 1997.
16. Wade PJ, Muir IF, Hutcheon LL: Primary flexor tendon repair:
the mechanical limitations of the modified Kessler technique,
J Hand Surg (Br) 11:71–76, 1986.
17. Papandrea R, Seitz WH, Shapiro P, et al: Biomechanical and
clinical evaluation of the epitenon-first technique of flexor
tendon repair, J Hand Surg (Am) 20:261–266, 1995.
18. Amadio P, An KN, Ejeskar A, et al: IFSSH Flexor Tendon Com­
mittee report, J Hand Surg (Br) 30:100–116, 2005.
19. Strickland JW, Glogovac SV: Digital function following flexor
tendon repair in Zone II: A comparison of immobilization
and controlled passive motion techniques, J Hand Surg (Am)
5:537–543, 1980.
20. Tottenham VM, Wilton-Bennett K, Jeffrey J: Effects of delayed
therapeutic intervention following zone II flexor tendon
repair, J Hand Ther 8:23–26, 1995.
21. 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) 86:320–327, 2004.
22. 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.
23. Pettengill KM: The evolution of early mobilization of the
repaired flexor tendon, J Hand Ther 18:157–168, 2005.
24. 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.

161
Michael A. Tonkin, MD, FRCS Ed Orth, FRACS, and
Richard D. Lawson, MBBS, FRACS
G
The Australian Experience
followed by supervised active movement within the
inner range, once or twice a day. The splint was main­
tained for 5 weeks.
These methods of repair and rehabilitation were still
the mode of practice until the early 1970s. Since then,
flexor tendon surgery has evolved significantly and Aus­
tralian hand surgeons have made many contributions
to the development of the increasingly sophisticated
techniques currently practiced.
Perhaps one of the most important but basic contri­
butions was David Pennington’s recognition in 1979
that a locking loop tendon suture increased the tensile
strength of the repair and decreased gapping at the
repair site.2 He demonstrated that the alteration in the
relationship of the transverse and longitudinal intraten­
dinous parts of the suture technique of Mason and
Allen,3 modified by Kessler and Nissim,4 was vital to
prevent pullout of the suture under load. Now termed
the modified Kessler suture, the longitudinal compo­
nents of the suture configuration must pass deep to the
transverse limb. As tension is increased, the loops
tighten and lock. In fresh cadaveric flexor tendons, he
demonstrated that failure of the locking loop suture was
due to breakage of the suture, rather than suture pullout
that occurred with a nonlocking technique. Pennington
also performed some experimental repairs of cadaveric
tendons with wire and monofilament nylon, conclud­
ing that the stiffness of wire made it an unsuitable
material.2
The modified Kessler suture became the “work horse”
core suture for many years and is still favored by many,
although the use of monofilament nylon was largely
superseded by braided polyester sutures, and more
recently by other materials such as Fiberwire (Arthrex,
Naples, FL).
Pennington’s innovative approach to improving
flexor tendon surgical techniques included a discussion
of the use of a suction catheter technique to deliver the
retracted proximal end of a severed tendon to the repair
site when proximal digital milking of the tendon failed.5
Perhaps the most common technique in current practice
is that described by McGrouther and colleagues, from
the United Kingdom, in 1987, in which a fine Silastic
catheter is passed up the tendon sheath from a proximal
palmar wound and is sutured to the retracted tendon,
OUTLINE
This subchapter describes the contribution of Austra­
lian authors to the basic science and clinical manage­
ment of flexor tendon injuries. It then discusses the
current surgical management and rehabilitation of
flexor tendon injuries in Australia and ends with a
brief outline of our preferred approach to flexor
tendon repair.
In Australia, hand surgery developed in parallel with
and perhaps because of the necessity for hand surgery
services created by limb injuries sustained in World War
II. Sir Benjamin Rank, from Melbourne, though mostly
involved in secondary reparative surgery during and
immediately after the war years, established the place of
primary repair for hand injuries which became rife with
the mushrooming of light industry. Rank and Alan
Wakefield published Surgery of Repair as Applied to Hand
Injuries in 1953,1 being joined for the last two of five
editions by John Hueston. Their recommendations for
the place of primary and secondary flexor tendon recon­
struction and the techniques of the time are documented
in this text, which can best be summarized in the fol­
lowing: Primary repair in the digit was practiced for
flexor digitorum profundus (FDP) division only if the
repair did not interfere with the intact flexor digitorum
superficialis (FDS). If the primary repair of an FDP
tendon was likely to compromise FDS function, the FDP
was not repaired, relying on a tenodesis or an arthrod­
esis of the distal interphalangeal joint; alternatively, a
primary or delayed tendon graft from palm to terminal
interphalangeal joint was used. If both tendons were
divided, a graft was indicated. Results from primary
grafting did not compare favorably with those of sec­
ondary grafts. This became one of the exceptions to
the rule of primary repair in a tidy wound, which was
championed for all other soft tissue injuries. Primary
skin closure with delayed or secondary grafting was
favored. For primary repair, the suture used was a 5-0
polyester suture, inserted according to the method
described by Bunnell. The 6-0 sutures were used for
auxiliary apposition stitches, so as not to leave any open
tendon end exposed. Post-operative treatment main­
tained absolute immobilization until the third week,

162
Section 2:  Primary Flexor Tendon Surgery
release knot is disengaged by pulling on the trailing tail
suture at the palmar wound (Figure 13[G]-1A–D).
Wayne Morrison, from Melbourne, has been another
major contributor to many aspects of flexor tendon
surgery in Australia. Writing with Pribaz and Macleod,
he reported satisfactory results using the Becker repair
and early active motion.8 Becker’s beveling and overlap­
ping of the tendon ends reduced the problem of gap
formation but did require profundus tendon shortening
which is then drawn into the repair site.6 Michael
Sandow, from Adelaide, suggested a modification of this
technique in his article “A Further Tendon Retrieval
Trick,” published in 1997.7 Instead of a simple knot to
secure the catheter to the tendon, a loop release knot is
used, which allows the tendon to be advanced with the
tube; once the tendon is delivered and stabilized dis­
tally, the knot can be released without the need to allow
the tendon to retract to the proximal wound. The loop
Figure 13(G)-1  Diagram for tendon retrieval technique described by Michael J. Sandow. A, Loop release knot using 4-0
nylon is tied with a double throw first. B, Loop release knot: second throw. C, Loop release knot: completed securing catheter
to tendon. D, Catheter is pulled distally advancing tendon, the tendon is secured distally, the loop release knot is disengaged
by pulling on the trailing suture tail, the suture is removed, and the catheter is disengaged leaving the tendon in a good
position for repair. (From Sandow MJ: A further tendon retrieval trick, J Hand Surg [Br] 22[1]:125-127, 1997.)
Pull to
advance
tendon
A
B
C
D
1
Secure
tendon
distally
2
Pull to
release
knot
(distal
catheter
counter
traction)
3
Pull to
undo
suture
(distal
catheter
counter
traction)
4
Pull to
remove
catheter
5

163
Chapter 13G:  The Australian Experience
continued at a hectic pace. Most “improvements” have
centered on the development of stronger core and epi­
tendinous suture constructions which allow early post­
operative active mobilization. Savage, from the United
Kingdom, demonstrated the significantly improved
strength of repair gained by increasing the number of
strands within the core suture.15 Strickland’s graph is a
very helpful illustration of this concept comparing the
force generated by different postoperative rehabilitation
programs with the tensile strength of two-, four-, and
six-strand core sutures16 (Figure 13[G]-2). Descriptions
of modifications, all claiming some improvement in
one or more of the measured parameters, abound within
the current hand surgery literature. One is reminded of
the words of Guy Pulvertaft, in whose Derby hand unit
many Australian surgeons were trained: “it is not diffi­
cult to suture tendons and prepare the ground for sound
union; the real problem is to obtain a freely sliding
tendon capable of restoring good function.”17 Pulver­
taft’s point is pertinent. No tendon repair, regardless of
the number of strands in the core suture, technique of
epitendinous suture, or gauge of suture, will protect
against inappropriate postoperative activity against
resistance. Gelberman and others, from the United
States, have shown that the amount of movement
required to gain optimal promotion of tendon healing
and overcome adhesion formation is small.18,19 It would
seem logical to balance the determination to improve
the strength of repair against the increases in adhesion
formation, bulk, and resistance to glide, which may
Figure 13(G)-2  A strength-versus-force graph showing
two-, four-, and six-strand repairs plotted against passive,
light active flexion, and strong grip. The data are adjusted for
friction, edema, and stress. (Data from Strickland JW: 25th
anniversary presentation: Development of flexor tendon
surgery: Twenty-five years of progress, J Hand Surg [Am]
25[2]:214-235, 2000.)
0
2000
4000
6000
8000
Force in grams
Strong
grip
6-strand
4-strand
2-strand
Light
active
Passive
Repair
1 week
3 weeks
6 weeks
Time in weeks
of around 7 mm. Morrison’s ruptures (3 of 43) occurred
in those patients with crush injuries and stiff joints.
Morrison, in collaboration with Callan, described a
new and simple approach to gain access to the divided
flexor tendons within the flexor tendon sheath.9 They
recommended a transverse incision 1 cm distal to the
distal stump of the severed tendon. This allows delivery
of 1 cm of distal tendon through the sheath incision, be
it within the synovial sheath or within a pulley. The
proximal segment is retrieved and delivered through
the same incision by whatever technique is necessary,
including milking, flexion of proximal joints, and
catheter-assisted delivery, and is stabilized with a
23-gauge needle, allowing the core suture to embrace
1 cm of each end of the tendon, with ready access pro­
vided to the front and back walls for insertion of the
epitendinous suture. These authors claim that the lon­
gitudinal component of Lister’s L-shaped sheath inci­
sions created the possibility of narrowing the sheath
with repair, however many still prefer the increased
access gained by the L-shaped incisions. Tonkin and
Lister found no clinical evidence in their study to
support the necessity for closure of the synovial compo­
nent of the tendon sheath.10
Morrison, O’Brien, and co-authors investigated a
number of innovations designed to improve the results
of staged flexor tendon surgery. They developed a model
in which a flexor tendon was placed into bone prior to
using that tendon as a flexor tendon graft.11 Incorpora­
tion of the tendon into bone allowed the bone block to
be used as superior fixation into the digital phalanx at
the time of tendon grafting. This prefabricated bone-
tendon graft offered some theoretical advantages per­
taining to distal pullout strength but has not been
commonly adopted. Another fascinating study from the
same group compared the results of vascularized and
nonvascularized tendon grafts in a primate experimen­
tal model.12 The authors concluded that the concept of
vascularized tendon grafting may be advantageous in
scarred tendon beds. This work was developed during a
time of euphoria associated with the application of
microsurgical techniques to hand surgery but, although
it has some significant theoretical advantages, it has not
displaced the conventional approach of staged flexor
tendon grafting. Honner and Meares confirmed the reli­
ability of this technique in their review of 100 cases in
1977.13 The consistency of the results of flexor tendon
grafting performed with precise attention to detail was
also reported by Tonkin and Hagberg’s review of the
experience of Kleinert’s Louisville center, a fertile train­
ing ground for Australian hand surgeons.14
RECENT ADVANCES, 2000–2010
More recent Australian contributions relate to biome­
chanical testing of alternative suture configurations.
The search for optimal flexor tendon repair results has

164
Section 2:  Primary Flexor Tendon Surgery
laboratory, applied a cyclical testing protocol to a com­
parison of in vitro repairs using a modified Kessler core
suture or a four-strand modified Savage core suture,
with two different epitendinous techniques. They found
the modified Savage repair to be superior.
Vizesi and colleagues analyzed the stress relaxation
and creep properties of three materials used for flexor
tendon repair—Prolene (monofilament polypropyl­
ene), Ethilon (monofilament nylon), and Ticron
(braided polyester).26 Their study addressed the selec­
tion of suture material based on the temperature- and
time-dependent mechanical properties of a single-
strand suture. They reported that, based on static and
viscoelastic mechanical properties, Ticron was the most
suitable suture for flexor tendon repairs compared with
Prolene and Ethilon.
A number of studies from the same department have
investigated optimal methods of securing the FDP
tendon distally following avulsion. Schreuder et al
found that gap formation in such repairs was signifi­
cantly less when Ethibond (a braided polyester) was
used compared with Prolene or Supramid (a synthetic
polyfilament ensheathed by Caprolactin).27 Latendresse
et al28 compared the repair of FDP tendons using a
single micro-Mitek anchor with a modified Bunnell
two-strand pullout technique, using a monofilament or
follow excessive handling of the tendon when placing
more sophisticated suture configurations. Increased
adhesion formation was not found in the experimental
studies of Tonkin’s group in Sydney, which compared
adhesion formation in two- and four-strand configura­
tions with and without epitendinous sutures in a live
chicken model.20 However, this group concluded that
the expertise of the surgeon may play a substantial role
and suggested that complicated techniques in inexperi­
enced hands increased the potential for a less than sat­
isfactory result. It would seem appropriate to establish
surgical techniques that obtain reproducible results in
the hands of those who, although competent, may have
less experience than those who are able to confidently
perform more complex techniques.
Sandow found the 6-strand Savage configuration to
be too complex. His modification is similar to that
made by others and involves a 4-strand repair.21 This
“Adelaide” repair has been adopted by many Australian
hand surgeons. It provides four strands to the core
suture, is less likely to gap than a modified Kessler core
suture, is relatively easy to place in the tendon ends, and
may be combined with most of the currently preferred
epitendinous suture configurations. Its strength permits
protected active postoperative mobilization (Figure
13[G]-3).
Many Australian hand surgeons have collaborated
with William Walsh’s biomechanical laboratory at the
University of New South Wales in Sydney. Dona and
Gianoutsos, along with Walsh and others, investigated
the ideal size bite of locking loops for cruciate core
repairs, finding that each loop should optimally include
25% of the tendon’s width.22 In an earlier publication
they demonstrated the biomechanical superiority of an
interlocking horizontal mattress epitendinous suture
technique over that of an interlocking cross stitch, a
simple cross stitch, and a simple running repair.23 There
were statistically greater loads to failure, 2-mm gap for­
mation, and stiffness within those cadaveric tendons
repaired with the interlocking horizontal mattress
repair. Both studies used single progressive tensile
loading rather than cyclic loading when testing to
mechanical failure. Sandow and colleagues compared
the theoretical and actual repair strength of a multiple
strand repair in a single tension test with the strength
of a repair subjected to cyclic loading, finding a signifi­
cant decrease in ultimate tensile strength when the
repair was subjected to cyclical loading.24 This reduction
equated to the number of strands multiplied by the
strength of the knotted strand and was explained by the
change in stiffness of the knotted strand after cyclical
loading. They concluded that cyclic loading is more
representative of physiological loading after acute flexor
tendon repair and suggested that this should be the
testing model of choice in suture tenorrhaphy studies.
Matheson et al,25 again in collaboration with Walsh’s
Figure 13(G)-3  A, Modified Kessler core stitch. B, Inherent
gapping of modified Kessler under load. C, Adelaide
tenorrhaphy (four-strand single-cross grasp repair). (Adapted
from Sandow M, Kay S: Flexor tendon injuries. In Prosser R,
Conolly WB [eds]: Rehabilitation of the Hand & Upper Limb,
London, 2003, Elsevier Health Sciences, p. 47.)
A
B
C

165
Chapter 13G:  The Australian Experience
epitendinous repair applied to the palmar surface was
effective in improving the strength of repair compared
with a core suture alone and was effective in decreasing
gap formation. However, the load to failure was signifi­
cantly less than that of a core suture combined with a
full circumferential suture.
Hile and colleagues34 were concerned about the prev­
alence and consequences of piercing the thread of a
braided polyester suture with the needle during complex
multistrand repairs. They concluded that cutting needles
should not be used in tendon repairs because they can
damage or completely sever core sutures. They also
found that crushing of the suture between the teeth of
a needle holder significantly weakened the involved
segment.
The work of the investigators that was just outlined
is not inclusive of all such work that has been and is
being performed by Australian surgeons and their col­
leagues. However, the time and effort directed toward
obtaining optimal results from flexor tendon repair
within Australia and worldwide are indicative of the
importance of this pursuit. Although some thought
needs to be given to the worth of in vitro cadaveric
model experiments in which biology is removed from
the assessment, it is undeniable that such studies have
resulted in an improved understanding of the most
appropriate suture material, its gauge, and, importantly,
the surgical techniques and configurations of core and
epitendinous sutures that best prevent rupture, avoid
adhesions, and allow an optimal postoperative rehabili­
tation program. Despite these improvements, future
advances would appear to rely on our ability to modify
healing processes at a molecular level to increase early
strength of repair and to avoid adhesion formation.
CURRENT PRACTICE IN FLEXOR TENDON
REPAIR AND REHABILITATION
To understand the current state of primary flexor tendon
repair in Australia, we conducted a survey of the mem­
bership of the Australian Hand Surgery Society. Approx­
imately 40% of the active membership responded. A
very wide range of approaches was apparent, particu­
larly in the more controversial areas such as manage­
ment of the critical pulleys, and in methods of
postoperative rehabilitation.
In Australia, hand surgeons perform on average one
or two repairs a month (mean of 17 per year). Surgeons
with a large trauma practice perform substantially more.
Many other repairs are done by training residents and
by plastic, orthopedic, and general surgeons.
The most commonly used core suture material is
some form of braided polyester suture, typically Ticron
or Ethibond. For a large tendon, such as the FDP to the
middle finger, around two thirds prefer to use a 3-0 core
suture, but a significant minority prefer to use a 4-0 core
suture. Around 20% of the respondents who used a
a braided polyester suture. Load to failure in the modi­
fied Bunnell technique was superior to the micro-Mitek
anchor technique. Significant gap formation was present
when a monofilament suture was used but this problem
needed to be balanced against the difficulty of removal
of a braided polyester pullout suture. In the search for
alternatives to pullout sutures, with their potential for
creation of nail plate deformities, nail fold necrosis, and
infection, this group studied the influence of anchor
orientation when inserting the micro-Mitek anchor.29 In
an in vitro biomechanical study using 3-0 Ethibond
suture anchors in cadaver specimens, they did not find
any support for the theory that varying the angle of
anchor insertion improved the load to failure. However,
inserting the anchor in a retrograde fashion at 45°
appeared to result in fewer failures at the bone anchor
interface, with failure more likely to occur at the join
between suture and the eyelet of the anchor. Other sur­
geons in the same laboratory advised avoidance of nail
problems following reinsertion of FDP tendons distally
by placing the suture through two drill holes in the base
of the distal phalanx.30 They demonstrated that this
repair method was biomechanically sound with compa­
rable gap formation and load to failure to other com­
monly used repairs, was technically straightforward, and
did not impair the integrity of the distal phalanx.
In 2006, Dona and Walsh31 described a V-Y plasty
within pulleys to increase the cross-sectional area within
the pulley and thus decrease the friction of the repaired
tendon, allowing easier glide but maintaining a mechan­
ically sound pulley. They did not advocate this tech­
nique for routine use but as an alternative to venting or
resection in those circumstances in which gliding of the
repaired tendon was found to be unsatisfactory.
Much of the above experimental work relates to the
optimal types of suture material and techniques of core
and epitendinous sutures for flexor tendon repair. A
number of recent Australian studies have contributed to
an improved understanding of some other nuances of
flexor tendon surgery. Stewart and coauthors reminded
us of the ability of intact vincula to flex interphalangeal
joints despite tendon division.32 In an experimental
cadaveric model, they demonstrated that in the immedi­
ate postinjury period, the vinculum breve can hold a
divided tendon within a few millimeters of its insertion.
In these circumstances they quantified the motion
present at proximal and distal interphalangeal joints
as being 93% and 69% of normal, respectively. They
emphasized that testing of the injured digit against resis­
tance is important if one is to avoid missing the diag­
nosis of a tendon injury.
Lawson and coworkers addressed the difficulty of per­
forming a full circumferential epitendinous repair when
the tendon injury comes to lie beneath the A4 pulley
and access to the dorsal aspect of the tendon is com­
promised.33 They reported that a half-circumferential

166
Section 2:  Primary Flexor Tendon Surgery
Australian hand surgeons. Our current practice is sum­
marized as follows:
1. The approach to the flexor sheath begins with
débridement of the existing wound, followed by
extension using a combination of Bruner, hemi-
Bruner, and midlateral incisions.
2. The A2 and A4 pulleys are preserved during the
approach, with access to the tendon achieved via
Lister’s L-shaped windows between these critical
pulleys. This technique acts to funnel the tendon
beneath the pulley and maximize the chance of
smooth running of the repair.
3. If access to the tendon ends is adequate, a modi­
fied Savage four-strand (Adelaide) repair is per­
formed, using 3-0 Ticron in larger tendons and 4-0
or 5-0 Ticron in smaller tendons (usually little
finger FDP and pediatric tendons, respectively).
A simple running or Silfverskiöld epitendinous
repair using 6-0 Prolene with a taper needle is
performed, depending on access. The back wall
repair is ideally done first to keep the tendon ends
snugly opposed. If access to the back wall is par­
ticularly difficult or not possible because of the
proximity of the A4 pulley, only the front half of
the tendon receives an epitendinous suture.
4. If the distal tendon ends lies beneath the A4 pulley,
combined proximal/distal repairs as described by
Lister are used. This technique precludes the use
of the modified Savage repair in our hands, and
instead a modified Kessler with a supplementary
horizontal mattress core suture is performed. This
technique permits retention of the A4 pulley while
allowing for a 10-mm purchase with the core suture.
5. If gliding of the repair is impeded, the pulleys can
be stretched with pediatric urethral dilators. Small
strands of tendon projecting from between the
epitendinous sutures can be carefully removed
with microscissors.
6. The incised sheath is laid over the tendons but no
formal repair is performed if there is any threat to
gliding.
7. Postoperative rehabilitation is tailored to the
quality of the repair, and the compliance of the
patient, with a tendency to err toward a less aggres­
sive form of mobilization. In cooperative patients
with high quality repairs an active range of motion
technique is used; in the less optimal repair or
patient a modified Duran regimen with passive
flexion and active extension is prescribed. A dorsal
extension blocking splint is used to protect the
repair. In all cases it is particularly important to
guard against the development of a proximal
interphalangeal joint flexion contracture.
8. The splint is typically removed at 6 to 8 weeks, at
which point a gentle strengthening program is
braided suture as their core stitch preferred to use a
nonbraided suture for contaminated wounds.
There were two core suture configurations commonly
used: the four-strand modified Savage (often referred to
as an Adelaide repair) was used by 40% of the surveyed
surgeons and a four-strand modified Kessler was also
used by around 40%. Other favored techniques included
the six-strand Savage repair, and the Tsai-Lim six-strand
technique taught at Kleinert’s center in Louisville. Flex­
ibility in the choice of core suture is practiced by some,
depending on the ease of access to the cut ends of the
tendon around the pulley system.
The epitendinous suture of choice was polypropylene
(Prolene), with 60% of respondents preferring 6-0
Prolene and 25% favoring 5-0 Prolene. A simple running
epitendinous repair was performed by 70% of surgeons,
with the next most popular technique being the Silfver­
skiöld technique, used by around 20%.
Management of the A2 and A4 critical pulleys varied,
with forcefully held and disparate opinions; 12% of the
surveyed surgeons would not release any part of the
critical pulleys, while 88% would perform partial or full
releases. Of those who released the A4 pulley, 20%
would do so at the beginning of the repair as part of
their exposure; the remainder would do so if exposure
proved difficult during the repair or if the repaired
tendon did not glide adequately.
Two-thirds of respondents did not attempt to repair
divided pulleys, with some stating that they thought a
satisfactory repair was not possible; some of those who
would try to repair the pulley stated they would perform
a V-Y plasty to increase the diameter of the repaired
pulley. Closure of the synovial sheath/A3 pulley was
advocated by 55%.
Teaching of flexor tendon repair to trainees should
involve a sound grounding in basic principles, and 60%
of the surgeons surveyed believed, some very strongly,
that junior physicians should not be taught to release
the critical pulleys, believing they should first learn to
perform good repairs while working around the pulleys,
and perhaps when more expert would be able to make
an informed choice with respect to pulley release.
Rehabilitation protocols varied widely: 25% of
respondents used the same protocol for all of their
patients, while the other 75% tailored their approach to
the demands of each patient. The most popular regimen
was some form of immediate active motion, used by
65%. Passive flexion/active extension (Duran-Hauser)
was used by 19%, passive flexion/active hold (place and
hold) was used by 11%, and Kleinert traction was used
by 6%. Unrestricted use was allowed by 55% at 12
weeks, by 15% at 8 weeks and, by 15% at 6 weeks.
AUTHORS’ PREFERRED APPROACH
As shown, there is a considerable variety of approaches
to the management of flexor tendon injuries by

167
Chapter 13G:  The Australian Experience
by Rank, Morrison, and Pennington. The methods of
repair used by hand surgeons in Australia include use
of a four-strand core suture, careful consideration of
pulley management, and a trend toward active move­
ment in rehabilitation, but the precise methods vary
widely and are the subject of healthy debate.
started. Full grip is prohibited until 3 months post
repair.
SUMMARY
Australian surgeons have made important contributions
in the field of flexor tendon repair, including key work
References
1. Rank BK, Wakefield AR: Surgery of Repair as Applied to Hand
Injuries, Edinburgh, 1953, E & S Livingstone.
2. Pennington DG: The locking loop tendon suture, Plast Recon­
str Surg 63:648–652, 1979.
3. Mason ML, Allen HS: The rate of healing of tendons: An
experimental study of tensile strength, Ann Surg 113:424–
459, 1941.
4. Kessler I, Nissim F: Primary repair without immobilization
of flexor tendon division within the digital sheath. An experi­
mental and clinical study, Acta Orthop Scand 40:587–601,
1969.
5. Pennington DG: Atraumatic retrieval of the proximal end of
a severed digital flexor tendon, Plast Reconstr Surg 60:468–
469, 1977.
6. Sourmelis SG, McGrouther DA: Retrieval of the retracted
flexor tendon, J Hand Surg (Br) 12:109–111, 1987.
7. Sandow MJ: A further tendon retrieval trick, J Hand Surg (Br)
22:125–127, 1997.
8. Pribaz JJ, Morrison WA, Macleod AM: Primary repair of flexor
tendons in no-man’s land using the Becker repair, J Hand Surg
(Br) 14:400–405, 1989.
9. Callan PP, Morrison WA: A new approach to flexor tendon
repair, J Hand Surg (Br) 19:513–516, 1994.
10. Tonkin M, Lister G: Results of primary tendon repair with
closure of the tendon sheath, Aust N Z J Surg 60:947–952,
1990.
11. Singer D, Doi K, Morrison WA, et al: Comparative study of
the use of prefabricated bone tendon grafts and conventional
tendon grafts in flexor tendon reconstruction, J Hand Surg
(Am) 14:830–836, 1989.
12. Singer DI, Morrison WA, Gumley GJ, et al: Comparative study
of vascularized and nonvascularized tendon grafts for recon­
struction of flexor tendons in zone 2: An experimental study
in primates, J Hand Surg (Am) 14:55–63, 1989.
13. Honner R, Meares A: A review of 100 flexor tendon recon­
structions with prosthesis, Hand 9:226–231, 1977.
14. Tonkin M, Hagberg L, Lister G, et al: Post-operative manage­
ment of flexor tendon grafting, J Hand Surg (Br) 13:277–281,
1988.
15. Savage R: In vitro studies of a new method of flexor tendon
repair, J Hand Surg (Br) 10:135–141, 1985.
16. Strickland JW: Development of flexor tendon surgery: Twenty-
five years of progress, J Hand Surg (Am) 25:214–235, 2000.
17. Pulvertaft RG: Repair of Tendon Injuries in the Hand: Hunt­
erian Lecture delivered at the Royal College of Surgeons of
England on 6th February, 1948, Ann R Coll Surg Engl 3:3–14,
1948.
18. Boyer MI, Gelberman RH, Burns ME, et al: Intrasynovial
flexor tendon repair. An experimental study comparing low
and high levels of in vivo force during rehabilitation in
canines, J Bone Joint Surg (Am) 83:891–899, 2001.
19. Silva MJ, Brodt MD, Boyer MI, et al: Effects of increased in
vivo excursion on digital range of motion and tendon strength
following flexor tendon repair, J Orthop Res 17:777–783,
1999.
20. Strick MJ, Filan SL, Hile M, et al: Adhesion formation after
flexor tendon repair: A histologic and biomechanical com­
parison of 2- and 4-strand repairs in a chicken model, J Hand
Surg (Am) 29:15–21, 2004.
21. Sandow M, Kay S: Flexor tendon injuries. In Prosser R,
Conolly WB, editors: Rehabilitation of the Hand & Upper Limb,
London, 2003, Elsevier Health Sciences, pp 46–52.
22. 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.
23. Dona E, Turner AW, Gianoutsos MP, et al: Biomechanical
properties of four circumferential flexor tendon suture tech­
niques, J Hand Surg (Am) 28:824–831, 2003.
24. Gibbons CE, Thompson D, Sandow MJ: Flexor tenorrhaphy
tensile strength: reduction by cyclic loading: in vitro and ex
vivo porcine study, Hand (NY) 4:113–118, 2009.
25. Matheson G, Nicklin S, Gianoutsous MP, et al: Comparison
of zone II flexor tendon repairs using an in vitro linear cyclic
testing protocol, Clin Biomech (Bristol, Avon) 20:718–722,
2005.
26. Vizesi F, Jones C, Lotz N, et al: Stress relaxation and creep:
viscoelastic properties of common suture materials used
for flexor tendon repair, J Hand Surg (Am) 33:241–246,
2008.
27. Schreuder FB, Scougall PJ, Puchert E, et al: Effect of suture
material on gap formation and failure in type 1 FDP avulsion
repairs in a cadaver model, Clin Biomech (Bristol, Avon)
21:481–484, 2006.
28. Latendresse K, Dona E, Scougall PJ, et al: Cyclic testing of
pullout sutures and micro-mitek suture anchors in flexor digi­
torum profundus tendon distal fixation, J Hand Surg (Am)
30:471–478, 2005.
29. Schreuder FB, Scougall PJ, Puchert E, et al: The effect of mitek
anchor insertion angle to attachment of FDP avulsion inju­
ries, J Hand Surg (Br) 31:292–295, 2006.
30. Stewart DA, Smitham PJ, Nicklin S, et al: A new technique for
distal fixation of flexor digitorum profundus tendon, J Plast
Reconstr Aesthet Surg 61:475–477, 2008.
31. Dona E, Walsh WR: Flexor tendon pulley V-Y plasty: An alter­
native to pulley venting or resection, J Hand Surg (Br) 31:133–
137, 2006.
32. Stewart DA, Smitham PJ, Gianoutsos MP, et al: Biomechani­
cal influence of the vincula tendinum on digital motion after
isolated flexor tendon injury: A cadaveric study, J Hand Surg
(Am) 32:1190–1194, 2007.
33. Ansari U, Lawson RD, Peterson JL, et al: Effect of partial versus
complete circumferential repair on flexor tendon strength in
cadavers, J Hand Surg (Am) 34:1771–1776, 2009.
34. Ihsheish W, Smith BJ, Hile MS, et al: Suture handling reduces
suture strength. 2010 Australian Hand Surgery Society Annual
Scientific Meeting, March 17-20, 2010, Canberra, Australia.

168
Mark A. Rider, MBChB, FRCS, FRACS
H
The Wellington Experience
impinging on the tendon repair. The A2 and A4 pulleys
were preserved during suturing of the tendons. However,
subsequent venting was allowed if there was impinge­
ment of the repaired tendon during passive on-table
testing. There was a varying degree of consultant super­
vision, with most of the work done by trainees. The
rehabilitation protocol was unchanged throughout all
periods (“modified Belfast” early active, commencing at
48 hours). The main outcome measures were rupture
rate and range of movement by original Strickland at 12
weeks, assessed by the hand therapists in a standard
manner. After review of each period’s results, one or two
aspects of the repairs were changed, and the process
repeated. The results of these are summarized in Table
13(H)-1.
The first review was conducted in September 2000
and was retrospective over the previous 6 months. No
protocols existed in the unit at that time, resulting in a
wide range of repair and suture types, modified Tajima
with 3-0 and 4-0 Prolene predominating. The data were
markedly incomplete, but of the 10 patients who were
followed, 2 had rupture. The initial recommendation
was to start a prospective audit.
An 18-month prospective review was completed in
October 2002. The repairs were mainly modified Tajima
with 3-0 and 4-0 Ticron or Ethibond, and a simple
running epitendinous, on 37 patients with 49 tendons.
The follow-up rate was low, but of the 21 patients fol­
lowed up, 7 had rupture! The obvious recommendation
was to move to a multistrand repair. We selected a cross-
locked cruciate repair (Figure 13[H]-1).
The next review covered a 2-year period to August
2005. The repairs were all Adelaide with 3-0 or 4-0
Ethibond, plus a variable epitendinous repair, most
usually Silfverskiöld. The rupture rate plunged to 1 of
23 patients, with 82% excellent or good.
For the next audit, the repair type was left unaltered
but the suture material was changed to 4-0 Fiberwire
(Arthrex Naples, FL). This final review covered an
18-month period. There were no ruptures, and 96% of
fingers achieved Strickland excellent or good results.
DISCUSSION
Benefits of Serial Audit
I hope the improvement in our (initially appalling)
results is a convincing demonstration of the benefits of
OUTLINE
A series of prospective reviews is presented, showing
how stepwise changes in repair techniques lead to
dramatic improvements in outcomes. Our currently
preferred technique, and the rationale for it, is dis­
cussed. Some aspects of working as a hand surgeon in
New Zealand are mentioned. Finally, some guidelines
for preventing repair rupture are offered.
Multiple revolutions of the audit cycle are a power­
ful means of improving outcomes. The outcomes have
been improved drastically with progressive decreases
in the rupture rates of the tendons over the past 10
years with different surgical repair methods. Currently
the four-strand Adelaide repair with Fiberwire and
postoperative controlled active motion is a successful
combination in our practice.
“If you cannot measure it, you cannot improve it” is one
of many quotes attributed to Lord Kelvin. It is certainly
possible to measure the results of flexor tendon repair,
but this is not commonly done by surgeons. I submit
that measuring the results of our current practice, and
then making incremental changes based on the outcomes
and literature reviews, may yield great benefits for our
patients. Some surgeons rarely change their techniques,
ignoring important new developments; some change
constantly, following every fashion but having little idea
of their results. For me, a middle way has proved useful,
and I present a series of flexor tendon reviews to illus­
trate this.
THE AUDITS
The research method was the same for each period. All
adult patients presenting with completely divided flexor
tendons within zone 2 of the fingers were enrolled. The
study patients were treated in the same way as all other
patients with hand injuries who were admitted to the
unit; although the aim was to perform surgery on the day
of presentation, delays of up to 48 hours were common.
Both deep and superficial tendons were repaired. If
the flexor digitorum superficialis (FDS) was large enough
to accept a core suture, this was the same type as used
for the flexor digitorum profundus (FDP). Partial inju­
ries of critical pulleys were not repaired. The synovial
sheath was loosely repaired if this could be done without

169
Chapter 13H:  The Wellington Experience
The Nature of Hand Injuries and Their
Management in New Zealand
New Zealand has in the past had a much higher
incidence of occupational injury than the safest of
the Northern European countries, although rates are
improving. The occupations with the highest rate of
hand injuries are machine operators, construction, for­
estry workers, and meat processing workers.5 The last
two occupations are relatively common in New Zealand
compared to more industrialized countries and may in
part explain the high numbers of hand injuries. Perhaps
surprisingly considering the high cost of these injuries,
etiological and preventable factors for work-related
hand injuries are poorly studied worldwide.6 In our
experience, meat processors were the group with the
highest work-related cause of flexor tendon injury. Pos­
sible risk factors include blunt knives, removing protec­
tive gloves, the cold environment, and the high turnover
rate of staff. From a surgeon’s point of view, the injuries
were usually sharp and tidy, although they carried a risk
of infection from raw meat.
Targets
An essential part of audit is to identify best practice from
the literature. Establishing a realistic target for rupture
rates is not a simple matter. During the 1990s, rates of
well over 4% were regularly reported. It is difficult to
judge these. On the one hand, they came from well-
known centers and could perhaps be presumed to be
best practice. On the other hand, many of the larger
series were from British public hospitals and undoubt­
edly included the results of unsupervised junior
surgeons.
In reviewing more recent publications, we can con­
clude that rupture rates from two-strand repairs are not
improving with time. However, there are a number of
reports of multistrand repairs with very low rupture
rates, although with small numbers of patients. One of
the largest recent reports of four-strand repairs gives a
headline rupture rate of 2%.7 If we extract from the
authors’ results only the zone 2 repairs, the rate rises
to 3.7%. Almost all the repairs were performed by
trainees.
serial audits. Much of the improvement was no doubt
due to the change to a four-strand repair. However, I
think there are other, less tangible benefits, particularly
the greater involvement by senior surgeons and the
general focus on quality and excellence by the whole
unit.
The Change to a Four-Strand Repair
We anticipated that the selection of the cross-locked
cruciate repair (known locally as the Adelaide repair)
would provide a good compromise between strength
and difficulty of insertion. Our unit is typical of many
large public hospital services in that many of the flexor
tendon repairs are performed by unsupervised, rela­
tively junior trainees. More complex repair patterns may
be stronger when performed by experts, but in our “real
world” situation are also more likely to be poorly
inserted. Studies confirming the favorable properties of
the Adelaide repair are now starting to appear.1
Fiberwire
As we completed our penultimate audit, we became
interested in Fiberwire. Some early reports attested to its
high strength compared to preexisting suture materials,
Fiberwire showing less gap formation at all loads and
the greatest load to failure. A subsequent study con­
firmed its strength with locking repairs.2 Some of the
benefit may be illusory due to its nonstandardized size:
a 4-0 Fiberwire may be much stronger than a 4-0 Ticron,
but it is also thicker.3 We remain happy to use Fiberwire,
but note some recent studies raising caveats, in particu­
lar poor knot holding requiring more throws (and
therefore greater bulk of suture material).4
Figure 13(H)-1  Four-strand locked cruciate repair method.
Table 13(H)-1  The Results of Tendon Repairs for Each Period Reviewed
Year
Patients
Fingers
Tendons
Patients
Lost to
Follow-Up
Fingers
Followed-Up
Fingers
Good or
Excellent (%)
Ruptures
Patients
Fingers
Rate (%,
Fingers)
2000
15
21
26
5
15
7
2
3
20
2002
37
42
49
16
23
43
7
7
33
2005
33
41
61
10
28
82
1
1
4
2007
32
39
43
10
28
96
0
0
0

170
Section 2:  Primary Flexor Tendon Surgery
It seems likely that ruptures will continue for the
foreseeable future, despite improvements in repair types
and ongoing research into biological factors. We cer­
tainly do not expect our rupture rate to remain at zero.
Patient irresponsibility is a factor in many ruptures.8 In
our series, we saw early ruptures in one patient playing
competitive rugby and another using a chainsaw having
first climbed a ladder. These incidents occurred despite
presumably adequate education from an expert and
well-staffed hand therapy service.
I suggest that a specialist hand unit, with some repairs
performed by trainees, should obtain a rupture rate in
zone 2 of less than 4% and Strickland good/excellent
result of greater than 80%. An experienced hand surgeon,
reviewing his or her own repairs, should aim for a
rupture rate approaching zero and good/excellent results
in greater than 90%.
CONCLUSION
We have shown how multiple revolutions of the audit
cycle are a powerful means of improving outcomes. We
believe this should be a mandatory activity in every
hand unit. The Adelaide repair Fiberwire and controlled
active motion is, for us, a successful combination.
References
1. Croog A, Goldstein R, Nasser P, et al: Comparative biome­
chanic performances of locked cruciate four-strand flexor
tendon repairs in an ex vivo porcine model, J Hand Surg (Am)
32:225–232, 2007.
2. Miller B, Dodds SD, deMars A, et al: Flexor tendon repairs:
the impact of fiberwire on grasping and locking core sutures,
J Hand Surg (Am) 32:591–596, 2007.
3. Scherman P, Haddad R, Scougall P, et al: Cross-sectional area
and strength differences of fiberwire, prolene, and ticron
sutures, J Hand Surg (Am) 35:780–784, 2010.
4. Waitayawinyu T, Martineau PA, Luria S, et al: Comparative
biomechanic study of flexor tendon repair using Fiberwire,
J Hand Surg (Am) 33:701–708, 2008.
5. Burridge JD, Marshall SW, Laing RM: Work-related hand
and lower-arm injuries in New Zealand, 1979 to 1988, Aust
N Z J Pub Health 21:451–454, 1997.
6. Sorock GS, Lombardi DA, Courtney TK, et al: Epidemiology
of occupational acute traumatic hand injuries: A literature
review, Saf Sci 38:241–256, 2001.
7. Caulfield RH, Maleki-Tabrizi A, Patel H, et al: Comparison of
zones 1 to 4 flexor tendon repairs using absorbable and
unabsorbable four-strand core sutures, J Hand Surg (Eur)
33:412–417, 2008.
8. Harris SB, Harris D, Foster AJ, et al: The aetiology of acute
rupture of flexor tendon repairs in zones 1 and 2 of the fingers
during early mobilization, J Hand Surg (Br) 24:275–280,
1999.

171
CHAPTER
14 
PARTIAL TENDON
LACERATIONS
Morad Askari, MD, and Peter C. Amadio, MD
on complete laceration of flexor tendons with much
smaller focus on partial tendon laceration.
The controversy centered on the repair of partial
tendon laceration was initiated by the work of Wray and
Weeks2 in a chicken model where nonrepaired partially
lacerated flexor tendons were deemed better than those
repaired. This result was similarly confirmed by Reyn­
olds et al3 in a similar model. Just a few years later, the
concept of not repairing partially lacerated tendons was
challenged by Kleinert and others.4 They advocated that
partially lacerated tendons should be treated similarly
to complete lacerations, emphasizing that this type of
injury required the same diligence in surgical repair and
postoperative care. In their series, they reported signifi­
cant rise in complications following nonrepair of partial
flexor tendon lacerations. Similarly, around the same
time, Janecki5 reported on two cases of entrapment fol­
lowing nonoperative treatment of partially lacerated
flexor tendons. The sentiment to repair this type of
tendon injury was further echoed by Schlenker and
coworkers,6 who reported a significant increase in trig­
gering, entrapment, and rupture among other complica­
tions when these lacerations were not repaired. They
recommended the repair of all lacerations greater than
25%. They favored modified Kessler core suture tech­
nique along with a running epitendinous suture for
lacerations greater than 50% while for those between
25% and 50%, a running braided polyester suture was
suggested. For laceration less than 25%, excision of the
distally based flap was recommended to avoid triggering
or entrapment in the pulley. However, since these
studies, a series of in vitro biomechanical studies along
with in vivo animal studies has resulted a better under­
standing of the properties of partially lacerated tendons
and a shift in paradigm of treatment of partial tendon
injury.
DIAGNOSIS AND PHYSICAL EXAMINATION
Clinically most patients have normal active motion, and
cutaneous wounds are either rather small or healed.
Identifying partial tendon laceration is a challenge.
Partial tendon laceration classically presents as painful
and weakened tendon excursion as opposed to absence
OUTLINE
Many patients with partial tendon laceration come
to the clinic several weeks after the injury due to
persistence of symptoms, by which time the cutane­
ous wound may be healed and closed. Evidence of
triggering or entrapment is adequate indication for
exploration of the wound. Partial tendon lacerations
can also be seen in a large wound together with com­
plete lacerations of other tendons, which are easy to
diagnose.
The degree of laceration directly impacts tensile
strength of the injured tendon. However, with even
larger degrees of laceration, the tendon retains enough
tensile strength to withstand most active motion. The
volar location of the laceration may lower a clinician’s
threshold for surgical repair. A lesser degree laceration
(<40% to 50%) may not need surgical repair. A tendon
laceration >50% to 60% usually needs surgical repair.
Repair of partial tendon lacerations decreases the
chance of tendon triggering and improves tendon
excursion. The partially lacerated tendon can be treated
with a circumferential suture with or without core
suture technique. Core sutures are used in large degree
(>60% to 75%) partial laceration. The lacerated parts
can be trimmed to reduce triggering in partial tendon
lacerations up to 75%. Complications of untreated
lacerations include triggering, entrapment, and rupture
of the tendons.
While little controversy exists today on necessity of
complete tendon laceration repair, the topic of treating
partial tendon laceration remains a subject of debate in
hand surgery. Opinions are divided on whether man­
agement of partial tendon laceration is best achieved
through surgical or nonsurgical approach.1 This con­
troversy, in part, stems from fewer in vivo studies of
partial as opposed to complete tendon laceration in the
literature many with conflicting results. In part, this
could also be due to fewer cases of partial tendon lacera­
tion properly diagnosed in clinical practice. Thus, major­
ity of studies have used in vitro cadaveric or animal
models. Additionally, a majority of these studies focus

172
Section 2:  Primary Flexor Tendon Surgery
tendon with 30% laceration had histological evidence
of amorphous connective tissue in the tendon gap with
nascent collagen fibers bridging across the tendon edges,
the 60% and 90% lacerated tendons showed little evi­
dence of healing with no collagen present in the gap at
35 days.
BIOMECHANICAL PROPERTIES OF PARTIAL
TENDON INJURY
The majority of experimental data on partial tendon
injuries is the result of years of investigation using in
vitro cadaveric and animal tendons in addition to a
smaller number of in vivo studies in animal models.
The popular animal models for their proximity to
human tendon anatomy are the canine, chicken, sheep,
and pig models. Majority of effort in understanding
partial tendon injury has been associated with flexor
tendons specifically in zone 2.
Gap formation in the context of partial tendon injury
is an important point of clinical interest because its
increase is associated with higher gliding resistance, trig­
gering, and delay in healing.14,15 Gapping beyond 2 mm
is harmful to tendon healing.14
Tensile Strength and Cross-Sectional Area
Feared sequelae of improperly treated partial tendon
laceration include triggering, loss of entrapment, adhe­
sion formation, or late rupture. It is intuitive that partial
lacerated tendons would be weaker than intact tendons.
Additionally, some had suggested formation of a scar
mass at the laceration sites if a partial injury were to go
nonrepaired, thus adversely affecting tendon function.4
It is our presumption that many such tendons if they
were to be repaired would like be subjected to more
rigorous immobilization postrepair than if they were
left alone. In the following sections, the evidence that
has shaped our current approach to partial tendon
injury is reviewed. Of note, the great focus in study
partial tendon lacerated has involved mainly flexor
tendons and attention specifically to zone 2 injuries in
part due to the historical poor outcome of injury in this
area. Many times, the clinicians inadvertently extrapo­
late this information to treatment of flexor tendon in
other zones and to extensor tendon injuries.
In the recent decades, several studies have pointed
toward early active finger motion to decrease adhesion
formation and improve tendon gliding.16-18 While
rupture of partially lacerated tendons has been dis­
cussed as a complication of early motion and no repair,
many have shown this to be a more minimal concern
than previously thought. In a series of 34 partial tendon
lacerations ranging from 25% to 95% laceration of the
cross-sectional area of the tendon, no rupture occurred
following non surgical management even in the 11 cases
with more than 75% laceration. All but one case had
good to excellent function.19 Several studies have
of active finger flexion or any excursion as would be
expected with complete laceration. However, without
direct visualization through an open wound or an
imaging modality, determining the degree of laceration
based on clinical exam is not possible. In majority of
cases, an opening in the skin at the site of trauma can
serve as a window for initial examiner to characterize
the degree of tendon laceration as a percentage of intact
tendon width. This measurement is commonly done
using a caliper. Interestingly, it has been showed that
neither caliper measurement nor estimation with naked-
eye provides consistent and reliable means of assessing
the degree of laceration when managing partial tendon
injuries.7 Yet, these are tools used by most in estimating
degree of tendon injury.
Many patients with partial tendon laceration may
present several weeks after the injury due to scheduling
or persistence of symptoms, by which time the wound
may be healed and closed. Evidence of triggering or
entrapment is adequate indication for exploration of the
wound. If a moderate index of suspicion exists for explo­
ration, an imaging study such as ultrasound or magnetic
resonance imaging is beneficial. Partial tendon lacera­
tion can also occur with other tendons lacerated com­
pletely in a larger wound, which is easy to diagnose.
PHYSIOLOGY OF TENDON HEALING
Through significant contribution early in the 20th
century by Mason, Allen, and Shearon8,9 an understand­
ing of various stages of tendon healing was developed.
Following injury, the tendons enter an exudative phase
of tendon healing (0 to 14 days) followed by a reparative
phase (15 to 35 days). An understanding of the phases
in tendon healing is important in choosing appropriate
time point for biomechanical and biological studies. The
healing tendon is strongest at day 1 with this strength
dropping by day 5. The lost strength is gradually restored
starting around day 15 and continues until the end of
the reparative phase. The exact length of time for the
tendon to reach maximum strength appears to extend
beyond the 5th week. The above authors also demon­
strated that the weakest sight on an undamaged tendon
corresponded to the musculotendinous junction and
the tendon–bone insertion. This group was among the
first to suggest that early motion resulted in increase in
strength and function of partially lacerated tendons. In
addition to tensile strength, the concepts of tendon
gliding and resistance against motion were emphasized
in the context of tendon repair. Many have shown the
deleterious effects of immobilization on tendons and
the subsequent change in substance constituents and
weakening of their insertion sites.10-12
The degree of laceration has an effect on tendon
healing. In an in vivo canine model of partial flexor
tendon injury, Cooney and others13 noted that the
greater lacerations had less evidence of healing. While

173
Chapter 14:  Partial Tendon Lacerations
Early Mobilization and Tendon Remodeling
As hesitations to mobilize patients with partial tendon
lacerations due to risk of rupture are lessened through
the mentioned studies, early mobilization was advo­
cated to increase tensile strength and to improve mor­
phology of the nonrepaired partially lacerated tendons.
Wray and others26 showed that early motion accelerates
return of tensile strength to partially lacerated tendons.
Similarly, Reynolds and others3 showed that immobi­
lized repaired tendons were weaker than nonrepaired
partially lacerated tendon following unrestricted
motion. This finding may be due to a more pronounced
intrinsic healing with early motion.27 Using an in vivo
canine model, Bishop and others24 studied the relative
effects of immobilization, early protected mobilization,
tendon repair, and no repair by paired comparison to
contralateral side. Early motion improved stiffness and
excursion significantly. Under scanning electron micros­
copy, these tendons had more nearly normal morphol­
ogy compared with immobilized tendons. They also
found a decrease in breaking strength and stiffness in
tendons that had undergone repair compared with no
repair. They concluded that tendon lacerations up to
60% of the cross-sectional area of the tendon should be
treated with early protected mobilization with no
attempt to repair.24 Kubota and others,28 in their in vivo
chicken model, documented that early motion and
tension on partial tendon injuries resulted in superior
biological remodeling. Grewal and others29 compared
canine tendon healing 3 weeks after passive to active
rehabilitation following 60% tendon laceration. The
tendons were repaired with modified Kessler suture in
one and were left nonrepaired in the other group. In
each group, a subgroup underwent immediate active
mobilization while the other subgroup underwent
passive motion. No difference was found in tendon
excursion, stiffness, or load to failure between passive
versus active mobilization or repair versus nonrepair
groups. However, gapping was significantly increased
among the active rehabilitation group compared to the
passive rehabilitation group. The authors concluded
that active rehabilitation is safe in lacerations up to
60%, but gapping at 3 weeks raises the concern for
future complication in tendon healing.29
Tendon Gliding and Resistance
Unresisted gliding is important for normal tendon func­
tion. Proponents of surgical repair of partial tendon
laceration have argued that presence of flaps at the non­
repaired site may result in triggering, entrapment in the
pulley, and eventual rupture of the tendon.4-6 However,
following surgical repair, the bulk of surgical repair and
rough tendon surfaces can result in higher gliding resis­
tance. This in turn may adversely affect the rehabilita­
tion process.30,31 In an in vivo chicken study, Ollinger
and others24 showed not only a decrease in tensile
characterized the strength of partially lacerated tendon.
While the tensile strength of the tendon decreases as the
cross-sectional area of laceration increase, these studies
support that a partially lacerated tendon in most cases
retains sufficient tensile strength to withstand motion
and loading. In chickens, Reynolds and others3 showed
the mean tensile strength of unsutured tendons to be
higher than the sutured repairs at both 2 and 4 weeks
of unrestricted motion after surgery. Dobyns and others20
demonstrated that 30% of lacerated tendons retain 80%
of their original strength, while 75% of lacerations
maintain 40% and 90% of lacerations have 25% of the
strength. Hariharan and others7 looked at the tensile
strength of 50% and 75% volarly lacerated flexor
tendons in cadavers and compared the failure loads to
in vivo forces measured in human flexor tendons during
unresisted active finger movement. Failure loads for
50% lacerated tendon were almost twice as much as
the failure load for 75% lacerated ones. However, the
failure values for both degrees of laceration far exceeded
the in vivo values required for unresisted active finger
movement.21
McCarthy and others7 noted a tendency for tendons
to fail at the site of partial laceration when the cross-
sectional area of laceration was equal or greater than
60% in an in vitro canine model. At this degree of lac­
eration, they noted a 22.8% decrease in stiffness, 41.5%
decrease in failure loads, 15.6% decrease in percent
elongation, and 56.2% decrease in energy absorbed
compared with intact tendons. The structural properties
change adversely with increasing degree of laceration.7
As inflicted lacerations through tendons are most
commonly not transverse, the effect of the direction of
laceration and its impact on tensile strength becomes
important. Tan and others22 have shown that obliquity
of tendon laceration affects the strength of partially lac­
erated tendons. Using pig tendons lacerated to 90% of
their diameter, they found that lacerations with 45° and
60° had significantly less ultimate tensile strength com­
pared with transverse, 15° or 30° oblique laceration.22
Tensile Strength and Repair
Several studies have investigated the effect of suture
repair and subsequent immobilization on tensile
strength of partially injured tendons. Suturing may
impair the vascularity of the tendon and damage tendon
cells.23 Ollinger and others24 found that sutured tendons
(Bunnell method) after partial laceration had less tensile
strength than nonrepaired controls. Similar results were
replicated by Bishop and others.25 In a canine model,
areas of tendon adjacent to the repair site had necrosis,
which explains the diminished tensile strength.25
Cooney and others12 showed in a similar model, that
for 30%, 60%, and 90% lacerated canine flexor tendons,
nonrepaired tendons had a greater mechanical strength
at 14 and 35 days compared with the repaired tendons.

174
Section 2:  Primary Flexor Tendon Surgery
results have resulted in a general consensus among clini­
cians to repair tendon laceration of 50% and greater
based on a survey by McCarthy and others.1 Tendon
retains adequate tensile strength despite greater than
50% laceration of their cross section. Yet, structural
properties of the tendon are detrimentally affected with
greater laceration. Tan and others39 have found benefi­
cial effect to peripheral suture repair versus no repair of
60% to 90% lacerated pig tendons. They noted a signifi­
cant increase in load to failure and increase in gap for­
mation forces with this type of repair. On the other
hand, Boardman and others40 found no significant dif­
ference in repair versus nonrepair of a 70% canine
tendon laceration. Stahl and others41 found no signifi­
cant benefit to repair of the tendon lacerations up to
75% in children versus no repair.
Peripheral Suture or Core Suture Repair
Using an in vitro sheep model, Haddad and others42
studied the effect of repair versus nonrepair on 75%
lacerated flexor tendons. Repairs included a group with
only peripheral sutures and a group with peripheral
sutures plus a core sutures. After cyclic loading, the
nonrepaired group had 2mm or greater gap formation,
while either repair resulted in 1mm or smaller gap with
no significant difference between the two repair groups.
Load to failure values between the two repair groups
were also not different. Using cadaveric tendons with
75% laceration, Zobitz and others43 compared different
core sutures combined with a peripheral suture to
peripheral sutures alone (Figure 14-1). With static
testing, they did not find any difference in gap forma­
tion among the groups though an increase was recorded
in load to failure and stiffness among the repairs with
core sutures.43 Neither of these two studies found a
mechanical disadvantage to having a core suture on the
integrity of the tendon. This is in contrast to earlier
studies looking at lesser laceration injuries where core
suture weakened the tendon.24,25 Thus, it can be sug­
gested that high degree lacerations (>70%) benefit from
repair. We recommend surgical repair with a large degree
of partial laceration.
Another concern with repairing partial tendon lacera­
tion is its effect on gliding resistance. Presence of suture
was shown to increase tendon gliding resistance by four
times that of intact tendons in zone 2.34 In high-degree
laceration where a repair may be advocated, it is impor­
tant to use a suture technique that minimizes tendon
gliding resistance. As discussed previously, Zhao and
others34 had demonstrated in 80% lacerated tendons,
repair with modified Kessler technique plus a peripheral
suture results in least resistance compared with other
repairs in cadaveric tendons. Since adhesion formation
is related to gliding resistance, it is intuitive that the
suture techniques with lesser resistance will result in
fewer adhesions. In the same model, Zhao and others45
strength but also a decrease in tendon gliding following
exposure and tenorrhaphy of partial tendon lacerations.
Early postoperative motion can restore smooth tendon
surface and improve gliding.32 Al-Qattan and others33
studied triggering and tendon flap formation over an
8-week period in an in vivo sheep model with 50%
lacerated flexor tendons without surgical repair. Trigger­
ing occurred not due to bulbous scar formation but due
to “bunching” of the tendon fibers proximal and distal
to the area of laceration. With time and mobilization,
these fibers became incorporated into the healing
process resulting in spontaneous resolution of trigger­
ing in majority of tendons.33 Various suturing tech­
niques result in different degree of friction and resistance
at tendon–pulley interface. Zhao and others34 character­
ized gliding resistance with several suture techniques on
the 80% lacerated tendons using cadaveric tendons.
Modified Kessler repair had the least resistance com­
pared to Kessler, Savage, Tsuge, or Becker techniques.
Yet, on average, there was an average increase of 1.08 N
in gliding resistance of the tendon following repair,
which is sufficient to limit tendon gliding during passive
motion (average 0.49 N).34
Location as well as the degree of laceration of the
tendon also impacts the gliding resistance. Erhard and
others35 compared volar and laterally transected tendons
in an in vitro cadaver model. The degree of laceration
in two different groups was 50% and 75%. Increase in
laceration resulted in greater gliding resistance. More
interestingly, volar partial laceration resulted in greater
gliding resistance in comparison to lateral transections
for both 50% and 75% lacerations.35
Tendon Nutrition and Adhesion Formation
Formation of adhesion limits gliding ability of the
tendon. Aside from potentially creating large physical
mass that impedes proper excursion of the tendon at
the pulley, the act of tendon repair may result in more
inflammation and adhesion formation. Suture repair
and tendon immobilization contribute to adhesion
formation.36 Chow and Yu37 observed more extensive
adhesion and decrease in tendon gliding when unre­
paired incomplete tendon lacerations of greater than
50% were immediately moved, compared with those
repaired and immobilized.
SURGICAL REPAIR OF PARTIAL
TENDON INJURIES
Surgical Repair or Nonrepair
The exact extent of the lacerated tendon cross section
that warrants repair remains unknown. Schlenker and
others6 had recommended repairing laceration greater
than 50% with modified Kessler and running peripheral
suture. Balk and others38 recommended similar repair
for a slightly higher cutoff of 60% laceration. These

175
Chapter 14:  Partial Tendon Lacerations
reported positive result from trimming partial lacera­
tions of some greater than 50%,. Erhard and others35
showed that in cadaveric tendons lacerated to 50%
and 75%, trimming tendon edges would result in less
gliding resistance than circumferential running repair
(Figure 14-2). Interestingly, both trimming and running
repair resulted in greater gliding resistance at 50% lac­
eration compared to no repair. This relationship changed
in fair of trimming or running repair at 75% volar lac­
erations compared to no repair. Trimming is indicated
for minor partial laceration as well, to reduce the chance
of triggering.
POSTOPERATIVE REHABILITATION
Protected early mobilization was shown to benefit
outcome of tendon repair including repair of partial
injuries. Zhao and others46 have pointed to the impor­
tance of the synergistic wrist motion in postoperative
rehabilitation following partial flexor tendon lacera­
tions. Following repair of 80% lacerated flexor tendons
in their in vivo canine model, gliding excursion of the
group that underwent synergistic wrist and finger motion
therapy was significantly greater than the group in which
the tendon were rehabilitated with wrist fixation (45°
flexion) and finger motion at 1, 3, and 6 weeks. Addi­
tionally, synergistic wrist motion has been shown to
result in significantly less severe adhesions and less
adhesion breaking strength at weeks 3 and 6.47
COMPLICATIONS
In management of partial tendon injuries, the incidence
of complications is related to the degree of laceration of
the cross-sectional area of the tendon. The prevalent
have shown at 3 and 6 weeks, the adhesion breaking
strength was significantly lower in tendon repaired with
modified Kessler core suture technique.
Direction of tendon laceration is an important factor
in consideration. Tan and others22 had shown that as
the direction of laceration becomes more oblique
(≥45°), the tensile strength of the tendon is decreased
for the same cross-sectional area of laceration. They
found a peripheral tenorrhaphy to significantly increase
gap formation forces and load to failure in such lacera­
tion. Therefore, this may lead one to consider surgical
repair for oblique laceration with smaller cross-sectional
area than a transverse laceration. The same group has
also shown that an oblique direction to the laceration
can weaken the conventional modified Kessler or cruci­
ate repair but their mechanical strength was strength­
ened by orientating the repair strands parallel to the
laceration.45 Thus, the orientation of the laceration
needs particular attention when choosing a core suture.
Trimming of the Partially Lacerated Tendons
An alternative to surgical repair of partial tendon lacera­
tions (≤75%) is trimming. The approach has been advo­
cated to be the more superior treatment even in light of
higher-degree laceration. Schlenker and others6 recom­
mended trimming beveled edges of tendons with less
than 25% of lacerations, while Al-Qattan and others33
Figure 14-1  Six popular core suture techniques for repair
of partially lacerated flexor tendons. Modified Kessler
technique was noted to have the lowest gliding resistance.
Kessler
Modified Kessler
Augmented Becker
Savage (modified)
Lee (modified)
Tsuge
Figure 14-2  Trimming of partial tendon injury. In an effort
to prevent bulky scar formation and provide smooth gliding,
the lacerated edges of the tendon are trimmed as shown.
This technique is as effective as suture repair in reducing
gliding resistance of laceration up to 75%.
Excursion
V
V`

176
Section 2:  Primary Flexor Tendon Surgery
the method of tendon repair and the motion regimen.48
With or without core suture technique, adding circum­
ferential sutures results in equivalent results when the
laceration is less than 75%. We recommend core suture
repair of the greater degree of partial laceration. Trim­
ming can be as effective as suture repair in reducing
gliding resistance of laceration up to 75%. Early passive
or active motion is recommended after surgery.
Over the past 50 years, the paradigm of management
of partial tendon laceration has shifted thanks to the
active work, observations, and investigations of many
astute clinical scientists. We now have a wealth of infor­
mation regarding the biomechanical properties tendons
in the context of various degrees of laceration, different
modes of repair, and rehabilitation. Yet, many more
years of research is necessary to create a consensus in
different aspects of partial tendon injuries.
Figure 14-3  Complications of nonrepair of partial tendon
laceration include entrapment, triggering or flapping, or
ultimately rupture of the tendon. This rate has been related
to the extent of laceration, direction, or location of the injury
on the tendon.
COMPLICATIONS
Tag or flap
Entrapment
Flapping
Rupture
Tendon sheath
Figure 14-4  Triggering in a patient sustained a laceration
over his left thumb metacarpophalangeal joint crease
3 weeks after he was treated with skin closure and
discharged from emergency department. He subsequently
had triggering and persistent pain at the A1 pulley site.
A, Intraoperatively, tendon laceration was found smaller
than 50%, with a reactive scar mass at the site of injury.
B, The mass along with lacerated edges of the tendon
were trimmed. Patient was asked to flex the thumb
intraoperatively to assure resolution of triggering at the
pulley. The pulley was left intact.
A
B
complications are triggering, entrapment, and rupture
(Figures 14-3 and 14-4). Interestingly, for lesser lacera­
tion, complications are more significant when repaired,
while for greater laceration, complications are more fre­
quent when no repair is performed. The exact cutoff in
the degree of the laceration between these groups again
is not exactly known. Cooney and others13 demon­
strated a higher complication rate with repair of 30%
and 60% lacerated flexor tendons but lower incidence
of complications with repair of 90% laceration flexor
tendons in a dog model at 35 days.
SUMMARY
Partially lacerated tendon has been shown to be stron­
ger than previously assumed. The degree of laceration
directly impacts tensile strength of the injured tendon.
Yet, with even large degrees of laceration (<75% lacera­
tion), the tendon retains enough tensile strength to
withstand most active motion. The degree of obliquity
and the volar location of the laceration are features that
may lower a clinician’s threshold for surgical repair of a
lesser degree laceration.
Many have pointed out the benefits of surgical repair
for large partial lacerations (>60% to 75%). Repair in
these cases will improve tendon excursion, lower the
resistance during rehabilitation, and protect the integrity
of intact tendon portion. Improving postoperative
smoother tendon gliding surface and avoiding triggers
are the major reasons for repair of the partially cut
tendon rather than restoring tensile strength, as the
tendon usually retains sufficient strength, but the cut
portion can be entrapped during tendon motion. The
entrapped partially cut tendon may be torn entirely
or resist tendon motion. Early active motion and syner­
gistic wrist motion are beneficial for proper healing
and remodeling of the tendon and lower the chance for
complications. Postoperative tendon gliding depends on

177
Chapter 14:  Partial Tendon Lacerations
22. Tan J, Ming L, Jia ZJ, et al: Repairs of partial oblique tendon
injuries: A biomechanical evaluation, J Hand Surg (Br) 29:
381–385, 2004.
23. Wong JK, Cerovac S, Ferguson MW, et al: The cellular effect
of a single interrupted suture on tendon, J Hand Surg (Br)
31:358–367, 2006.
24. Ollinger H, Wray RC Jr, Weeks PM: Effects of suture on tensile
strength gain of partially and completely severed tendons,
Surg Forum 26:63–64, 1975.
25. Bishop AT, Cooney WP 3rd, Wood MB: Treatment of partial
flexor tendon lacerations: the effect of tenorrhaphy and early
protected mobilization, J Trauma 26:301–312, 1986.
26. Wray RC Jr, Ollinger H, Weeks PM: Effects of mobilization on
tensile strength of partial tendon lacerations, Surg Forum
26:557–558, 1975.
27. Gelberman RH, Woo SL, Lothringer K, et al: Effects of early
intermittent passive mobilization on healing canine flexor
tendons, J Hand Surg (Am) 7:170–175, 1982.
28. Kubota H, Manske PR, Aoki M, et al: Effect of motion and
tension on injured flexor tendons in chickens, J Hand Surg
(Am) 21:456–463, 1996.
29. Grewal R, Saw SSC, Varitimidus S, et al: Evaluation of passive
and active rehabilitation and of tendon repair for partial
tendon lacerations after three weeks of healing in canines,
Clin Biomech (Bristol, Avon) 21:804–809, 2006.
30. Coert JH, Uchiyama S, Amadio PC, et al: Flexor tendon-
pulley interaction after tendon repair: A biomechanical study,
J Hand Surg (Br) 20:573–577, 1995.
31. Lieber RL, Silva MJ, Amiel D, et al: Wrists and digital joint
motion produce unique flexor tendon force and excursion in
the canine forelimb, J Biomech 32:175–181, 1999.
32. Gelberman RH, Vande Berg JS, Lundborg GN, et al: Flexor
tendon healing and restoration of the gliding surface: An
ultrastructure study in dogs, J Bone Joint Surg (Am) 65:70–80,
1983.
33. al-Qattan MM, Posnick JC, Lin KY: Triggering after partial
tendon laceration, J Hand Surg (Br) 18:241–246, 1993.
34. Zhao C, Amadio PC, Zobitz ME, et al: Gliding resistance
after repair of partially lacerated human flexor digitorum pro­
fundus tendon in vitro, Clin Biomech (Bristol, Avon) 16:696–
701, 2001.
35. Erhard L, Zobtiz ME, Zhao C, et al: Treatment of partial lac­
erations in flexor tendons by trimming. A biomechanical in
vitro study, J Bone Joint Surg (Am) 84:1006–1012, 2002.
36. de Klerk AJ, Jonck LM: Primary tendon healing. An experi­
mental study, S Afr Med J 62:276–281, 1982.
37. Chow SP, Yu OD: An experimental study on incompletely cut
chicken tendons—a comparison of two methods of manage­
ment, J Hand Surg (Br) 9:121–125, 1984.
38. Balk ML, Sotereanos DG: Partial flexor digitorum profundus
lacerations, Oper Tech Orthop 8:67–72, 1998.
39. Tan J, Wang B, Tan B, et al: Changes in tendon strength after
partial cut and effects of running peripheral sutures, J Hand
Surg (Br) 28:478–482, 2003.
40. Boardman ND 3rd, Morifusa S, Saw SS, et al: Effects of tenor­
raphy on the gliding function and tensile properties of par­
tially lacerated canine digital flexor tendons, J Hand Surg
(Am) 24:302–309, 1999.
41. Stahl S, Kaufman T, Bialik V: Partial lacerations of flexor
tendons in children: primary repair versus conservative treat­
ment, J Hand Surg (Br) 22:377–380, 1997.
42. Haddad R, Sherman P, Peltz T, et al: A biomechanical assess­
ment of repair versus nonrepair of sheep flexor tendons
lacerated to 75 percent, J Hand Surg (Am) 35:546–551,
2010.
References
1. McCarthy DM, Boardman ND, Tramaglini DM, et al: Clinical
management of partially lacerated digital flexor tendons: A
survey of hand surgeons, J Hand Surg (Am) 20:273–275,
1995.
2. Wray RC, Weeks PM: Experimental comparison of technics of
tendon repair, J Hand Surg (Am) 5:144–148, 1980.
3. Reynolds B, Wray RC Jr, Weeks PM: Should an incompletely
severed tendon be sutured? Plast Reconstr Surg 57:36–38,
1976.
4. Kleinert HE: Should an incompletely severed tendon be
sutured? commentary, Plast Reconstr Surg 57:236, 1976.
5. Janecki CJ Jr: Triggering of the finger caused by flexor-tendon
laceration: A report of two cases, J Bone Joint Surg (Am)
58:1174–1175, 1976.
6. Schlenker JD, Lister GD, Kleinert HE: Three complications of
untreated partial lacerations of flexor tendons—entrapment,
rupture, and triggering, J Hand Surg (Am) 6:392–398, 1981.
7. McCarthy DM, Tramaglini DM, Chan SS, et al: Effect of
partial laceration on the structural properties of the canine
FDP tendon: An in vitro study, J Hand Surg (Am) 20:795–800,
1995.
8. Mason ML, Allen HS: The rate of healing of tendons: An
experimental study of tensile strength, Ann Surg 113:424–
459, 1941.
9. Mason ML, Shearon CG: The process of tendon repair, Arch
Surg 25:615–692, 1932.
10. Noyes FR: Functional properties of knee ligaments and altera­
tions induced by immobilization, Clin Orthop Relat Res 123:
210–242, 1977.
11. Viidik A: Tensile strength properties of Achilles tendon
systems in trained and untrained rabbits, Acta Orthop Scand
40:261–272, 1969.
12. Woo SL, Gelberman RH, Cobb HG, et al: The importance
of controlled passive mobilization on flexor tendon healing.
A biomechanical study, Acta Orthop Scand 52:615–622,
1981.
13. Cooney WP, Weidman KA, Malo DS, et al: Partial flexor
tendon lacerations. In Huner JM, Schneider LH, Mackin E,
editors: Tendon Surgery in the Hand, St Louis, 1987, CV Mosby,
pp 148–155.
14. Amadio PC: Friction of the gliding surface: implications for
tendon surgery and rehabiliation, J Hand Ther 18:112–119,
2005.
15. Gelberman RH, Boyer MI, Brodt MD, et al: The effect of gap
formation at the repair site on the strength and excursion of
intrasynovial flexor tendons. An experimental study on the
early stages of tendon healing in dogs, J Bone Joint Surg (Am)
81:975–982, 1999.
16. Gelberman RH, Manske PR: Effects of early motion on the
tendon healing process: experimental studies. In Hunter JM,
Schneider LH, Mackin EJ, editors: Tendon Surgery in the Hand,
St Louis, 1987, CV Mosby, pp 170–177.
17. Gelberman H, Menon J, Gonsalves M, et al: The effects of
mobilization on the vascularization of healing flexor tendons
in dogs, Clin Orthop Relat Res 153:283–289, 1980.
18. Seyfer AE, Bolger WE: Effects of unrestricted motion on
healing: A study of posttraumatic adhesions in primate
tendons, Plast Recontr Surg 83:122–128, 1989.
19. Wray RC Jr, Weeks PM: Treatment of partial tendon lacera­
tions, Hand 12:163–166, 1980.
20. Dobyns RC, Cooney WP, Wood MB: Effect of partial lacera­
tions on canine flexor tendons, Minn Med 65:27–32, 1982.
21. Hariharan JS, Diao E, Soejima O, et al: Partial lacerations
of human digital flexor tendons: A biomechanial analysis,
J Hand Surg (Am) 22:1011–1015, 1997.

178
Section 2:  Primary Flexor Tendon Surgery
43. Zobitz ME, Zhao C, Erhard L, et al: Tensile properties of
sutures methods for repair of partially lacerated human flexor
tendons in vitro, J Hand Surg (Am) 26:821–827, 2001.
44. Zhao C, Amadio PC, Momose T, et al: The effect of suture
technique on adhesion formation after flexor tendon repair
for partial lacerations in a canine model, J Trauma 51:917–
921, 2001.
45. Tan J, Wang B, Xu Y, et al: Effects of direction of tendon lac­
erations on strength of tendon repairs, J Hand Surg (Am)
28:237–242, 2003.
46. Zhao C, Amadio PC, Zobitz ME, et al: Effect of synergistic
motion on flexor digitorum profundus tendon excursion,
Clin Orthop Relat Res 396:223–230, 2002.
47. 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) 84:78–84, 2002.
48. Zhao C, Amadio PC, Momose T, et al: Remodeling of the
gliding surface after flexor tendon repair in a canine model
in vivo, J Orthop Res 20:857–862, 2002.

179
CHAPTER
15 
FLEXOR TENDON INJURIES
IN CHILDREN
Shian Chao Tay, PhD, and Steven L. Moran, MD
identification of these injuries include difficulties or
inadequacy of physical examination. Many flexor
tendons injuries in children are due to lacerations with
broken glass, which may produce a small skin laceration
or puncture but is often associated with a high rate of
injury to underlying vital structure.2
Advances in flexor tendon repair techniques and the
introduction of early postoperative rehabilitation pro­
tocols have improved the results of flexor tendon surgery
in adults over the last two decades.3-7 Unfortunately,
these advances are not always applicable to children
with flexor tendon injuries due to the size of their
tendons and their compliance with therapy. The size of
the profundus tendon can be less than 4 mm in young
children, making the placement of multiple core sutures
difficult, particularly if larger suture material is used.
Finally, early motion protocols can result in flexor
tendon rupture if the child is noncompliant with splint­
ing or therapy. The purpose of this chapter is to review
the pertinent issues in the management of flexor tendon
injuries in children, including clinical pearls for identi­
fication, repair, and rehabilitation of these injuries in an
effort to improve outcomes of flexor tendon injuries in
this age group.
INCIDENCE
Finger flexor tendon injury in the pediatric population
is less frequent than in adults. In a report from Helsinki,
the annual incidence is 3.6 per 100,000 children (<16
years). Peak incidence is at 7 years of age. The ratio of
boys to girls was 3:1.8 Broken glass is the most common
mechanism of injury with some authors reporting
nearly 80% of such injuries being caused by it.9-12 As in
adults, zone 2 injuries are the most common injuries in
children with the little finger being the most commonly
injured.10,13,14 The rate of concurrent digital nerve injury
range from 36% to 58%.8,11,12,14-16
DIAGNOSIS
The assessment of flexor tendon injuries in the pediatric
population is difficult (Figure 15-1). Young children,
particularly those under 6 years of age, cannot be
expected to cooperate well with the examiner during the
OUTLINE
Finger flexor tendon injury in the pediatric population
is less frequent than in adults and is often caused by
a sharp cut, requiring primary repair. Advances in
tendon repair techniques and the postoperative reha­
bilitation protocols are not always applicable to chil­
dren. The size of the tendon can be less than 4 mm in
young children, making the placement of multiple
core sutures difficult. Depending on the size of the
tendon, core suture size of 4-0 can be used, and a four-
strand repair should be attempted. Epitenon sutures
should be one size smaller than core sutures, and sizes
5-0 through to 7-0 can be used. In older children,
four-strand or even six-strand repairs can be per­
formed. The principle is to have as strong a repair as
possible without making the repair site too bulky. In
zone 2 injuries, both the flexor digitorum profundus
(FDP) and flexor digitorum superficialis (FDS) tendons
should be repaired. Repair of FDP alone and discard­
ing the lacerated FDS is not recommended. We recom­
mend that all children under 7 years of age or who are
not cooperative be immobilized in an above elbow
cast for 4 weeks after surgery. Children who are older
than 7 years and who are judged to be compliant can
be started on early mobilization regimens. Unre­
stricted exercises are started after 4 weeks. In delayed
cases with extensive scarring, tendon grafting will be
necessary. If the pulley system is excessively damaged,
a two-stage tendon grafting procedure is indicated.
Pediatric flexor tendon injuries remain a complicated
problem for the hand surgeon. The factors that contrib­
ute to the difficulties in treating these injuries in chil­
dren include difficulties in diagnosis, the technical
demands of the repair, and the child’s compliance with
postoperative therapy. Five decades ago, more than 90%
of flexor tendon injuries in children were missed at the
time of the initial injury and thus were treated late with
flexor tendon grafting.1 Unfortunately, within the
present day, a significant percentage of flexor tendon
injuries in children still present late making primary
repair more difficult. Reasons for continued delays in

180
Section 2:  Primary Flexor Tendon Surgery
Figure 15-1  A 2-year-old child suffered a laceration to
the small finger while playing on a slide. The laceration was
initially closed by the local emergency room physician 2
weeks prior to presentation to the hand surgeon for noted
lack of motion of the small finger. A, Note the lack of normal
cascade in repose. B, The lack of flexion of the small finger
with wrist extension tenodesis. C, Additional maneuvers to
identify flexor tendon injury include forearm compression
which should result in some flexion of the fingers, but none
is seen within the small finger of this child. D, The finger
was explored through a Bruner incision where a complete
laceration of the FDP and FDS tendons was noted in zone 2.
E, Tendon repair was performed with a four-strand core
repair using a modified Kessler technique followed by a
locked running epitendinous suture.
A
B
C
D
E

181
Chapter 15:  Flexor Tendon Injuries in Children
Radiographs of the injured finger or hand may be
useful in excluding a retained foreign body or underly­
ing fracture. Ultrasound has limited diagnostic accuracy
in an uncooperative child. Magnetic resonance imaging
(MRI) is expensive and requires a general anesthetic for
young children.13 Both modalities can be considered if
the child is sedated.
Concurrent injury to neurovascular structures should
always be assessed. Vascularity of the finger is assessed
in the usual manner by observation of color and contour
of the finger pulp. Capillary refill can be checked by
indenting the finger pulp and observing for the speed
of refill. In dark-skinned individuals, finger pulp color
may be difficult to assess. In such cases, the nail bed
is blanched by distal pressure on the nail and the
speed of refill of the nail bed is observed when the
pressure is released. Hand-held Doppler exam may be
used to aid in identification of unilateral digital arterial
injuries.
Examining for digital nerve injury in the child can be
virtually impossible if they are not able to understand
or cooperate with the examination. As such, the usual
methods employed in the adults such as comparative
light touch and two-point discrimination tests may not
work. Thus, in children, nerve injury should always
be suspected based on the location of the surface
wound. The loss of resistance to glide using a plastic pen
can help in diagnosing the loss of sweating in a nerve
injury. Alternatively, the hand may be immersed in
water at room temperature until skin wrinkling occurs.
As only innervated skin wrinkles, an area of unwrinkled
skin indicates a likely nerve injury.13 In older injuries,
dryness or trophic changes are clues to sensory nerve
disturbances.1 Surgical exploration under anesthesia
may be the only way to verify the condition of the
nerves.9,13
hand exam.1 Observation and additional techniques
must be used to determine the status of the flexor
tendon in children. Previous reports by Bell and Mason
have also noted the importance of observation of the
child’s hand at rest and during play to look for signs of
active flexion or a loss of the normal cascade within the
hand.1,9 Children with older injuries will learn to adapt
to their injury and can gasp the injured finger with a
normal adjacent finger during flexion, thereby giving
the impression that active motion is being performed.1
This trick maneuver, called “trapping,” may initially
mislead both parents and examiners into thinking that
there is no disability.13
Missed diagnosis resulting in a delayed diagnosis
still accounts for 15% to 30% of pediatric patients
with flexor tendon injuries.8,11,14 The risk of delayed diag­
nosis and treatment is particularly high in children who
are less than 6 years of age8 and, paradoxically, in chil­
dren who sustain injuries with small surface wounds
that bleed little.1 Wounds that bleed profusely often
compel parents to seek medical attention and the likeli­
hood of a missed flexor tendon injury is thus lower.
Regardless, a high index of suspicion should always be
adopted when faced with a child with a hand or finger
laceration.
Examination begins with simple observation of the
resting posture of the hand. In the normal situation, the
fingers of the hand lie in a cascade with some flexion
being present in all the finger joints. A finger that is
observed to be out of alignment from the normal finger
cascade may have a flexor tendon injury. If the joints in
that finger are carefully observed, it may be possible to
even determine which flexor tendon is injured. In com­
plete lacerations of both the flexor digitorum profundus
(FDP) and flexor digitorum superficialis (FDS) tendons,
both the proximal interphalangeal (PIP) joint and distal
interphalangeal (DIP) joint will be extended. In isolated
lacerations of the FDP tendon, only the DIP joint will
be in extension with residual flexion present in the PIP
joint. In isolated FDS injury, the finger will lie out of the
normal finger cascade but there will be residual flexion
present in both the PIP and DIP joints. If the child
permits, the tenodesis effect can be elicited by perform­
ing passive wrist flexion and extension to check for
integrity of the flexor tendons (Figure 15-2). In a normal
situation, one should observe increasing flexion of all
finger and thumb joints with passive wrist extension.
Assuming that the child is cooperative and is able to
relax, flexor tendon injury should be suspected if there
is either absence of flexion or decreased amount of
flexion in a finger joint. Passive flexor tendon excursion
can also be performed by compression of the flexor
tendons in the distal forearm. Again, if there is absent
flexion or decreased amount of flexion, this would
suggest a complete flexor tendon laceration (see Figure
15-1C).
Figure 15-2  Loss of tenodesis of the long finger with wrist
extension.

182
Section 2:  Primary Flexor Tendon Surgery
tendon injury, up to 46% of flexor tendon injuries in
children are still treated with plaster immobilization.8,14
Most authors now agree that no benefit can be found in
instituting early mobilization protocols in children.11,14-16
The Mayo Clinic group found that children started on
early mobilization had a TAM of 78%, whereas those in
the immobilized group, 82%, using the Strickland and
Glogovac formula.14 Fitoussi et al reported a mean TAM
of 87% for the group with early mobilization versus 86%
in the immobilized group.16 Berndtsson and Ejeskär
reported 79% mean TAM in the group with Kleinert
protocol versus 74% in the immobilized group.11 None
of these differences were statistically significant.
It is well known that flexor tendon healing is more
rapid in children.9,10 In children, the distances between
the vinculi are shorter so that the relative blood supply
per square centimeter of tendon substance is consider­
ably better than in the adult. The better vascularity of
the flexor tendon encourages healing and resultant
adhesions are less extensive and more pliable.9,19 As
such, prolonged immobilization beyond 4 weeks is
unnecessary in children. It is also documented that
immobilization for longer than 4 weeks significantly
deteriorates functional outcome.15,16
Fitoussi et al recommend that an above elbow cast
be routinely used for postoperative immobilization to
avoid tendon ruptures.16 In their study, five ruptures
occurred in 58 fingers reviewed. Of these, two of the
ruptures were attributed to noncompliance to splint.
The remaining three ruptures occurred in the group of
patients immobilized in a below elbow plaster cast. No
ruptures were found in the group immobilized in an
above elbow cast. Incidentally, four of the five ruptures
above occurred in children who were 5 years or younger.
The benefit of above elbow casting for postoperative
immobilization is corroborated by Kato et al’s report.
They looked at the long-term outcome of 12 children
less than 6 years of age with zone 2 flexor tendon lac­
erations repaired with a modified Kessler technique
and immobilized with an above elbow cast for 4 weeks.
They reported a mean TAM of 89% evaluated with the
Strickland formula, which was a very favorable result.12
OUTCOMES
As stated earlier, there are many factors that can affect
the outcome of flexor tendon repairs in children includ­
ing type of repair and zone of injury. Other variables are
more controversial; the effect of age on functional
outcome continues to be an area of debate. The Mayo
Clinic group reported that age was not a determinant of
outcome.14 On the other hand, Berndtsson et al reported
that older children tended to have better results.11
Fitoussi et al compared the outcomes between those
aged 0 to 5 years, 5 to 10 years, and 10 to 15 years.16
Although they did find that the youngest age group was
associated with higher rupture rates, they did not find
TREATMENT
In 1963, Entin stated that flexor tendon surgery in
children is more difficult and demanding than in
adults and requires meticulous attention to handling of
tissues as the structures are so delicate.9 Many authors
have also advocated that only surgeons trained in hand
surgery should perform such operations.3,4,9,10 Absorb­
able sutures should always be used for wound closure
so as to avoid another anesthetic exposure for suture
removal.13
The options for tendon repair in children are as
varied as they are in the adult patient. The literature is
scant with regard to comparative studies, but most series
favor some type of four-strand repair with an epitendi­
nous suture. Navali and Rouhani reported one rupture
in a group of children less than 4 years of age with a
two-strand repair while there were no ruptures in the
group with four-strand repairs.17 Nietosvaara et al
reported no ruptures in their group of children less
than 16 years of age who had multistrand repairs, but
three postoperative ruptures occurred in their group
with two-strand repairs.8 In both reports, other than the
ruptures, no difference in functional outcomes was
found between the two-strand and multistrand repair
groups. It appears that in children, multistrand repair
may protect against postrepair rupture but not have any
significant effect on functional outcomes.
ADJUNCTS TO SURGERY
Tuzuner et al20 reported on the adjunctive use of botu­
linum toxin type A for children less than 6 years with
zone 2 flexor tendon repairs. The principle was to tem­
porarily paralyze the muscle unit of the repaired flexor
tendon to avoid ruptures due to poor compliance. All
the patients received intramuscular injections into the
specific muscles of the repaired tendons. Postopera­
tively, all patients were started on Duran’s passive
motion program for 4 weeks, after which active motion
was allowed. The average duration of the botulinum
effect was 6 weeks and all seven patients achieved satis­
factory results with an average total active motion (TAM)
of 84%. There were no tendon ruptures. However, a few
concerns were noted. Forearm atrophy was noted in
three of the seven patients. Rehabilitation was pro­
longed when otherwise it would have taken only 4
weeks. Other potential concerns of this experimental
adjunctive therapy include allergic reactions. At this
point in time, the use of botulinum toxin A is experi­
mental and should be performed within the confines of
a clinical trial.
REHABILITATION
Young children cannot always be expected to cooperate
with a rehabilitation program. In this day and age when
early mobilization is the norm in adults with flexor

183
Chapter 15:  Flexor Tendon Injuries in Children
required tenolysis. Both achieved satisfactory outcomes
after secondary repairs and tenolysis, respectively.14 Kato
et al reported that only one of their 12 patients required
a tenolysis, which achieved excellent outcome at final
follow-up.12 Fitoussi et al reported a 9% rupture rate.16
They report that risk factors for rupture were a noncom­
pliant patient, age group less than 5 years, and immo­
bilization in a below elbow cast compared to an above
elbow cast.16 Two of the ruptures were treated by direct
secondary repair after lengthening at the musculotendi­
nous junction. Both achieved satisfactory outcomes at
final follow-up. The other 3 underwent two-stage flexor
tendon grafting with one achieving good outcome and
the other 2 fair outcomes. Navali and Rouhani had one
rupture in one child with a two-strand repair. In this
case, no further treatment was given as the parents
declined.17
Digital growth disturbances may occur after flexor
tendon injury during childhood. Kato et al reported that
there is a small risk of the finger being shorter at the
middle and distal phalanx as a result of previous flexor
tendon injury. The range of shortening was 2 to 4 mm
for each digit.12 A mean shortening of 3% of digit length
was also reported by Tuzuner et al.20
FLEXOR TENDON GRAFTING IN CHILDREN
Primary tendon repair has been shown to be superior
to tendon grafting and staged reconstruction.21-24 As
such, flexor tendon grafting should be done only when
direct repairs are not possible. Courvoisier et al recom­
mend that one-stage flexor tendon grafting can be per­
formed in a digit with moderate scarring with intact A2
and A4 pulleys and full passive mobility of the PIP and
DIP joints.25 Two-stage flexor tendon grafting should be
reserved for cases where scarring within the flexor sheath
is associated with extensive pulley damage with PIP or
DIP joint contractures. In his series, Courvoisier et al
reported 38% (3 of 8 children) satisfactory outcomes
for one-stage flexor tendon grafting and 42% (5 of 12
children) satisfactory outcomes for two-stage flexor
tendon grafting.25 Darlis et al painted a better picture in
a group of children who underwent two-stage flexor
tendon reconstruction with the modified Paneva-
Holevich technique.26 Eight of their 9 children achieved
satisfactory outcomes with a mean TAM of 75%.
However, they had one child who had a complication
of deep infection that required removal of the silicone
rod. They were able to reinsert the rod 3 months later
when the infection cleared.
Valenti and Gilbert reported satisfactory outcomes in
73% of the 27 children they treated with two-stage
flexor tendon grafting.27 They noted that they achieved
better outcomes in older children. The mean TAM for
children aged 10 to 15 years was 81.5%, whereas chil­
dren aged 1 to 3 years had a mean TAM of 53%. They
had a complication rate of 27%: there were 4 distal
any significant difference in functional outcome among
the three age groups.
Concomitant injuries within the affected fingers may
also influence outcomes. The Mayo Clinic group
reported lower TAM in zone 2 for combined FDP/FDS
injuries compared with isolated FDP injuries. They also
reported that in patients with concomitant digital nerve
injury, the outcome was poorer.14 O’Connell et al also
reported poorer TAM in patients with concomitant
palmar plate or digital nerve injury.15 The one exception
to these findings was reported by Berndtsson and
Ejeskär, who evaluated their outcomes in zone 2 flexor
tendon repairs and found that concurrent superficialis
tendon injury did not appear to impact negatively on
functional outcome.11
Injuries within zone 2 are still presumed to produce
the poorest outcomes. Fitoussi et al reported that all
flexor tendon repairs in zones 1, 4, and 5 in children
achieved good or excellent outcomes.16 Outcomes in
zones 2 and 3 were good or excellent in 77% and 71%
of the digits in terms of TAM, respectively.16 The Mayo
Clinic group also reported better outcomes in zone 1
injuries compared to zone 2.14
Surprisingly, delayed repairs have not been shown to
consistently result in a poorer outcome in children. Ber­
ndtsson and Ejeskär found no significant difference in
outcome between children who had primary repair
versus those with delayed repairs (mean of 58 days after
injury, range 10 days to 1 year). In addition, none of
their patients with delayed treatment required tendon
grafting.11
Functional outcome in children following flexor
tendon repair appears to improve with time. This has
been attributed to better remodeling and continuing
digital growth, which contributes to the rupture of ten­
dinous adhesions.12,14,18 O’Connell et al reported con­
tinuing improvements in TAM beyond 18 months of
follow-up.15 In their report, a further 17% improvement
of TAM was noted in a subgroup of patients who were
available for extended reevaluation 18 to 120 months
later.
COMPLICATIONS
Grobbelaar and Hudson reported that complications
after flexor tendon repair in children were rare.10 In their
series of 38 patients, there were three cases of tendon
ruptures that they attributed to technical factors. All
underwent direct secondary repair and eventually
achieved excellent to good outcomes. None of their 38
cases required flexor tenolysis. However, they com­
mented that if tenolysis were required, they would prefer
to wait at least 18 months before contemplating it, as
adhesions in children are more pliable.10
In the Mayo Clinic group of 35 patients, complica­
tions were noted in only two patients.14 One had a
tendon rupture and the other had a stiff finger that

184
Section 2:  Primary Flexor Tendon Surgery
For zone 1 repairs without sufficient distal tendon
stump, the FDP tendon may be repaired directly to bone
using a nonabsorbable suture13 or a Bunnell pull-out
repair using 4-0 monofilament suture.14 Care should be
taken not to damage the epiphysis if drill holes are
made in the distal phalanx.
In zone 2 injuries, both the FDP and FDS tendons
should be repaired. Presently, repair of FDP alone and
discarding the lacerated FDS is not a recommended
option. The tendons are repaired with nonabsorbable
sutures. All skin wounds should be closed with absorb­
able sutures so as to avoid suture removal, which may
be distressing for the child.
Postoperatively, we recommend that all children
under 7 years of age or children who are not cooperative
be immobilized in an above elbow cast with the elbow
in flexion for 4 weeks. Children who are older than 7
years and who are judged to be compliant and coopera­
tive can be started on early mobilization regimens. In
both cases, unrestricted motion exercises are started
after 4 weeks.
In delayed flexor tendon injuries with significant
scarring making direct repair impossible, flexor tendon
ruptures occurring at the distal implant–tendon inter­
face, 2 silicone synovitis requiring synovectomy, and
one deep infection requiring removal of the silicone
rod. Four of these 7 cases with complications ended up
with unsatisfactory outcomes.27
CURRENT TREATMENT RECOMMENDATIONS
A high index of suspicion should always be adopted
when evaluating children for flexor tendon injury. If
there is any doubt, exploration under general anesthesia
should be performed to confirm the status of the flexor
tendon and digital neurovascular bundles.
Surgery should always be performed under general
anesthesia, in a bloodless field using pneumatic tourni­
quet control, fine instruments, and magnification. If the
injury is more than 1 week old, the parents and the
patient should be informed of the possibility for graft­
ing of the flexor tendon. Direct tendon repairs should
be attempted even in delayed cases. Flexor tendon graft­
ing procedures should be reserved for salvage situations
where there is extensive injury to the flexor tendon
sheath or irreparable damage to the pulleys. Tenodesis
and arthrodesis should be avoided in children as they
may further affect finger growth adversely.
The flexor tendons should be approached through
a Bruner extension of skin laceration (Figure 15-3).
Windows in the flexor tendon sheath can be created
using L-shaped incisions to expose the tendons. As far
as possible, A2 and A4 pulleys should be preserved. If
necessary not more than half the length of the A2 and
A4 pulleys can be released to aid in exposure of the
flexor tendon for repair.
Depending on the size of the tendon, core suture size
of 4-0 can be used, and a four-strand repair should be
attempted. Epitenon sutures should be one size smaller
than core sutures, and sizes 5-0 through to 7-0 can be
used. Cutting needles are preferred as they allow the
needle to pass through the tendon with less trauma than
with round needles.
Whether a two-strand or multistrand repair is per­
formed will depend on the size of the tendon. In very
small caliber tendons, a four-strand repair may not be
possible. In older children, four-strand or even six-
strand repairs can be performed. The principle here is
to have as strong a repair as possible without making
the repair site too bulky. Tendon repair techniques
reported in children include Tajima,17 modified Kessler,14
and Lim.8,28
Once the repairs are complete, the finger should be
passively flexed and extended to make sure that the
repair can glide smoothly through the pulleys and the
flexor sheath. Occasionally, partial release of the unyield­
ing ends of the A4 or the A2 pulleys29 may be needed to
prevent the repair site from snagging on the pulley edges.
We prefer leaving the L-shaped windows of the flexor
sheath unrepaired to avoid constriction to the tendon.
Figure 15-3  A, A Bruner incision is used to expose the
profundus and superfiicalis tendon injury of the middle
finger in a 5-year-old child. B, Flexion cascade is restored
following tendon repair.
A
B

185
Chapter 15:  Flexor Tendon Injuries in Children
will be necessary, which has been detailed by Valenti
and Gilbert.27 After stage 1 surgery, passive motion exer­
cises are started and continued for at least 3 months
before proceeding to stage 2 surgery. After stage 2, the
wrist and elbow are immobilized for 4 weeks before
beginning active motion exercises.27
grafting will be necessary. If the pulley system is intact
and there are no joint contractures, one-stage flexor
tendon graft using the palmaris longus or the plantaris
tendon should be performed with the proximal tendon
juncture sited in the palm.27 If the pulley system is exces­
sively damaged, a two-stage tendon grafting procedure
References
1. Bell JL, Mason ML, Koch SL, et al: Injuries to flexor tendons
of the hand in children, J Bone Joint Surg (Am) 40:1220–1230,
1958.
2. Provencher MT, Allen LR, Gladden MJ, et al: The underestima­
tion of a glass injury to the hand, Am J Orthop (Belle Mead
NJ) 35:91–94, 2006.
3. Kleinert HE, Kutz JE, Atasoy E, et al: Primary repair of flexor
tendons, Orthop Clin North Am 4:865–876, 1973.
4. Creekmore H, Bellinghausen H, Young VL, et al: Comparison
of early passive motion and immobilization after flexor
tendon repairs, Plast Reconstr Surg 75:75–79, 1985.
5. Savage R: In vitro studies of a new method of flexor tendon
repair, J Hand Surg (Br) 10:135–141, 1985.
6. Savage R, Risitano G: Flexor tendon repair using a “six strand”
method of repair and early active mobilisation, J Hand Surg
(Br) 14:396–399, 1989.
7. Small JO, Brennen MD, Colville J: Early active mobilisation
following flexor tendon repair in zone 2, J Hand Surg (Br)
14:383–391, 1989.
8. Nietosvaara Y, Lindfors NC, Palmu S, et al: Flexor tendon
injuries in pediatric patients, J Hand Surg (Am) 32:1549–
1557, 2007.
9. Entin MA: Flexor tendon repair and grafting in children, Am
J Surg 109:287–293, 1965.
10. Grobbelaar AO, Hudson DA: Flexor tendon injuries in chil­
dren, J Hand Surg (Br) 19:696–698, 1994.
11. Berndtsson L, Ejeskär A: Zone II flexor tendon repair in chil­
dren. A retrospective long term study, Scand J Plast Reconstr
Surg Hand Surg 29:59–64, 1995.
12. Kato H, Minami A, Suenaga N, et al: Long-term results after
primary repairs of zone 2 flexor tendon lacerations in chil­
dren younger than age 6 years, J Pediatr Orthop 22:732–735,
2002.
13. Havenhill TG, Birnie R: Pediatric flexor tendon injuries, Hand
Clin 21:253–256, 2005.
14. 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.
15. O’Connell SJ, Moore MM, Strickland JW, et al: Results of zone
I and zone II flexor tendon repairs in children, J Hand Surg
(Am) 19:48–52, 1994.
16. Fitoussi F, Lebellec Y, Frajman JM, et al: Flexor tendon injuries
in children: factors influencing prognosis, J Pediatr Orthop
19:818–821, 1999.
17. Navali AM, Rouhani A: Zone 2 flexor tendon repair in young
children: a comparative study of four-strand versus two-
strand repair, J Hand Surg (Eur) 33:424–429, 2008.
18. Friedrich H, Baumel D: The treatment of flexor tendon injuries
in children, Handchir Mikrochir Plast Chir 35:347–352, 2003.
19. Arons MS: Purposeful delay of the primary repair of cut flexor
tendons in “some-man’s-land” in children, Plast Reconstr Surg
53:638–642, 1974.
20. Tüzüner S, Balci N, Ozkaynak S: Results of zone II flexor
tendon repair in children younger than age 6 years: botuli­
num toxin type A administration eased cooperation during
the rehabilitation and improved outcome, J Pediatr Orthop
24:629–633, 2004.
21. Boyes JH, Stark HH: Flexor-tendon grafts in the fingers and
thumb. A study of factors influencing results in 1000 cases,
J Bone Joint Surg (Am) 53:1332–1342, 1971.
22. Ejeskar A: Flexor tendon repair in no man’s land. II: Early
versus late secondary tendon repair ad modum Kleinert,
Scand J Plast Reconstr Surg 14:279–283, 1980.
23. Vahvanen V, Gripenberg L, Nuutinen P: Flexor tendon injury
of the hand in children. A long-term follow-up study of 84
patients, Scand J Plast Reconstr Surg 15:43–48, 1981.
24. Amadio PC, Wood MB, Cooney WP 3rd, et al: Staged flexor
tendon reconstruction in the fingers and hand, J Hand Surg
(Am) 13:559–562, 1988.
25. Courvoisier A, Pradel P, Dautel G: Surgical outcome of
one-stage and two-stage flexor tendon grafting in children,
J Pediatr Orthop 29:792–796, 2009.
26. Darlis NA, Beris AE, Korompilias AV, et al: Two-stage flexor
tendon reconstruction in zone 2 of the hand in children,
J Pediatr Orthop 25:382–386, 2005.
27. Valenti P, Gilbert A: Two-stage flexor tendon grafting in chil­
dren, Hand Clin 16:573–578, 2000.
28. Lim BH, Tsai TM: The six-strand technique for flexor tendon
repair, Atlas Hand Clin 1:65–76, 1996.
29. Amis AA, Jones MM: The interior of the flexor tendon sheath
of the finger. The functional significance of its structure, J Bone
Joint Surg (Br) 70:583–587, 1988.

186
CHAPTER
16 
PRIMARY REPAIR OF
THE FLEXOR POLLICIS
LONGUS TENDON
David Elliot, MA, FRCS, BM, BCh
tension as a result of the greater retraction of the FPL
muscle remains unknown. Murphy believed this
problem to be due to the fact that the FPL tendon is
completely separated from the other flexor tendons,
making retraction easier than where a tendon is
restrained by others. An alternative explanation may lie
in the difference of muscle configuration between the
FPL muscle and those of the finger flexors.
THE AVASCULAR ZONE 2 SITE OF THE FPL
Recently, we re-examined data from our earlier studies
and found that the rate of rupture after repair in zone
2 is twice as common as after zone 1 FPL repairs.3
Zone 1 FPL repairs had a rate of rupture similar to that
of primary finger flexor repairs.4 The increased rate of
rupture of zone 2 repairs largely explains the higher
rupture rate of the FPL tendon when compared with
finger flexor repairs in zones 1 and 2. A lack of extrin­
sic vascular supply of the FPL tendon immediately
palmar to the metacarpophalangeal joint was identi­
fied by Lundborg.5 Hergenroeder et al6 also found that
there is a zone of relative avascularity of the tendon at
the same point, and this was illustrated by Tubiana
et al.7 This avascular point coincides with the zone
2 site of FPL repair and may explain the particular
susceptibility of zone 2 repairs to rupture, contributing
to the higher rupture rate of FPL tendon primary
repairs.
TREATMENT OF FPL DIVISION BEFORE 1989
Reports published in the 20 years after Murphy’s report,
when many repairs would now be classified as delayed
primary, or even secondary surgery, repeatedly high­
lighted the difficulty of direct repair of the FPL after any
delay and many recommended that primary repair be
avoided in most instances.8 To overcome this problem,
most authors favored tendon reconnection by interposi­
tion grafting,1,9-19 although some reported direct repairs
of the tendon, usually in cases seen within a few
hours.1,11,14-18,20-29 Tendon lengthening, either in the
muscle or in the tendon at the wrist, was also used
by several authors as an alternative to interposition
OUTLINE
A series of changes of primary treatment of divisions
of the flexor pollicis longus tendon were carried out
in the St. Andrew’s unit in a total of 216 thumbs over
18 years in an attempt to reduce the rupture rate
from the initial high level of 17%. It was possible,
eventually, by strengthening the sutures and including
the fingers in the splint, to reduce the rupture rate to
zero. A simplification was then made and others dis­
cussed, whereby elimination of mechanical ruptures
of the repairs could be achieved using simpler suture
methods that are more appropriate for use by the
training surgeons who normally perform these repairs
worldwide.
More difficulty is encountered in approximation of the
severed ends than in any other tendon of the hand.
Murphy, 1937
THE PROBLEM OF FLEXOR POLLICIS
LONGUS RETRACTION
It has been known that the flexor pollicis longus (FPL)
tendon is more difficult to repair than the finger flexors
for over 60 years. In 1937, Murphy identified this
problem and highlighted the fact that retraction of the
proximal end of the cut FPL tendon was almost always
greater than that of the proximal ends of cut finger flexor
tendons.1 Our clinical experience, and that of others
more recently, would agree that the greater retraction of
the cut FPL tendon is due to shortening of the FPL
muscle as there is often considerable difficulty pulling
the proximal tendon end out to length if FPL repair is
attempted after even a short delay beyond 48 hours. It
would seem likely that this increased tension is also the
cause of the higher rupture rate in the repaired FPL
tendon compared to that of finger flexor tendon repairs.
More recently, the presence of a retracted proximal end
of the FPL tendon at surgery has been shown to be det­
rimental to the outcome of the tendon repair.2 Whether
all FPL repairs, or only those with visible retraction of
the proximal tendon at surgery, are exposed to increased