Hand Surgery
1st Edition

37
Early Repairs of Flexor Tendon Injuries
Daniel P. Mass
One of the many unique things that allows the human hand to be a finely tuned organ is the ability to independently move the fingers at each of the 14 joints of the hand for precision motion, yet also be able to move them simultaneously as a function of power. Humans use the muscles in their forearm to motor long tendons through a complex pulley system, allowing for efficient conversion of motor power into grasping motions and forces. When the fingers and wrist are held in a straight or neutral position, one can feel tension on both the flexor and extensor muscles in the proximal forearm. A passive collapse of the wrist into extension allows the passive tenodesis of the finger flexors to bend the fingers partially. Muscle contraction then provides the controlled force for both further motion and power for pinch and grip.
The loss of the ability to actively bend one or all fingers can occur from injuries to nerves proximal to the elbow or injuries to the motor or tendon units themselves. This chapter discusses injuries to the muscular-tendinous units. For the hand to function properly, tendons need to be strong and also have a long gliding excursion through the wrist and fingers. Therefore, after an injury and subsequent repair of a tendon, the return of function depends on the healing of the tendon in a manner that allows for both strength and gliding. Traditionally, this outcome has been a problem, particularly in the fingers. Because of the poor vascularity to the tendons, it was thought that the tendons were inert and could not heal without extrinsic adhesions. These adhesions led to a stronger repair without the ability to glide in many instances. Bunnell in 1918 admonished surgeons that “it is better to remove the tendons entirely from the finger and graft in new tendons throughout its length” (1). In the 1960s, surgeons such as Verdan (2,3), Kleinert et al. (4), and Kessler and Nissim (5) challenged the concept of grafting by performing primary tendon repairs. Because there were perceived differences in tendon healing in different parts of the hand with results of primary tendon repairs being compared in the literature, Verdan (2,3) proposed dividing the forearm and hand into five anatomic zones. This chapter uses these zones as a basis for describing the anatomy, repair, rehabilitation, and results of flexor tendon injuries.
ANATOMY
The five zones of the forearm and hand as defined by Verdan (2,3) are shown in Figure 1. Zone V is made up of the muscles of the forearm. This includes the median nerve innervating the flexor carpi radialis (FCR), the palmaris longus (PL), the flexor digitorum superficialis (FDS) as a superficial layer (Fig. 2A), a deep layer made up of the radial half of the flexor digitorum profundus (FDP), and the flexor pollicis longus (Fig. 2B). The ulnar innervated muscles include the flexor carpi ulnaris (FCU) in the superficial layer and the ulnar half of the FDP deep. The median and ulnar nerves and the radial and ulnar arteries run the length of the forearm under the surface of the superficial layer.
There are two potential poor prognosis problems associated with injury to the muscles: first, the recognition and treatment of injuries to the nerves and arteries, and, second, partial denervation of the muscle, leaving a small active proximal segment of muscle to motor the distal scar segment plus the tendon to obtain finger flexion.
Tendon anatomy, which functions like a piece of rope, is complex. The tendon itself is 70% longitudinal type I collagen fibers, with very few cells (the tenocytes) (6). The central core of the tendon is surrounded by a thin, cross-hatched layer of type I collagen, with many more fibroblasts (the epitenon). The tendon is also split centrally near the middle with an internal epitenon that breaks the tendon into two fascicles. If at all possible, it is this internal epitenon that should be grabbed by instrument when performing surgery. This layer provides a smooth, almost friction-free gliding surface between the tendon and its surrounding environment, the pulley system, allowing for the inflow and egress of nutrients from the surrounding
P.680

synovial fluid. The layers of the cross-hatched design allow it to act as a Chinese finger trap, tightening to longitudinal pull and strengthening the central core the way a waist belt aids a weight lifter (Fig. 3).
FIGURE 1. Flexor zones of the hand and forearm as defined by Verdan (2,3). These zones are used for describing the location of the injury and the prognosis for repair.
The third major structure in the tendon is the vascular, or vincular, supply (Fig. 4) that enters the tendon through the dorsal epitenon. The vascular supply only supplies the dorsal two-thirds of the tendon with blood flow; there are known watershed areas in the tendon between the vinculum (7,8,9 and 10).
The tendons themselves originate within the muscles of the forearm. They start as a thin structure in the middle of the muscle and enlarge until full width at the distal end of the muscle (11). The muscles end an average of 4 cm proximal to the distal wrist crease. The tendons of the wrist flexors cross the wrist and insert onto the proximal second (FCR) or fifth (FCU) metacarpal. They are responsible for power wrist flexion in either radial or ulnar deviation. The PL, a weak central wrist flexor, is present in only 80% of humans and inserts onto the palmar fascia.
The other nine tendons enter the hand by traveling with the median nerve under the transverse carpal ligament through the carpal tunnel. Zone IV is the area under the transverse carpal ligament. This thick ligament at the base of the palm protects the tendons from injury. Because of the frictional forces within the carpal tunnel, the tendons are surrounded by synovial sheaths, which provide not only
P.681

lubrication but also nutrition, as the tendons have little intrinsic blood supply. The transverse carpal ligament also acts as a pulley to maintain the finger flexors close to the carpal bones. This pulley prevents bowstringing at the wrist and improves the mechanical function of the muscles in flexing the fingers (12). This pulley function has been named the A-O pulley (13). Because of the exposed location, injuries to the tendons just proximal to the carpal canal are common. Any injury to a tendon at this level must make one think of a potential injury to the median nerve, as it is more superficial than the nine tendons. Because of the synovial coverage of the tendon up to 4 cm proximal to the carpal tunnel, the author believes that the area of zone IV should be expanded to include the entire synovial area (Figs. 5 and 6).
FIGURE 2. A: Volar forearm muscle (superficial). From radial to ulnar: flexor carpi radialis, palmaris longus, superficialis III and IV, and flexor carpi ulnaris. B: Volar forearm muscle (deep). Superficialis muscle in fingers, median nerve in pickups, brachial artery splitting to radial and ulnar artery at scissors tip. Flexor digitorum profundus tendon deep.
FIGURE 3. A: Histology. Scanning electron micrograph of a human flexor tendon. The center has longitudinal collagen fibers. At the edge is the epitenon with cross-hatched collagen fibers. B: Chinese finger trap, demonstrating the cross-hatched design of the epitenon.
FIGURE 4. Vincula. Vascular supply to the dorsum of the tendons. The vessels are long for tendon motion, and there are no vessels volarly because of the tendon compression against the pulleys. FDP, flexor digitorum profundus; FDS, flexor digitorum superficialis; VBP, vinculum brevis profundus; VBS, vinculum brevis superficialis; VLP, vinculum longus profundus; VLS, vinculum longus superficialis.
As the tendons travel from under the protection of the transverse carpal ligament, they enter zone III, which extends from the distal end of the transverse carpal ligament to the A1 pulley at the level of the distal palmar crease. Because this area is between pulleys, the tendons have a better blood supply through the mesotenon and, therefore, should have better healing potential. The main problem with injuries in zone III is the collateral damage to the nerves and superficial arch, which are superficial to the tendons and tightly compacted in the center of the palm, fanning out to the fingers.
From the distal palmar crease to the distal insertion of the FDP tendon on the distal phalanx, the tendons run through a fibro-osseous pulley system (Fig. 7). This pulley system prevents bowstringing at the finger joints and improves the mechanical function of the muscles in flexing the fingers (Fig. 8). The pulley system also creates frictional and compressive forces between the tendons and the pulley. This in turn requires the tendon to be avascular in its volar one-third area and to have synovial fluid secreted by the flexor sheath lining the pulley system. The synovial fluid
P.682

not only provides for frictional lubrication, but also provides the main nutritional flow for the tendons through diffusion (14,15,16 and 17). This fibro-osseous canal is divided into two zones. Zone II is from the A1 pulley to the insertion of the superficialis tendon on the middle phalanx. Zone I is the continuation of the canal to the insertion of the profundus tendon on the distal phalanx. The reason for the two different zones is the implication of healing of injuries to one tendon alone in zone I versus the likelihood of injuries to two tendons in zone II, which historically have the worst results due to increased chances for adhesions between the tendons and sheath or bone or between each other. All of this creates increased risk to the final function.
FIGURE 5. Carpal tunnel (A-O pulley) and its synovial lining, seen filled with methylene blue.
FIGURE 6. The author’s proposed new zones because of the expanded synovial lining seen outside the carpal canal and A1 pulley (dotted line). Old zones are drawn on the left hand. New, proposed zones are drawn on the right hand.
BIOMECHANICS
Without the flexor tendon pulley system made up of the carpal tunnel and the fibro-osseous canal of the finger, the tendons would fall away from the joints during flexion, creating a greater distance of the tendon from the axis of joint rotation. This in turn creates a greater moment arm for generating flexion but a smaller total rotation for a given muscle contraction. With the pulleys intact, the tendons are held relatively close to the joints; therefore, there is a more efficient generation of joint rotation for the same given muscle contraction (18) (Fig. 8).
The studies of Urbaniak et al. (19) and Schuind et al. (20), using force transducers on flexor tendons during carpal tunnel releases, found unrestricted passive motion of the flexor tendons generated 2 to 4 N of force through the tendon. Active flexion with mild resistance, which is undefined, generated 9 N or 19 N of force.
TENDON HEALING
In the 1950s and 1960s, both Potenza (21,22 and 23) and Peacock (24,25) performed research on the healing of flexor tendons. Peacock established the three phases of the repair process. This process was similar to any healing organ. First, it required the migration and invasion of peripheral cells and blood vessels, as the tendon cells themselves were not capable of collagen production. In the second phase, the tendon ends unite through the surrounding tissues. Finally, tendon remodeling takes place secondary to attempted tendon function. Potenza concluded that it was the proliferation and migration of macrophages and fibroblasts from the synovial sheath that repaired the tendon. Peacock and Potenza found that tendons themselves were inert and required extrinsic adhesions to heal.
In the late 1970s, Lundborg and Rank (26) floated a tendon segment in the knee of a rabbit and demonstrated end encapsulation. When some criticized the finding, saying that the cells could have come from the synovium and floated to the tendon, Lundborg and Rank (27) and Lundborg et al. (28) first placed the tendon in a dialysis bag that blocked the cells. The tendon segment still end encapsulated. In 1985, Lundborg et al. (29) further placed the tendon in a dialysis bag and placed the bag in a subcutaneous pouch, and the tendon still end encapsulated. Becker et al. (30) was the first to culture tendon segment. After placing a
P.683

circular hole in a chicken tendon, they demonstrated that placing a plasma clot in the hole accelerated tendon cell migration and proliferation. They were also able to confirm the synthesis of collagen. This was the first work that demonstrated that tendon cells participate in the healing process. Manske et al. (31) demonstrated the intrinsic healing capability of monkey, chicken, dog, and rabbit tendon segments in culture. Mass and Tuel (32,33) provided similar evidence for human tendon segments in culture. Manske et al. (14,15,16 and 17,34,35 and 36) established, using radioactive tracers, that both perfusion and diffusion were used by the tendon for nutritional exchange. They concluded that diffusion was not only more effective than perfusion, but could provide all the nutrition to the tendon when there was no blood vessel. This information allows consideration of the possibility of intrinsic healing through the aid of perfusion. Almost all hand surgeons now believe that the true healing of tendons clinically is a combination of both intrinsic and extrinsic healing. The key to a successful repair is controlling the needed extrinsic healing to prevent rigid adhesions and finding a repair technique that is strong enough to start early motion to provide for diffusion.
FIGURE 7. A: Pulley system of the fingers palmar aponeurosis (PA) to A5 outlined with methylene blue. B: Sketch of the pulley system: The annular pulleys attach to the volar plates at the joints and to the bone over the phalanges. The pulleys keep the tendons close to the bone to prevent bowstringing. C: Pulley in flexion: The cruciates become horizontal as they collapse with flexion to complete a single pulley in flexion. MA, flexion movement arm.
FIGURE 8. Effect of loss of a pulley, demonstrating bowstringing.
Gelberman et al. (37,38,39 and 40), in elegant studies starting in 1983 using a dog model, demonstrated that primary repairs of flexor tendons treated with a passive-motion postoperative protocol healed with minimal to no adhesions (Fig. 9). Gelberman et al. (41) was able to demonstrate intrinsic tendon
P.684

healing without adhesions through a tendon callus (Fig. 10).
FIGURE 9. Tendon healing without adhesions.
Since 1941, when Mason and Allen (42) demonstrated that healing tendons first became weaker at day 5 and then regained their day 1 suture strength by day 19, there was a tradition of immobilizing flexor tendon repairs for 3 weeks before starting a therapy program. However, this conclusion was based on research using the FCU and flexor carpi radialis, nonsynovial tendons of dogs, and not digital flexor tendons. Also, Mason and Allen never stated the percentage drop in strength that occurred, although they did present the raw data. The work, previously mentioned by Peacock (24,25) and Potenza (21,22 and 23), only confirmed this clinical approach to postoperative tendon repair programs. Despite this research and before the newer work, Kleinert et al. (43,44 and 45) and Duran and Houser (46) in the 1960s had started to move the tendon repairs early. Duran and Houser used pure passive motion, isolating the proximal and distal interphalangeal joints. Kleinert et al. used rubber bands to flex the fingers and the intrinsic muscles to extend the fingers to give motion “without tension.” The clinical results presented by these authors and others who copied their protocols were much better than the results with immobilization (47).
Urbaniak et al. (19) in 1975 studied the healing strength of the digital flexor tendons of dogs over time. They also demonstrated a decrease in the initial repair strength at 5 days when using the Kessler suture repair technique and a return to the strength of the suture by 3 weeks. This paper did not indicate the percentage of strength lost nor the number of tendons studied. Looking at the raw data, there is less than a 50% loss in strength. Strickland (48) also performed a study that demonstrated only approximately a one-third drop in tendon strength at 1 week postoperatively.
In 1985, Savage (49) published a paper in which he presented the rationale for a new six-strand suture technique. By having a stronger suture repair technique, surgeons could allow immediate early active motion of the fingers after surgery. Savage used Urbaniak et al.’s 1975 work (19). He stated that finger flexion “against moderate resistance is 1.5 kilograms (14.7 N).” He further stated, “This strength in a repair would therefore be adequate to permit early active movement, but tendon softens after injury to about one-fifth
P.685

of its initial strength at about one week from injury. Therefore,… the initial repair strength should be five times 1.5 kilograms (14.7 N), that is 7.5 kilograms (74 N).” His new repair technique failed at 6.85 kg (67 N), not quite enough by his calculations. He then applied this technique to a series using active motion postoperatively (50). The problem with this paper is that every subsequent author has used the one-fifth loss of strength and, therefore, stated the need for a repair greater than 74 N. In the author’s review of Urbaniak et al.’s 1975 paper (19), there is only one graft from which to extrapolate a 46% drop in the initial repair strength at 5 days. Even if we go back to Mason and Allen’s 1941 paper (42), the drop in strength at 5 days was only approximately two-thirds. Using a worst case scenario and this new interpretation of Urbaniak et al.’s or Mason and Allen’s work, surgeons only need to look for repair techniques that provide greater than two to three times 14.7 N or 19 N for the increase in strength. This comes to 30 to 45 N by the measurements of Urbaniak et al. (19) for active flexion with mild resistance or 45 to 60 N using Schuind et al.’s data (20).
FIGURE 10. A: Healing of a human profundus tendon in culture. B: Sketch of tendon callus showing primary healing of the tendon.
REPAIR TECHNIQUES
Since the 1990s, there has been a proliferation of articles presenting new stronger repair techniques to meet the criteria of Savage (49) at 74 N, but few provide this strength. Before this, there were three main repair techniques: Bunnell’s criss-cross technique (51), which is thought to strangulate the blood supply of the tendon and therefore fell out of favor; the Mason and Allen (42) repair, which has much of the suture repair exposed outside of the tendon and is thought to cause too much scarring; and the most popular technique, the modified Kessler technique (5), a grasping suture modified to bury the knot between the two ends of the tendon repair (Fig. 11A). This last core technique with an epitendinous repair became the standard for most surgeons who followed the work of Kleinert et al. (43,44).
Research has demonstrated, and most surgeons agree on, two factors in tendon repair. One, there needs to be both a core suture and an epitendinous repair. Two, the more strands of the core suture that cross the repair site, the stronger the repair. More recently, a locking core suture technique has become popular (52). To fit into these factors, what should be looked for in the suture material? The suture should have high tensile strength; have easy knotability, with minimal loss of strength; not be extensible, for preventive gapping; have minimal tissue response; be absorbable late after healing of the tendon; and have ease of use. A review of the literature agrees with Trail et al. (53), who recommend 3-0 or 4-0 braided sutures because of their ease of placement, adequate strength, and minimum stretching. In the author’s hand surgery, 4-0 Tevdek (D&G) or 4-0 Ethibond (Ethicon) is used, as 3-0 is too bulky, and enough strength is provided with 4-0.
Reviewing the literature, there are at least six 4-strand (54,55,56,57,58 and 59), three 6-strand (49,50,60,61), and one 8-strand (62) repair techniques that have been presented and evaluated for primary rupture strength (Fig. 11B–D). All of these repairs were tested in a linear model that does not take into account the friction and compressive forces of the pulley system. Many different types of tendons (pig, chicken, and human) were used for testing with many different types of suture material and suture sizes. This makes comparison of techniques difficult. Using Urbaniak et al.’s more conservative numbers, most of the four-strand repairs reach or surpass the 45 N needed for early active motion. In the literature, of the four-strand repairs, only the Becker (54) and modified Becker (55) are strong enough for the 60-N repair strength required by Schuind et al. (20). In the latest work from the author’s laboratory, the locked cruciate (57) repair also has a strength of greater than 60 N (82). The problem with the original Becker technique is that, to create the bevel for the repair, the tendon has to be shortened by more than 1 cm. The modified Becker technique has been studied in the anatomic model for strength (63). The advantage of this model is that the tendon is exposed to the same forces that the tendon sees in vivo. The major problem with the modified Becker technique is that much of the suture and the knots are on the outside of the tendon, and this could cause increased friction and work of flexion. If the repair is placed correctly on the side of the tendon, there is little compressive force and, therefore, little friction on the knots. Previous studies in the author’s research laboratory have not demonstrated increased work while using this technique (63). Aoki et al. (64) demonstrated that knots away from the tendon repair were stronger in two-, four-, and six-strand sutures. Pruitt et al. (65) showed that this effect lasted for 6 weeks before becoming equalized. The locked cruciate is as strong as the modified Becker, has a lower work of flexion, and is easier to perform. The locked cruciate is the author’s repair of choice.
Besides the core suture, there have been two studies looking at the epitenon repair and its strength without a core suture. The epitenon repair was originally added to the repair to “tidy up” (66) the tendon repair (Fig. 12). Lister et al. (45) also demonstrated that a simple epitendinous repair strengthened the Kessler core. Kubota et al. (67) studied six different epitendinous or circumferential suture techniques using 6-0 Prolene with no core repair. The Lin-locking technique was significantly more resistant to rupture than the others, but none was better for gap strength. Although the Lin-lock was stronger, it was the only one that significantly increased the resistance to gliding. Diao et al. (68) demonstrated that a deep, grab, locking, epitendinous suture was the strongest. The author currently uses the simple locking epitendinous repair with 6-0 nylon (66) (Fig. 12), because it is simple and decreases the risk of cutting the core suture with a deep bite or a cross stitch. The combination of the locked
P.686

P.687

cruciate and the simple locking epitendinous repair gives a strength of 64.1 N (82) in the anatomic testing. Currently, the author uses a strong four-strand repair technique, either the modified Becker or locked cruciate (4-0 Ethibond) with the simple locking epitendinous suture (6-0 nylon) for the profundus tendon and one slip of the superficialis, based on the latest work by Paillard et al. (83). If the superficialis has divided proximal to Camper’s chiasm, the author uses a single core suture and epitenon repair like the profundus. Early, protected, active range of motion exercises are then started. Good to excellent results have been accomplished, with no ruptures in all the patients who cooperated with the exercise program, including two boys 10 and 12 years old (Fig. 13). Other four-strand techniques can be used, but they have not been tested in an anatomic model to see if they provide 60 N of strength to failure.
FIGURE 11. Sketches of core repair techniques. A: Two-strand techniques: Bunnell, Mason-Allen, Kessler, and modified Kessler. B: Four-strand techniques: Strickland, Lee, Robertson, Becker, modified Becker, and locked cruciate (McLarney). C: Six-strand techniques: Savage, Sandow, and Lim. D: Eight-strand technique: Silva.
FIGURE 12. Sketches of the running and running lock epitenon repair techniques.
CLINICAL AND HISTORY
Except for ruptures of the FDP from the distal phalanx, almost all tendon injuries, whether partial or complete, are caused by a laceration from a sharp object, such as a knife or broken glass. Typical histories include defending oneself against a knife attack, reaching into a sink full of soapy water and cutting oneself on a knife or a piece of broken glass, and falling onto glass or through a glass window or
P.688

door. For presurgical planning, it is helpful to know if the fingers were in extension rather than flexion when the injury took place, as both tendons will be cut at the same level. This usually means that it is easier to find the tendon ends through a smaller incision, but both repairs will be at the same level, increasing the risk of the healing process and causing the tendons to become one scar unit.
FIGURE 13. A 12-year-old boy sharply cut both tendons in the long, ring, and little fingers. A: Preoperative loss of arcade. B: Postoperative flexion. C: Postoperative extension.
PHYSICAL EXAMINATION
Any complete injury to the flexor mechanism is accompanied by loss of active flexion. Because of their close proximity, any laceration of a tendon can cause a laceration to a nerve; therefore, evaluation of the nerve function should take place before any lidocaine (Xylocaine) is used to anesthetize the wound. This is best performed by feeling for loss of sweat and measuring two-point discrimination of the digits involved, as sticking a patient with a pin can cause fear of potential pain (Fig. 14A). Then, examine for the tendon injury. Asking a patient to actively flex his or her finger through an open wound is painful. To best examine someone without pain, use the tenodesis effect mentioned in the first paragraph of this chapter (Fig. 14B). Allow the wrist to fall into full extension and observe the position of the fingers. They should collapse into a flexion arcade, with the little finger more flexed than the ring and the ring more flexed than the long, which is more flexed than the index. Even the thumb is flexed at the interphalangeal (IP) joint. All the joints of all the fingers should be flexed to form this arcade. One should be able to imagine holding an egg in one’s hand (Fig. 14C). Any break in the arcade is a cause for worry in the light of an injury. If the ring finger is straighter than the rest of the fingers but there is proximal IP joint (PIP) flexion,
P.689

then one can confirm the suspicion of a jersey finger (Fig. 14D). If this correlates with the history of a gripping injury while playing football or there is a laceration over the distal half of the proximal phalanx or PIP joint, then a laceration of the FDP has occurred without a complete injury to the FDS. A finger held with slight flexion of both IP joints but with a break in the cascade demonstrates a laceration of the FDS without a complete FDP injury, as the FDP tension maintains some flexion at both joints. Obviously, a completely straight finger means a complete laceration to both the FDS and FDP. This examination is good for zone I to III injuries and some zone IV injuries. The synovial interconnection in the carpal tunnel may mask this type of examination. In the proximal aspect of zone IV, the most common place for injury in suicide attempts and falling through a glass door or window, any injury to a tendon other than the PL should warrant suspicions of an injury to the median nerve. If the FCU is involved, then a close evaluation for injury to the ulnar artery and nerve should be performed. Injury to the FCR requires evaluation of the radial artery.
FIGURE 14. Physical examination. A: Loss of sweat tested by loss of friction. B: Two-point discrimination examination. C: Normal tenodesis. D: Jersey finger. Note partial break in arcade in the long finger.
After this type of physical examination, the wound edges should be injected with local Xylocaine or a digital or wrist block performed for a confirmatory evaluation of the wound and to irrigate the wound before a loose, temporary closure. All wounds should be explored to look for partial tendon and nerve injuries that would otherwise be missed in the physical examination.
IMAGING STUDIES
If the injury was caused by glass, then an x-ray is appropriate to see if there is any remaining glass in the wound. If the injury is a jersey finger, then an x-ray of the finger is needed to see if a fragment of bone was avulsed with the tendon and to determine where the fragment has held up the proximal migration of the tendon.
NONOPERATIVE TREATMENTS
If function is to be returned, there are no nonoperative treatments for lacerations of flexor tendons. However, frequently a jersey finger injury is ignored by the young football player and not brought to the attention of his parents until after the season. If the hand is functional and there is no blockage to PIP flexion or no painful nodule in the palm from retraction of the profundus tendon, then no surgery is necessary.
SURGICAL INDICATION
Zone I and II injuries can be repaired electively within 2 to 3 days as long as one neurovascular bundle is intact. Because of the likelihood of multiple tendon and nerve injuries in zone III and IV, surgical exploration and repair of these injuries should take place as soon as possible. If both of the digital arteries or the radial and ulnar artery are lacerated, then, obviously, this is an emergency. Because injuries in zone V can be superficial or deep into the muscle bellies, care must be taken. If the surgeon is confident the injury is superficial in the muscles only, then an emergency room closure of the fascia with a dissolving 4-0 suture and a loose skin closure are appropriate. If the wound is deep, an acute formal surgical exploration is indicated.
SURGICAL MANAGEMENT
All flexor tendon injuries should be operated on in a formal operating room with at least an axillary block. It is advisable to use loupe magnification and a tourniquet when performing this type of surgery. After the appropriate anesthesia and preparation, the injured hand is laid out on a hand table. The incisions are planned based on the original laceration and the extent of dissection anticipated to find the tendon ends. For zone I and II, if one anticipates finding the proximal ends of the tendon in the palm, digital and palmar incisions do not have to be contiguous, but proper planning should allow them to connect, if it becomes necessary. Depending on the location and angle of the original laceration and the anticipated collateral injury to one or both digital neurovascular bundles in a digit, one draws either Bruner (70) zigzag incisions or mid-lateral incisions, or both (Fig. 15A). For a transverse laceration with tendons and one nerve injury, better exposure for both the tendons and the one nerve is opening the transverse laceration and extending with a mid-lateral incision. The author prefers Bruner incisions for pure tendon injuries or wide exposure for multiple structure injuries on both sides of the digit. After sheath exposure, the location of the laceration is identified. If the injury is in the middle of the A2 or A4 pulley, and the pulleys are otherwise intact, then opening the pulley system proximal or distal to these pulleys to form a cone (71) allows for exposure of the tendon ends (Fig. 15B). If part of the A2 or A4 pulley has been destroyed, clean up the ends and preserve as much as possible. If the entire pulley has been destroyed, then it has to be rebuilt. The A2 pulley is functionally more needed than the other pulleys, so if either the proximal and/or distal one-fourth of the pulley are injured, one may consider partial pulley reconstruction. If the laceration is between the A2 and A4 pulleys, then open one side of the sheath for wider exposure without injuring the A2 or A4 pulleys. When there is a complete laceration of both tendons during a grabbing activity, the tension from the muscle may pull the proximal tendons back into the palm (44). If they are not in the sheath or easily milked out with the wrist and meta-carpophalangeal (MP) joints in flexion, then exploration of
P.690

the palm is necessary. Exposure of the palm through a zigzag incision allows both tendons to be popped into the operating field to place sutures that could be advanced through the flexor sheath a short distance. If a greater distance than a mosquito hemostat will reach is needed, then the tendons should not be exposed. A catheter should be used that is advanced through the tendon sheath. The tendons are then tied to the side of the advancing catheter and pulled out distally (72) (Fig. 15C). Either the catheter can be used to hold the tendon out distally while core sutures are placed, or a 25-gauge needle can be placed through the tendon and a pulley to keep the tendon from retracting while the core sutures are put into place. The distal joints must be flexed to find the distal tendon stumps. If exposure is not adequate, then a distal opening into the sheath is needed, and the repair is performed at that site.
FIGURE 15. A: Bruner incision. B: Sheath opening. C: Catheter to find tendon and pull out to distal tendon. FDP, flexor digitorum profundus; FDS, flexor digitorum superficialis.
Zone I injuries entail one of several different injuries to the FDP tendon. The first type is the jersey finger or tendon avulsion with a large fragment of bone (73). If seen early, the tendon has often retracted to the PIP joint where the bone fragment is caught in the superficialis bifurcation. Treatment is an open reduction and internal fixation of the fracture fragment, which repairs the tendon. If the fragment is small or there is a true tendinous avulsion, then the tendon should be reattached to the distal phalanx. Classically, this has been performed by placing a Bunnell or Kessler repair in the distal tendon, making a window in the distal phalanx, and with pull-out sutures tie the repair over a button that sits over the nail (Fig. 16A,B). This pull-out technique has worked well except that early motion cannot be started, there is a risk of infection spreading up the suture through the distal phalanx and to the tendon, and/or the risk of suture rupture if the button gets caught. Patients also do not like the button. Recently, some research laboratories, including mine, have been working with the use of small suture anchors for repairing these tendon avulsions. With the use of two micro Mitek suture anchors in the distal phalanx and two 4-0 Ethibond modified Becker repairs on each side of the tendon, we have demonstrated 70 N of repair strength versus 43 N with a pull-out button repair (74) (Fig. 16C). This repair strength increases to 78 N with acute cycling. This repair has been used by the author in several fingers followed by an active motion postoperative protocol. The patients did very well.
When the injuries are lacerations of the FDP, they can occur either within the last centimeter of the tendon or under the A4 pulley. When there is a distal laceration, the classic treatment was to advance the tendon end and attach it to the bone using the small insertion tail to reinforce the repair. The A5 pulley was excised for exposure. Although Malerich et al. (75) demonstrated that advancement of less than 1 cm was not harmful, most surgeons do not want to weaken the other profundus tendons by this advancement or have the finger develop a flexion contracture. This repair is performed in one of two ways by the author. If the distal stump is small, a 4-0 nylon Kessler repair is placed in the proximal stump and both ends of the suture are attached to
P.691

two Keith needles (Fig. 17A). Instead of driving the needles through the bone to the nail and therefore advancing the tendon, the needles are pulled through the distal stump and out the tip of the finger, where the suture is tied over a button at the right tension (Fig. 17B). The button can still cause a problem, but this is a good technique to have in reserve fore the infrequent case in which the next repair cannot be performed. Repairing the tendon ends with a locked cruciate repair and a volar epitendinous 6-0 nylon repair is preferred by the author. To perform this repair technique, one needs enough distal tendon to place three suture crossings, which requires approximately 0.75 cm. Repairs under the A4 pulley are performed with a locked cruciate repair and circumferential epitendinous repair technique. The trick is advancing the proximal tendon distal enough to perform the repair distal to the pulley. However, part of the A4 pulley is often injured. This should be excised and cleaned up. If more than half of the pulley is lost, consider a reconstruction with a portion of the dorsal wrist retinaculum.
FIGURE 16. A: Example of a profundus avulsion. B: Sketch of button. C: Sketch of micro Mitek. FDP, flexor digitorum profundus.
Zone II injuries are only more complicated than zone I inasmuch as there are two tendons potentially injured and, therefore, more scar tissue (Fig. 18A). Both tendons must glide to maintain both PIP and distal IP (DIP) joint flexion for optimal function. After the tendon ends have been brought into the same window between the pulleys, as indicated in Figure 14, the author repairs the FDS first. For most of this zone, the superficialis tendon is divided into two slips.
P.692

These slips fan out and twist around the profundus tendon until the two slips merge to insert on the middle third of the middle phalanx. In the past, many surgeons placed simple mattress sutures into the slips to approximate them and did not worry about the strength needed to protect the repair. Schuind et al. (20) have demonstrated that the superficialis tendon sees an average of 17 N of force with average index finger pinch; therefore, this repair might not be strong enough for early motion. In the author’s research laboratory, a modified Becker repair has been placed on each slip with the use of a simple suture needed to prevent rotation of the slip (Fig. 18B). This repair has 58 N of strength compared to the mattress’s 20 N (76). This repair holds the slip in better approximation so that the superficialis opening, or Camper’s chiasm, does not become narrowed around the FDP. If only one slip of the superficialis is lacerated, then that side of the tendon is excised to open the chiasm. If the second slip is injured, only one side is repaired (83). Next, the FDP is repaired with a 4-0 Ethibond or locked cruciate and 6-0 nylon circumferential repair technique.
FIGURE 17. Profundus pullout through finger tip. A: Kessler repair to proximal profundus end attached to two Keith needles that are pushed through the distal profundus and out the distal tip of the finger. B: Suture tied over a button at the tip. FDP, flexor digitorum profundus.
The author treats flexor pollicis longus injuries the same way as FDP lacerations with insertion into the distal phalanx or tendon repair between pulleys. An attempt is made to preserve or reconstruct the oblique pulley over the proximal phalanx of the thumb.
Partial tendon lacerations have been studied. It has been determined that if the laceration is less than 50% or 60% of the tendon, then repairing the tendon actually weakens the tendon (77,78,79 and 80). However, all of the studies create lacerations from one side to the other and not the volar part, as occurs in trauma. The epitenon surrounds the tendon and also divides it into two fascicles. One is larger than the other; 60/40 according to the literature (77). The research studies have been performed with side-to-side laceration that cuts only one fascicle or breaks into the second to test the strength of the tendon over time. This study determined that as long as the smallest fascicle is intact (40%), then the tendon strength remains adequate. When one fascicle is left intact, there is still a surrounding internal epitenon that completes a Chinese finger trap (Fig. 3B) and provides the strength. Most clinical wounds create a volar laceration of the tendon, breaking the epitenon of both fascicles. The author believes that this significantly weakens both fascicles of the tendon. The author’s entire approach
P.693

to partial tendon lacerations has been changed by two recent almost simultaneous events. Al-Qattan (80) wrote in the November 2000 Journal of Hand Surgery about his experience with exploration, trimming, and protected active range of motion for partial tendon lacerations (up to 90%). The author’s laboratory studied the strength of 90% partial tendon lacerations on an anatomic rupture model and demonstrated that the remaining tendon strength after cycling was 140 N, more than twice the strength of a tendon with a four-strand primary repair. For these reasons, the author agree with Al-Qattan that all lacerations of the digit need an exploration by a knowledgeable physician to determine if there is a partial laceration and if so how much of the tendon is involved and whether there is triggering and/or trapping with active motion. The tendon may need trimming, and the sheath can be closed. Protected active range of motion is started in a splint that prevents full extension of the finger for 4 weeks.
FIGURE 18. A: Superficialis tendon sketch. B: Sketch of repair of superficialis tendon. FDS, flexor digitorum superficialis.
Finally, in zone II the surgeon must consider the tendon sheath and pulleys. Based on the research that demonstrates that the predominant means of nutrition to the tendon is diffusion, it makes sense to repair the sheath that provides the nutrients (14,15,16 and 17). Lister (71) is the main proponent for closing the sheath, but the clinical studies to date are contradictory and fail to help with the decision. Therefore, if the sheath is easy to close without narrowing the canal, the author repairs it with 6-0 dissolving suture. If there is any risk of strangulation of the tendon, it is left open. To have an effective pulley system that allows for full flexion of the repaired tendons, the patient needs as much of the A2 and A4 pulley as possible. The other pulleys can be opened and even excised if destroyed by the injury. If the distal part of the A2 pulley plus the A3 pulley is missing, severe bowstringing across the PIP joint causes a flexion contracture and loss of full finger flexion unless the distal A2 pulley is reconstructed. If less than one-third of the A2 and A4 pulley remains, a reconstruction using the dorsal retinaculum (81) or the PL is advised (Fig. 19A,B).
Postoperatively, the patients are placed in a classic clam digger’s splint, with the wrist in slight flexion, the MP fully
P.694

flexed, and both the PIP and DIP joints straight (Fig. 20A,B). The patients are seen again within 2 to 3 days and placed in a Kleinert palmar bar rubber band splint for exercises at home. At night, they come out of the rubber bands and have their fingers held in place using Velcro straps, with both the PIP and DIP joints straight to prevent flexion contractures. They attend hand therapy daily, if possible, for protected active range of motion. The author starts with Strickland’s (59) block and hold technique and moves to wrist extension and active IP flexion, with the MP joint held in 40 degrees of
P.695

flexion. Protected blocking exercises can be started when the edema is down.
FIGURE 19. A: Dorsal retinacular reconstruction. Half of the dorsal retinaculum is harvested and sutured synovial layer down to the edges of the proximal or middle phalanx, as there is usually a scar rim of the original pulley. B: Palmaris longus reconstruction. The tendon is harvested from the forearm. The tendon is attached to the rim of the proximal or middle phalanx, passed over the tendon and around the phalanx, through both rims at an oblique angle, and then around the bone again. This gives a broad pulley, but is not synovial.
FIGURE 20. Exercises in clam digger’s splint. A: Rubber band extension exercise. B: Rubber band flexion exercise.
Classically, this repair is performed with a Kessler core and epitendinous repair. Then, either Kleinert rubber band motion or Duran passive motion is used postoperatively. The results range from 75% to 95% good and excellent. As Strickland (59) pointed out, stronger four-strand repair techniques provide for the possibility of early active protected motion. No randomized study between early active motion and passive motion has been reported.
Postoperative protocols must be changed for children. The author places young children in clam diggers’ long-arm casts and older children in clam diggers’ short-arm casts. The wrist is left in slight extension, the MP joint in 40 degrees of flexion, and both the PIP and DIP joints straight. A thick dressing is placed volarly under the fingers from the PIP joint distally. This dressing is pulled in the operating room after the cast has hardened. This provides an inch or so room for the fingers to be wiggled in a protected shell. The author has seen no ruptures with this technique.
Zone III injuries almost always include injuries to neurovascular structures and require a wide exposure, including opening the carpal canal to find the retracted tendons and nerves. Zone IV injuries require opening the carpal canal to have the exposure to repair the tendons and median nerve. When the carpal tunnel is open, it should be performed in a way as to be able to close the transverse carpal ligament at the end of surgery to prevent bowstringing of the tendons during the postoperative period of immobilization, particularly if the wrist is to be held in slight flexion as it is in most postoperative motion protocols. The transverse carpal ligament is opened in its proximal half on the radial side, making sure to protect the remainder of the median nerve (Fig. 21). The incision then crosses to the ulnar side, where the ligament opening is completed to protect the motor branch of the median nerve. After repair of the flexor tendons and nerve, if necessary, the two ends of the transverse carpal ligament are swung together and closed with a 2-0 dissolving suture. Occasionally, cutbacks are needed to complete the mobilization of each side of the transverse carpal ligament.
The trick to the repair in zone III and IV is properly identifying all the structures and repairing them in an orderly fashion. All of the structures, tendon, nerve, and arteries, should be repaired appropriately. The nerves and arteries require the use of an operating microscope to be available. These structures are repaired after the tendons, unless the ulnar artery is repaired first for a devascular hand. The author repairs the tendon as above with locked cruciate and circumferential epitendinous techniques. The major problem with this repair is that it is time-consuming for repairing many tendons. The results are worth the time, in the author’s opinion. Other four-strand techniques could work, but have not been tested for strength anatomically. In zone III, the lumbric origins are wrapped around the profundus repairs for a better blood supply and to reapproximate the lumbrical tension.
Zone IV injuries also include the wrist flexor tendons. When possible, the author repairs the FCR and FCU tendons with the locked cruciate technique, because the postoperative protocol requires at least passive flexion-extension of the wrist.
After the transverse carpal ligament is repaired, the wrist is splinted in neutral, with the MP joint in 40 degrees of flexion and both the PIP and DIP joints straight. A similar postoperative protocol is performed for zones I and II injuries. The tendon repairs of zone III injuries have traditionally done better than those for other flexor tendon injuries. The problem with zone IV repairs is that there are many tendons repaired in a relatively devascularized area once the synovium has been removed; therefore, there is a tendency for the tendon repairs to bind together. The excursion of the tendons is not as great at the wrist as it is in the finger, so that differential gliding is hard to obtain. Hopefully, with the stronger repair techniques and early active motion postoperatively, the results will improve.
FIGURE 21. Carpal tunnel release by Z-plasty and closure: see text for description.
Zone V injuries require wide exposure in the intramuscular planes, so as not to further injure the muscles. To obtain adequate visualization of the structures in the forearm, as there is usually a large intra- and extramuscular hematoma, irrigating the wound with 50% hydrogen peroxide and 50% saline is recommended, followed by a vigorous saline washout. This lyses the blood and allows for good visualization without injuring the muscle. All deep neurovascular structures are identified and repaired, if necessary. In the distal part of the muscle, thin tendinous structures can be milked out of the muscle bellies to provide a structure that holds a suture. The tendons are repaired with a modified Kessler technique, and the muscle fascia is repaired with a 4-0 dissolving suture. The arm, including the elbow, is immobilized for 3 weeks to allow for muscle healing before starting an exercise program. The results of this repair technique depend on whether the median or the ulnar nerve, or both, have been injured and how proximally the muscle has been denervated.
The ideal repair of any tendon allows for early, unprotected range of motion, because in many parts of the world hand therapy is not available or not practiced. In developed countries, hand therapy is available, but the repairs are less than ideal.
COMPLICATIONS
There are two major complications that occur during the postoperative period after the repair of flexor tendons: (a) rupture of the tendon repair and (b) adhesions causing decreased finger motion. There is also a third complication that can be seen when using the rubber band splinting technique. If the rubber bands are left on all the time, some patients develop significant PIP joint flexion contractures. It is important to have the patients remove the rubber bands nightly and use the Velcro straps to hold the fingers in extension during the night.
Tendon rupture is derived from failure of the repair and can occur with decreasing frequency as late as 8 weeks after surgery. This problem is reported at approximately 5% in the literature (45,46 and 47). Failure can occur because the repair technique is too weak for the postoperative protocol, because patients do not protect their repair and allow free full extension of the fingers before 6 weeks or hyperextension before 8 weeks, or because patients try to use their hand for strong resistive grip or pinch. If a finger has been moving and suddenly loses its motion and the arcade of the hand is broken, then a tendon rupture should be suspected. Acute attention to this problem is needed, and the repair should be redone. If the surgeon is comfortable with the new repair, then there should be no change in the postoperative protocol, except to
P.696

perform all motion under strict supervision. If the repair is weak, then traditional immobilization is needed, with the realization that a tenolysis is probably necessary to obtain functional motion from that finger. If the tendon is too soft to repair, then a primary graft can be placed (see Chapter 38).
Tendon adhesions form from the surrounding tissues in an attempt to help the tendon heal. Lister et al. (45) reported this occurred approximately 20% of the time. This was the classic way tendons were allowed to heal, because the sheath was removed above the repair, and the tendon was immobilized for 3 weeks. With early motion protocols, the idea is to either eliminate the surrounding adhesions or have them develop soft and elongated so that they do not block motion of the tendon. However, no matter what surgeons do, a percentage of patients develop firm adhesions either between the tendons themselves or the tendons and the sheath. The author believes that to perform a tenolysis one must wait for tissue balance before reoperating on the tendon. If performed too early, the skin incision may break down during the active rehabilitation, or the tendon will not have intrinsically healed and the removal of the adhesions allows the tendon to rupture. The author performs tenolysis no sooner than 4 months postoperatively and/or when the skin is soft and supple and the patient has near full passive range of motion of the joints.
The author prefers to perform tenolysis procedures under local anesthesia with sedation. A forearm tourniquet is used for the exposure and initial tendon release. The pulleys are carefully preserved by making windows between the pulleys to separate the tendons from the surrounding structures, tendon, pulley, and bone. The hardest part of the surgery is to separate the two tendons from each other without damaging one or both. Once the surgeon thinks that an adequate release has been performed and there is gliding of each tendon with passive finger motion, then the tourniquet is dropped. While hemostasis is obtained, the sedation is stopped. After approximately 10 minutes, the patient should be able to actively flex the finger or fingers. If there is a blockage to flexion, then further exploration is warranted. An 18-gauge Angiocath is placed in the carpal canal and connected with a short piece of sterile extension tubing. The patient is shown how to inject 3 to 4 cc of 0.5% Marcaine into the carpal canal every 4 to 6 hours. This provides excellent pain relief while the patient performs active flexion exercises. The patient should have a dorsal splint for several weeks to prevent hyperextension of the fingers and possible tendon rupture. Patients are then instructed in blocking exercises to obtain individual PIP and DIP joint flexion. The patient also needs supervised therapy. Any time that a surgeon attempts a tenolysis, he or she should be prepared to remove the scarred tendons and place a Hunter silicone rod under the pulleys. Any pulleys that are injured during the procedure should also be reconstructed. These procedures are described in Chapter 38.
REFERENCES
1. Bunnell S. Repair of tendons in the fingers and description of two new instruments. Surgery Gynecol Obstet 1918;26:103–110.
2. Verdan CE. Primary repair of flexor tendons. J Bone Joint Surg 1960;42A:647–657.
3. Verdan CE. Practical considerations for primary and secondary repair in flexor tendon injuries. Surg Clin North Am 1964;44:951–970.
4. Kleinert, HE, Kutz JE, Ashbell TS, Martinez E. Primary repair of lacerated flexor tendons in “no man’s land” [abstract]. J Bone Joint Surg 1967;49A:577.
5. Kessler I, Nissim F. Primary repair without immobilization of flexor tendon division within the digital sheath: an experimental and clinical study. Acta Orthop Scand 1969;40:587–601.
6. Gelberman RH, Goldberg V, An KN, Banes A. Tendon. In: Woo SLY, Buckwalter JA, eds. Injury and repair of the musculoskeletal soft tissues. Park Ridge, IL: American Academy of Orthopaedic Surgeons, 1988.
7. Armenta E, Lehrman A. The vincula to the flexor tendons of the hand. J Hand Surg 1980;5:127–134.
8. Caplan HS, Hunter JM, Merklin RJ. Intrinsic vascularization of flexor tendon. In: AAOS symposium on tendon surgery in the hand. St. Louis: Mosby, 1975.
9. Lundborg G, Myrhage R. The vascularization and structure of the human digital tendon sheath as related to flexor tendon function. An angiographic and histological study. Scand J Plast Reconstr Surg 1977;11:195–203.
10. Lundborg G, Myrhage R, Rydevik B. The vascularization of human flexor tendons within the digital synovial sheath region—structural and functional aspects. J Hand Surg 1977;2:417–427.
11. Brand PW, Beach RB, Thompson DE. Relative tension and potential excursion of muscles in the forearm and hand. J Hand Surg 1981;6:209–219.
12. Kline SC, Moore JR. The transverse carpal ligament. An important component of the digital flexor pulley system. J Bone Joint Surg 1992;74:1478–1485.
13. Kang HJ, Lee SG, Phillips CS, Mass DP. Biomechanical changes of cadaveric finger flexion: the effect of wrist position and of the transverse carpal ligament and palmar and forearm fasciae. J Hand Surg 1996;21A:963–968.
14. Manske PR, Bridwell K, Lesker PA. Nutrient pathways to flexor tendons of chickens using tritiated proline. J Hand Surg 1978;3:352–357.
15. Manske PR, Lesker PA. Nutrient pathways of flexor tendons in primates. J Hand Surg 1982;7:436–444.
16. Manske PR, Lesker PA. Flexor tendon nutrition. Hand Clin 1985;1:13–24.
17. Manske PR, Whiteside LA, Lesker PA. Nutrient pathways to flexor tendons using hydrogen washout technique. J Hand Surg 1978;3:32–36.
18. Idler RS. Anatomy and biomechanics of the digital flexor tendons. Hand Clin 1985;1:3–11.
19. Urbaniak JR, Cahill JD, Mortenson RA. Tendon suturing methods: analysis of tensile strengths. In: AAOS symposium on tendon surgery in the hand. St. Louis: Mosby, 1975.
20. Schuind F, Garcia-Elias M, Cooney WP, An KN. Flexor tendon forces: in vivo measurements. J Hand Surg 1992;17A:291–298.
P.697

21. Potenza AD. Tendon healing within the flexor digital sheath in the dog. J Bone Joint Surg 1962;44A:49–64.
22. Potenza AD. Critical evaluation of flexor tendon healing and adhesion formation within artificial digital sheaths. J Bone Joint Surg 1963;45A:1217–1233.
23. Potenza AD. Mechanisms of healing of digital flexor tendons. Hand 1969;1:40–41.
24. Peacock EE. Fundamental aspects of wound healing relating to the restoration of gliding function after tendon repair. Surg Gynecol Obstet 1964;119:241–250.
25. Peacock EE. Biological principles in the healing of long tendons. Surg Clin North Am 1965;45:461–476.
26. Lundborg G, Rank F. Experimental intrinsic healing of flexor tendons based upon synovial fluid nutrition. J Hand Surg 1978;3:21–31.
27. Lundborg G, Rank F. Experimental studies on cellular mechanisms involved in healing of animal and human flexor tendon in synovial environment. Hand 1980;12:3–11.
28. Lundborg G, Hansson HA, Rank F, Rydevik B. Superficial repair of severed flexor tendons in synovial environment: an experimental, ultrastructural study on cellular mechanisms. J Hand Surg 1980;5:451–461.
29. Lundborg G, Rank F, Heinau B. Intrinsic tendon healing. A new experimental model. Scand J Plast Reconstr Surg 1985;19:113–117.
30. Becker H, Graham MF, Cohen IK, Diegelmann RF. Intrinsic tendon cell proliferation in tissue culture. J Hand Surg 1981;6:616–619.
31. Manske PR, Gelberman RH, Van de Berg JS, Lesker PA. Intrinsic flexor-tendon repair. A morphological study in vitro. J Bone Joint Surg 1984;66A:385–396.
32. Mass DP, Tuel RJ. Intrinsic healing of the laceration site in human superficialis flexor tendons in vitro. J Hand Surg 1991;16A:24–30.
33. Mass DP, Tuel RJ. Human flexor tendon participation in the in vitro repair process. J Hand Surg 1989;14A:64–71.
34. Manske PR, Bridwell K, Whiteside LA, Lesker PA. Nutrition of flexor tendons in monkeys. Clin Orthop 1978;136:294–298.
35. Manske PR, Lester PA, Bridwell K. Experimental studies in chickens on the initial nutrition of tendon grafts. J Hand Surg 1979;4:565–575.
36. Manske PR, Lester PA. Comparative nutrient pathways to the flexor profundus tendons in zone II of various experimental animals. J Surg Res 1983;34:83–93.
37. Gelberman RH, Van de Berg JS, Lundborg GN, Akeson WH. Flexor tendon healing and restoration of the gliding surface: an ultrastructural study in dogs. J Bone Joint Surg 1983;65A:70–80.
38. Gelberman RH, Woo SLY, Lothringer K, et al. Effects of early intermittent passive mobilization on healing canine flexor tendons. J Hand Surg 1982;7:170–175.
39. Gelberman RH, Manske PR. Factors influencing flexor tendon adhesions. Hand Clin 1985;1:35–42.
40. Gelberman RH, Botte MJ, Spiegelman JJ, Akeson WH. The excursion and deformation of repaired flexor tendons treated with protected early motion. J Hand Surg 1986;11A:106–110.
41. Gelberman RH, Woo SLY, Amiel D, et al. Influences of flexor sheath continuity and early motion on tendon healing in dogs. J Hand Surg 1990;15A:69–77.
42. Mason ML, Allen HS. The rate of healing of tendons: an experimental study of tensile strength. Ann Surg 1941;113:424–456.
43. Kleinert HE, Kutz JE, Atasoy E, Stormo A. Primary repair of flexor tendons. Orthop Clin North Am 1973;4:865–876.
44. Kleinert HE, Forshew FC, Cohen MJ. Repair of zone I flexor tendon injuries. In: AAOS symposium on tendon surgery in the hand. St. Louis: Mosby, 1975.
45. Lister GD, Kleinert HE, Kutz JE, Atasoy E. Primary flexor tendon repair followed by immediate controlled mobilization. J Hand Surg 1977;2:441–451.
46. Duran RJ, Houser RG. Controlled passive motion following flexor tendon repair in zones II and III. In: AAOS symposium on tendon surgery in the hand. St. Louis: Mosby, 1975.
47. 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 1980;5:537–543.
48. Strickland JW. Flexor tendon injuries: I. Foundations of treatment. J Am Acad Orthop Surg 1995;3:44–54.
49. Savage R. In vitro studies of a new method of flexor tendon repair. J Hand Surg 1985;10B:135–141.
50. Savage R, Risitano G. Flexor tendon repair using a “six strand” method of repair and early active mobilisation. J Hand Surg 1989;14B:396–399.
51. Bunnell S. Repair of tendons in the fingers. Surg Gynecol Obstet 1922;35:88–97.
52. Hotokezaka S, Manske PR. Differences between locking loops and grasping loops: effects on 2-strand core suture. J Hand Surg 1997;22A:995–1003.
53. Trail IA, Powell ES, Noble J. An evaluation of suture materials used in tendon surgery. J Hand Surg 1989;14B:422–427.
54. Becker H, Davidoff M. Eliminating the gap in flexor tendon surgery. A new method of suture. Hand 1977;9:306–311.
55. Greenwald DP, Randolph MA, Hong H-Z, May JW Jr. Augmented Becker versus modified Kessler tenorrhaphy in monkeys: dynamic mechanical analysis. J Hand Surg 1995;20A:267–272.
56. Lee H. Double loop locking suture: a technique of tendon repair for early active mobilization. Part I. evolution of technique and experimental study. J Hand Surg 1990;15A:945–952.
57. McLarney E, Hoffman H, Wolfe SW. Biomechanical analysis of the cruciate four-strand flexor tendon repair. J Hand Surg 1999;24A:295–301.
58. Robertson GA, Al-Qattan MM. A biomechanical analysis of a new interlock suture technique for flexor tendon repair. J Hand Surg 1992;17B:92–93.
59. Strickland JW. Flexor tendon repair: Indiana method. The Indiana Hand Center Newsletter 1993;1:1–12.
60. Lim BH, Tsai TM. The six-strand technique for flexor tendon repair. Atlas Hand Clin 1996;1:65–76.
61. Sandow MJ, McMahon MM. Single cross-grasp six-strand repair for acute flexor tenorrhaphy: modified Savage technique. Atlas Hand Clin 1966;1:41–64.
62. Silva MJ, Hollstien SB, Fayazi AH, et al. The effects of multiple-strand suture techniques on the tensile properties of repair of the flexor digitorum profundus tendon to bone. J Bone Joint Surg 1998;80A:1507–1514.
P.698

63. Stein T, Ali A, Hamman J, Mass DP. A randomized biomechanical study of zone II flexor tendon repairs analyzed in an in vitro model. J Hand Surg 1998;23A:1046–1051.
64. Aoki M, Pruitt DL, Kubota H, Manske PR. Effect of suture knots on tensile strength of repaired canine flexor tendons. J Hand Surg 1995;20B:72–75.
65. Pruitt DL, Aoki M, Manske PR. Effect of suture knot location on tensile strength after flexor tendon repair. J Hand Surg 1996;21A:969–973.
66. Leddy JP. Flexor tendons—acute injuries. In: Green DP, eds. Operative hand surgery. New York: Churchill Livingstone, 1993.
67. Kubota H, Aoki M, Pruitt DL, Manske PR. Mechanical properties of various circumferential tendon suture techniques. J Hand Surg 1996;21B:474–480.
68. Diao E, Hariharan JS, Soejima O, Lotz JC. Effect of peripheral suture depth on strength of tendon repairs. J Hand Surg 1996;21A:234–239.
69. Reference deleted.
70. Bruner JM. The zig-zag volar-digital incision for flexor tendon surgery. Plast Reconstr Surg 1967;40:571–574.
71. Lister G. Indications and techniques for repair of the flexor tendon sheath. Hand Clin 1985;1:85–95.
72. Sourmelis SG, McGrouther DA. Retrieval of the retracted flexor tendon. J Hand Surg 1987;12B:109–111.
73. Leddy JP, Packer JW. Avulsion of the profundus tendon insertion in athletes. J Hand Surg 1977;2:66–69.
74. Brustein M, Pellegrini J, Choueka J, et al. Bone suture anchors versus the pullout button for repair of distal profundus tendon injuries: a comparison of strength in human cadaveric hands. J Hand Surg (in press).
75. Malerich MM, Baird RA, McMaster W, Erickson JM. Permissible limits of flexor digitorum profundus tendon advancement—an anatomic study. J Hand Surg 1987;12A:30–33.
76. Miller L, Mass DP. A comparison of four repair techniques for Camper’s chiasma flexor digitorum superficialis lacerations: tested in an in vitro model. J Hand Surg 2000;25A:1122–1126.
77. Bishop AT, Cooney WP, Wood MB. Treatment of partial flexor tendon lacerations: the effect of tenorrhaphy and early protected mobilization. J Trauma 1986;26:301–312.
78. Chow SP, Yu OD. An experimental study on incompletely cut chicken tendons—a comparison of two methods of management. J Hand Surg 1984;9B:121–125.
79. McGeorge DD, Stilwell JH. Partial flexor tendon injuries: to repair or not. J Hand Surg 1992;17B:176–177.
80. Al-Qattan MM. Conservative management of zone II partial flexor tendon lacerations greater than half the width of the tendon. J Hand Surg 2000;25A:1118–1121.
81. Lister GD. Reconstruction of pulleys employing extensor retinaculum. J Hand Surg 1979;4:461–464.
82. Angeles JG, Heminger H, Mass DP. Comparative biomechanical performances of 4-strand core suture repairs for zone II flexor tendon lacerations. J Hand Surg 2002;27A:508–517.
83. Paillard PJ, Amadio PC, Zhao C, et al. Pulley plasty versus resection of one slip of the flexor digitorum superficialis after repair of both flexor tendons in zone II. J Bone Joint Surg 2002;84A:2039–2045.