58 Stephens et al. Surgical Preparation Tendon repair should be performed under tourniquet control and an appropriate anes- thetic. The existing skin lacerations should be extended in a manner that balances access, flap vascularity, and limitation of wound length. Avoidance of longitudinal incisions passing volar to the joint axis is important in minimizing subsequent flexion contracture. Most often, a zigzag volar pattern incision is employed, as described by Bruner (56). The utmost care should be taken to identify and preserve the neurovascular bundles. The tendon sheath should be opened enough to allow delivery of both proximal and distal tendon stumps. Flexing the wrist and fingers and even “milking” the palm from a proxi- mal to distal direction to deliver a retracted proximal tendon stump is a way to facilitate this. Fine artery forceps passed down the flexor sheath can also be used; however, avoid blindly grasping down the sheath. A thrombus visible within the sheath indicates that the tendon stump is not too far away as these are often found on the end of the stump. If such measures fail to produce the tendon stump, an incision should be made within the palm or even at the wrist level (if necessary) to find it. Once found, it can then be passed distally down the sheath with the help of a fine catheter, such as an infant-feeding tube. Once both tendon stumps are within the surgical field, they can be secured in posi- tion by passing a fine needle through each and adjacent soft tissues, taking care to avoid the neurovascular bundles. If the tendon ends are jagged, these can be “cleaned” by excising a minimal amount, ensuring that the tendons are not shortened too much as this will result in a permanent finger flexion contracture. In the uncommon situation where the tendon is lacerated at a very oblique angle (greater than 10 mm of the length of the tendon) the repair could be performed using the Becker method (57). This method was proposed as a means of repairing all tendon injuries via beveling the ends, and although it provided a strong repair, it created a problematic shortening of the tendon and is therefore no longer routinely used. An alternate method of such oblique lacera- tions is to excise a minimal amount from each end to reduce the angle of the point of division. Then proceed to repair the tendon with both a “core” and a “peripheral” su- ture, as this is the preferred technique for all flexor tendon injuries. The aim is to have a repair site that is smooth enough to allow minimal gliding resistance within the sheath, and strong enough to withstand early mobilization so it does not rupture or allow significant gap formation. The gap retarding qualities of flexor tendon repair techniques are important; studies have shown that gap formation of as little as 1–3 mm can result in increased peritendinous adhesion formation and adverse clinical outcomes (58–60). Core Repair The most widely recognized form of tendon repair is that described by Kessler in 1973 (61). It is a two strand repair with two lateral “grasping” components on each side of the tendon division. Pennington (62) described the “locking” loop modification of the Kessler repair, where the transverse component of the suture passes anterior to the longitudinal strands, locking onto a bundle of tendon fibers when the tendon is subject to tensile forces. Locking repairs have been shown to afford more strength to the repair than simple “grasping” loops (63–65). Hatanaka and Manske (66) further demonstrated that increasing the cross-sectional area of the locking component increased the strength of the repair. Modified forms of Tendons of the Hand 59 the Kessler technique are probably the most commnonly applied surgical techniques in tendon repair. Numerous other techniques have been described, all of which aim to provide a stron- ger repair (48,50,67–74). One typical theme from these techniques is that the strength of the repair is proportional to the number of strands crossing it, with a four-stranded repair being twice as strong as a two-stranded repair and, consequently, a six-stranded repair being almost three times stronger. Several authors have even described eight- stranded repair techniques (73,74). Unfortunately, increasing the strands per repair also increases the techniqual complexity and, more importantly, the bulk of the repair. Therefore, many of these multistrand techniques have not yet been used in common practice. Four-stranded repair techniques are usually favored because they provide sufficiently increased strength without the bulk or tissue handling requirements of six- or eight- stranded repairs (48,67,68,71). Kubota et al. (67) described a four-strand single-knot modified Kessler repair, which provides four “locking” segments on either side of the tendon division. However, McLarney’s (72) four-strand cruciate repair is perhaps ideal. Although it is not as strong as Kubota’s four-strand modified Kessler technique (65), it does provide the strength of a four-strand repair with the simplicity and tendon han- dling requirements of only a two-stranded repair. Suture material and caliber are also important aspects of tendon repair. Various materials have been employed for the core suture. Larger caliber sutures are obviously stronger than those that are smaller. But again, a balance between suture size and mini- mal bulk must be met. For this reason, 3/0 or 4/0 suture material is generally preferred. The ideal suture material would be a monofilament for handling ease (to allow easy sliding within the tendon), maximal stiffness (to minimize the amount of stretch), and bioresorbable. Unfortunately, no suture material yet fulfils all these criteria. Uncertainties regarding the longevity of bioresorbable sutures, and the tissue reac- tion they may provoke during resorption, has generally precluded their use. Prolene and, to a lesser extent, Nylon, are typically the perferred sutures, as they are nonresorbable monofilaments and are thus easy to handle. However, they are flawed by their relatively poor stiffness qualities, resulting in a greater potential for stretch, with subsequent gap formation and repair failure. Braided sutures, e.g., Ticron and Ethibond, confer greater stiffness when compared with Prolene but can be difficult to handle, especially when using four or more strands in the repair. This difficulty relates to their inability to “slide” within the tendon once placed; therefore, adjusting indi- vidual suture strand lengths and tension is extremely difficult. This is an important issue, as Trail et al. (75) demonstrated that differential suture strand loading could reduce the repair strength. Ketchum et al. (76) showed that Supramid, a polyfilament ensheathed caprolactan (Supramid, S. Jackson, Inc., Alexandria, VA), is a good suture material. It is nonre- sorbable, slides within the tendon well, and has minimal extensibility. It is also one of the few sutures available with two strands per needle, thus making multistrand repairs relatively easy. Another area of debate is the placement of the suture knot. Aoki et al. (77) have shown using canine cadaver tendons that knots placed within the repair site can reduce the strength of the repair. They also demonstrated that the greater the number of knots, 60 Stephens et al. the worse the tensile strength. Using a canine in vivo study, Pruitt et al. (53) also showed that the early tendon tensile strength was worse when knots were placed within the repair site. However, at 6 wk, there was no significant difference between those that had knots placed either within or outside of the repair site. Despite the early biomechanical advantages of knot placement outside of the repair site, this does result in greater gliding resistance (78), with knots possibly catching on the tendon pulleys. McLarney’s (72) original description of the cruciate four-strand repair used an intratendinous placement of the suture knot, consequently avoiding these issues. However, in our experience, most surgeons prefer to place the knot between the two tendon ends. Ideally, utilizing a single continuous suture; thus, only one knot is required. PERIPHERAL SUTURE Much to the disappointment of the surgeon new to flexor tendon repair, the diagram- matic perfection of coaptation is not usually met in practice. To this end, a circumferential suture was used, originally as a tidy-up suture (35). However, it has been shown to confer significant biomechanical advantages to the overall integrity of the repair, improving the resistance to gap formation and providing as much as 50% of the ultimate load to failure and stiffness (3,67,68,79–81). This makes it an important consideration, particularly in those patients likely to commence on an early finger mobilization program. Although originally described as an epitenon repair (35), as it only involved the epi- tenon, Mashadi and Amis (82) showed the importance of the suture passing through the tendon fibers. Therefore, it is preferably referred to as a circumferential component of repairs, because all currently advocated methods pass through the tendon fibers. The suture material of choice is generally a 6/0 or even a 5/0 monofilament, such as Prolene. The simple running technique is likely the most commonly employed circumferen- tial suture. The suture strands of this repair run along the line of the tendon fibers. Attempts to improve this technique have been described with Diao et al. (79), who found an 80% increase in repair strength when the bite depth was increased. Several other modifications of the simple running repair have been described but are not rou- tinely used by most surgeons (83). Another method that has widespread use is the cross-stitch repair, described by Silfverskiold and Anderson in 1993 (84). The transverse component of this method pro- vides a stronger hold on the tendon as it runs perpendicular to the orientation of the tendon fibers. A further development is the interlocking horizontal mattress technique, recently described by some authors (85). This relatively easy method “locks” around the periphery of the tendon when subject to tensile forces and has been shown to be biome- chanically superior to the simple running and cross-stitch repairs in vitro. Various other horizontal mattress methods have been reviewed that provide greater biomechanical prop- erties in comparison to the simple running suture (35,82,86,87); however, many are flawed by their complexity in design and are therefore not routinely used by many sur- geons. Like the core suture, studies have demonstrated that increasing the number of strands crossing the repair site increases the repair strength (53). SHEATH AND PULLEY Every effort should be made to repair all aspects of the sheath, but this can be techni- cally challenging. Theoretically, this would be expected to reduce peritendinous adhe- Tendons of the Hand 61 sion formation, yet, no strong evidence exists to support this. If a decision to repair has been made, then this should be with a nonresorbable suture, such as 6/0 Prolene. Evidence exists to encourage preserving and repairing the pulleys; as discussed earlier, they are essential for normal finger functioning (10,78,88–90). An absent or incompetent pulley results in an increased moment arm, owing to tendon bowstringing, requiring increased tendon excursion to produce the same arc of motion. Clinically, this is associated with the loss of power, reduced range of motion, the risk of addi- tional pulley rupture, and the development of fixed flexion contractures (91). Biomechanically the most important pulleys are the A2 and A4 pulleys (88,89,92). It may be necessary to open (otherwise known as “vent”) one of the pulleys to permit greater access for tendon repair. If this is necessary, then this should be performed dorso-laterally, allowing sufficient dorsal cuff to aid in its closure after the tendon itself has been repaired. A repaired tendon that cannot move passively through its full excursion with minimal resistance is another reason a pulley may need to be vented or even partially excised. Kwai and Elliot (93) showed that in a series of 126 complete zone II tendon transections, venting was required in 64% of A2 (56%) and A4 (8%) pulleys. The mean size of their venting was 52% of the pulley length. Other biome- chanical studies have shown that up to 75% of both the A2 and A4 pulleys can be sacrificed with minimal decrease in the total finger flexion and without significant risk of rupture (94–96). In contrast to the A2 and A4 pulleys, the A3 pulley is considered to be of little biomechanical importance. However, Tang and Xie (97) demonstrated that A3 and the adjacent sheath spanning A2 to A4 does have an important part in restraining tendon bowstringing at the PIPJ. Zone Injuries If sufficient distal stump exists, then tendon repair should be performed in the same manner as described for zone II injuries. However, if the laceration is very distal, or if the FDP has avulsed from the distal phalanx, then a different approach should be taken. The traditional repair method was the pull-out suture tied over a button on the nail plate, originally described by Bunnell (98). As it is a pullout method, the preferred suture is therefore limited to a monofilament, such as Prolene. The duration of removal is usually 6 wk, at which stage, sufficient healing at the tendon bone interface should have occurred. Unfortunately, this technique has many disadvantages; it is unsightly, requires a second, albeit minor, procedure to remove the suture and button, can act as a source of suture tract infection, and it can cause discomfort to the nail region. Addi- tionally, the results of this repair method are poorly reported. Moiemen and Elliot (99) in a series of 14 such repairs, noted that six regained less than 50% of their normal DIPJ flexion. Much like zone II injuries, methods have been proposed to increase the strength of the pull-out repair through various suture techniques or by increasing the suture strands involved (17,100). However, this does not eliminate the problems encountered with pull-out techniques. Small bone anchors have been developed to secure soft tissues to bone, and several have been advocated for use in FDP tendon fixation (101–103). Bone anchors poten- tially make it a simple one-stage repair and allow the use of more than two strands per 62 Stephens et al. repair (103). As suture pull-out is not required, braided, less extensible sutures can be used with the potential for less gap formation at the tendon bone interface. But, like the pull-out button technique, there is still a paucity of clinical trials in the literature. Zones III–V injuries should be repaired in the same manner as zone II injuries already described. They tend to be less challenging in the absence of the digital sheath. Partial Lacerations The treatment of partially lacerated flexor tendons has been controversial. The pos- sible complications associated with an unrepaired partial laceration include trigger- ing, entrapment, or rupture (104–110). Triggering creates discomfort during finger motion, as the laceration site gets caught on the pulley edge. Entrapment is caused by the inability of the laceration site to enter the pulley, resulting in limited range of motion. Several authors have demonstrated that suturing of a partially divided tendon causes a reduction of the tensile strength of the tendon in vivo (111,112). Other authors have shown that the threshold load levels to rupture human flexor tendons with major divi- sions of their cross-sectional area (CSA) of up to 75% are higher than the physiologic load levels measured during active motion, suggesting that these partial lacerations can withstand in vivo forces associated with active mobilization (113). In a clinical study of 34 patients by Wray and Weeks (114), functional results with partial lacerations of up to 95% were excellent in 92% of cases. A similar study of 15 patients by Al-Qattan (115) with a mean partial laceration of 71% (range 55–90%) showed an excellent outcome in 93% at 6 mo. The logical approach to partial flexor tendon lacerations would be to intraopera- tively assess for any triggering. If such triggering exists, and the laceration is less than 75% of the tendon CSA, then the edges should be trimmed or beveled. If the laceration is greater than 75% of the tendon CSA, then a standard core and circumferential suture should be employed to coapt the edges. Although the remaining intact portion of the tendon is almost certainly biomechanically superior than any repair technique, it does facilitate tendon healing and maintains a normal caliber. Extensor Tendon Repair Extensor tendon injuries are more frequently encountered than flexor tendon inju- ries, because they are less protected than flexor tendons. As highlighted previously, their different anatomy entails that the treatment of such injuries differs to that of flexor tendons. The type of injury, surgical approach, and potential deformity vary according to the injury zone. Injuries and their respective treatments are categorized into the eight zones described by Verdan (116). Zone I Injury (Mallet Finger) Disruption of the extensor tendon over the DIPJ produces the characteristic flexion deformity known as a mallet finger. A mallet finger may be open, but it is more com- monly closed. They are classified into four types: Type I: closed, with or without an avulsion fracture. Type II: laceration at or proximal to the DIPJ with loss of tendon continuity. Type III: deep abrasion with loss of skin, subcutaneous cover and tendon substance. Tendons of the Hand 63 Type IV: which is designated into three categories: A: transepiphyseal plate fracture of the distal phalanx in children. B: hyperflexion injury with fracture of the articular surface of 20–50%. C: hyperextension injury with fracture of the articular surface usually greater than 50% and with early or late palmar subluxation of the distal phalanx. Each type of injury warrants a different treatment, but in the majority of cases, splint- ing alone will suffice. Type I: continuous splinting of the DIPJ in full extension for 6 wk, followed by 2 wk of night splinting. Type II: a simple figure-of-eight suture to repair the tendon. The DIPJ is then splinted in extension for 6 wk, followed by 2 wk of night splinting. Type III: immediate soft tissue coverage and primary grafting or late reconstruction using a free tendon graft. Type IV: with the following treatments: A: closed reduction. The extensor mechanism is attached to the basal epiphysis; thus, closed reduction followed by splinting for 3–4 wk results in correction of the defor- mity. B: splinting for 6 wk with 2 wk of night splinting. C: owing to the palmar subluxation, this injury is best managed operatively with open reduction and internal fixation with a Kirschner wire. This should also be protected with a splint for 6 wk, followed by wire removal and DIPJ mobilization. Zone II Injury (Middle Phalanx) A zone II injury is usually secondary to a laceration or crush mechanism. If less than 50% of the tendon width is divided, then treatment involves routine wound care and splintage for 7–10 d, followed by active mobilization. Injuries greater than 50% of the tendon should be repaired with either a continuous running suture or several figure-of- eight sutures using a nonresorbable 4-0 or 5-0 suture, which is followed by 6 wk of splinting. Zone III Injury (Boutonniere Deformity) Disruption of the central slip overlying the PIPJ results in a Boutonniere deformity with loss of extension of the PIPJ and hyperextension of the DIPJ. This may be a closed or open injury. As a general rule, all open injuries over the PIPJ should be explored in the operating theater. An early injury may not necessarily be associated with a Bouton- niere deformity, as this usually develops 10–14 d after the initial injury, especially in closed trauma (117). As the tendon ends do not retract in this area, treatment is rela- tively simple and requires several figure-of-eight sutures using a nonresorbable 4-0 or 5-0 suture, followed by up to 6 wk of splinting. In closed injuries, localized swelling without the classic deformity are characteristi- cally seen early. Diagnosis is best made by splinting the finger in extension for a few days and reexamining the finger after the swelling settles down. Weak or absent exten- sion of the PIPJ suggests central slip disruption (118). Initial treatment of closed injury should be splinting the PIPJ in extension for 4–6 wk and reapplication of a splint if the deformity recurs. Surgical indications for a closed Boutonniere deformity are: displaced avulsion frac- ture at the base of the middle phalanx; axial and lateral instability of the PIPJ associ- ated with loss of active or passive extension of the joint (119); and failed nonoperative treatment. 64 Stephens et al. Surgical treatment consists of securing the central slip to the middle phalanx with or without the bony fragment. This may be helped by the use of a bony anchor in a non- fracture avulsion injury. Splinting the PIPJ in extension is then required. If a bony fracture is involved, mobilization is usually commenced after radiographic evidence of union exists. If primary repair of the central slip is not possible, both lateral bands can be longitudinally split and sutured together along the dorsal midline, thus recreating the central slip. Another option is to create a turnover flap from the proximal portion of the central slip, which is sutured to the distal end of the central slip. The proximal defect is then closed primarily. These methods help prevent the development of a Bou- tonniere deformity and will allow active flexion of the PIPJ. Zone IV Injury (Proximal Phalanx) These injuries are more often partial and tend to only involve the broad extensor retinaculum, not the lateral bands. For such partial injuries, if no loss of extension is present, then repair is often not required, and early motion should be considered. Alter- natively, a simple running or figure-of-eight suture using a nonresorbable 5-0 suture to coapt the edges may be used. For complete lacerations, primary repair should be per- formed followed by 6 wk of splinting in extension. Zone V Injury (MCPJ) An important aspect of injuries in this zone is that they are often owing to human bites, and the potential complications can be significant. When associated with a human bite, the incidence of complications is directly related to the time from injury to treatment. Surgical exploration, thorough irrigation, and primary repair are indi- cated. Arthrotomy should be considered if there are any concerns regarding MCPJ involvement. All involved structures, including partial injuries, should be repaired. Lateral bands should be repaired to prevent lateral migration of the extensor digi- torum communis tendon and subsequent loss of metacarpophalangeal extension (120,121). Repairs are satisfactorily carried out using figure-of-eight sutures with a nonresorbable 4-0 or 5-0 suture. For definite bite-related injuries, broad-spectrum antibiotics are always required, and the wound is usually left open to heal by secondary intention or closed several days if no clinical infection has developed. Zone VI Injury (Dorsal Hand) As the tendons are thicker and more oval shaped, repair should be performed using a four-strand core suture with a nonresorbable 4-0 suture similar to that used in flexor tendons. A peripheral suture is not required. Importantly, single or partial tendon lac- erations in this zone may not result in a loss of extension at the MCPJ, because extensor forces are transmitted from adjacent extensor tendons through the juncturae tendinum. Hence, clinical examination cannot be relied on to assess tendon integrity, and all wounds must be formally inspected with appropriate anesthetic support. Zone VII Injury (Wrist) Exposure in this region is limited because of the coverage by the extensor retinacu- lum. Therefore, partial division or venting of the retinaculum is required to enable a formal tendon repair, which should consist of a four-strand core suture as per zone VI Tendons of the Hand 65 injuries. Efforts should be made to preserve some segment of the retinaculum to pre- vent tendon bowstringing. Zone VIII Injury (Dorsal Forearm) Repair of this zone can be difficult with complications in identifying individual ten- dons with multiple tendon injuries and fibrous septa retracting into the muscle bellies when the division occurs at the musculotendinous junction. Significant consideration should be given to restoring wrist and thumb extension. For muscle bellies, multiple figure-of-eight sutures are required. Splinting of the wrist in 45° extension, the meta- carpophalangeal joint in 15–20° flexion, and, occasionally, the elbow at 90° of flexion is required. TENDON HEALING IN THE HAND Microanatomy and Nutrition of the Tendon in the Hand Tendons are remarkably organized structures adapted to transmit force generated by muscle. Ancient surgeons commented on the tendon injuries and proposed various treat- ments, including tendon repair. Due to Galen’s work and his profound influence on the practice of surgery, tendon injuries were allowed to heal without intervention—this method of treatment continued until the 19th century. Although early observations on tendon morphology date back to the 18th and 19th centuries, it was not until the 20th century that knowledge was truly advanced (122). The complexity of tendon structure can be appreciated on histological sectioning. The major constituent of tendons is type I collagen (>80% of dry weight). Three chains of collagen are coiled in a right-handed triple helix that are held together by covalent bonds, forming a collagen molecule. The molecule is approx 1.5 nm in diameter and 300 nm long. By arranging collagen molecules in a quarter-stagger, oppositely charged amino acids are aligned. Collagen molecules combine to form very strong structures: microfibrils. Microfibrils combine to form subfibrils, which then form fibrils. Fibrils are immersed in a matrix that is rich in proteoglycans, glycoproteins, and water. Tightly packed fibrils become fascicles (Fig. 1). The living components of the tendon—collagen-producing fibroblasts—reside between fibrils. The longitudinal histological section demonstrates cell bodies, which appear char- acteristically spindle-shaped and are orientated in rows between collagen bundles (Fig. 2). Fibroblasts stain darkly when using basic histological stains, such as hema- toxylin and eosin. Fascicles within the tendon are bound by loose connective tissue: the endotenon. The endotenon contains fibroblasts, microvessels, lymphatics, and nerve end- ings. The cross-section of the intrasynovial flexor tendon also reveals the existence of a loose connective tissue layer surrounding the tendon, which constitutes the visceral syn- ovial membrane or epitenon. Tendons that are not supported by the synovial sheath have a layer of loose connective tissue surrounding them: the paratenon. Early observations regarding the vascular pattern of tendons were linked to those of tendon healing and the formation of adhesions. However, the German and French anato- mists presented more systematic investigative work in the mid-19th century. Follow- ing his cadaveric studies, Berkenbush in 1887 (122) first described the vascular pattern of the human flexor tendon. He concluded that the blood supply to the sheathed tendon comes from the perimysium, periosteum, and the surrounding tissue via vessels in the 66 Stephens et al. sheath. Berkenbush was probably the first investigator to establish that both super- ficialis and profundus tendons had consistent areas of avascularity. He also postulated that the tendon was poorly nourished owing to this pattern of vascular supply. Research by Arai in the early 20th century confirmed Berkenbush’s findings and suggested the importance of diffusion in the process of tendon nutrition. In the 1960s and 1970s several investigators revisited and further examined the theory of nutrition by diffusion. In 1963 Potenza (123) and later Lundborg et al. (124) confirmed that the avascular segments in the tendons are predominantly nourished by diffusion, whereas the vascular parts derive their nutrition from vascular perfusion (122). Manske and coauthors (90,125) further defined and compared the effectiveness of both processes in several animal species. Their work published in the 1980s concluded Fig. 2. Longitudinal section of a tendon. Fig. 1. Tendon structure. Tendons of the Hand 67 that although both diffusion and perfusion were important nutrient pathways to the flexor and extensor tendons, diffusion was more effective than perfusion and could, in the absence of perfusion, support most of the tendon (19,90,126). Others, including Lundborg et al. and Hooper et al. (124,127–129), further quantified the relative effi- ciency of both processes using radioactive tracers. Tendon Healing The healing process of the injured tendon has been well documented (130). After the tendon has been incised and sutured, the healing process is initiated. It consists of three stages: the inflammatory stage, collagen production, and scar maturation and remodel- ling. Following surgical repair, the incision site fills with a blood clot containing the inflammatory cells, their mediators, and fibrin. Within the next few days, fibroblasts arrive into the area. Collagen production can be detected on the third postoperative day. Initially, the collagen products can be detected in the cytoplasm, but within days, the collagen fibrils are visible under the microscope. The angiogenic activity in the tendon stumps is markedly increased by d 7 (131). After 2 wk, the tendon stumps appear fused by the fibrous bridge that is formed by the migrating collagen type I fibers. At this stage collagen fibers are positioned perpendicularly to tendon fibrils. Within the next 2 wk, the remodeling process ensures a progressive parallel organiza- tion of collagen fibers. Interestingly, only the collagen fibers within and around the tendon organize in this parallel manner. The fibroblasts in the repair zone appear to be of both intrinsic and extrinsic origin. Further remodeling coincides with increased strength and reduced mass of scar tissue and continues only in the presence of longitudinal stressing. The remodeling process continues for approx 4 mo. In the 1960s and 1970s, many researchers made significant contributions to the cur- rent understanding of the healing cascade. Peacock (132–137) focused predominantly on the cellular aspects of the healing and recognized the three phases of the healing cascade as previously described. Potenza (123,138,139) examined the importance of proliferation and migration of the fibroblasts into the repair site. Both Peacock and Potenza believed that the formation of fibrous attachments or adhesions was signifi- cant in the delivery of cells and nutritional support to the repair zone. Later, cells exter- nal to the tendon were identified as also having a role in the repair process, the extrinsic repair (140–145). Lindsay identified the inflammatory nature of adhesions and disputed their impor- tance for the healing process. For the first time, his work demonstrated the existence of the intrinsic and extrinsic processes of the cellular repair in chickens (146–148). Lundborg and associates (124,149–151) then examined the intrinsic contribution to the healing cascade. They investigated the cellular behavior of cut rabbit tendons by placing the excised segment in remote in vivo locations like the synovial cavity or subcutaneously in the dialysing pouch. The results of these studies, which confirmed the intrinsic capacity of the tendons to heal in the absence of adhesions, were later disputed, as the environment in which tendons were placed was found to influence the healing. However, it was not long before the in vitro studies of Becker and Graham (152,153) once again demonstrated that the injured chicken tendons were capable of [...]... 40:587–601 37 Verdan CE Primary repair of flexor tendons J Bone Joint Surg Am 1960;42:647–657 Tendons of the Hand 79 38 Kleinert HE, Kutz JE, Ashbell TS Primary repair of lacerated flexor tendons in no man’s land (abstract) J Bone Joint Surg Am 1967;49:577 39 Kleinert HE, et al Primary repair of flexor tendons Orthop Clin North Am 19 73; 4: 865–876 40 Ejeskar A Flexor tendon repair in no-man’s-land: results of. .. 1985;1: 43 53 71 Shaieb MD, Singer DI Tensile strengths of various suture techniques J Hand Surg [Br] 1997;22:764–767 72 McLarney E, Hoffman H, Wolfe SW Biomechanical analysis of the cruciate four-strand flexor tendon repair J Hand Surg [Am] 1999;24:295 30 1 73 Dinopoulos HT, et al The resistance of a four- and eight-strand suture technique to gap formation during tensile testing: an experimental study of repaired... Peacock Jr EE Inter- and intramolecular bonding in collagen of healing wounds by insertion of methylene and amide cross-links into scar tissue: tensile strength and thermal shrinkage in rats Ann Surg 1966;1 63: 1–9 133 Peacock Jr EE Biology of tendon repair N Engl J Med 1967;276:680–6 83 134 Peacock Jr EE Dynamic aspects of collagen biology I Synthesis and assembly J Surg Res 1967;7: 433 –445 135 Peacock Jr... anatomy of the extensor tendons of the digits Hand Clin 1997; 13: 51–62 31 Rockwell WB, Butler PN, Byrne BA Extensor tendon: anatomy, injury, and reconstruction Plast Reconstr Surg 2000;106:1592–16 03 32 Doyle JR, Blythe WF Anatomy of the flexor tendon sheath and pulleys of the thumb J Hand Surg 1977;2:149–151 33 Bayat A, et al The pulley system of the thumb: anatomic and biomechanical study J Hand Surg... new method of suture Hand 1977;9 :30 6 31 1 58 Seradge H Elongation of the repair configuration following flexor tendon repair J Hand Surg [Am] 19 83; 8:182–185 59 Ejeskar A, Irstam L Elongation in profundus tendon repair A clinical and radiological study Scand J Plast Reconstr Surg 1981;15:61–68 60 Gelberman RH, et al The effect of gap formation at the repair site on the strength and excursion of intrasynovial... Slip-knot flexor tendon suture in Zone II allowing immediate mobilization Hand 19 83; 15 :35 2 35 8 176 Wada A, et al Comparison of postoperative early active mobilization and immobilization in vivo utilising a four-strand flexor tendon repair J Hand Surg [Br] 2001;26: 30 1 30 6 177 Savage R The influence of wrist position on the minimum force required for active movement of the interphalangeal joints J Hand... Orthop 2001 ;38 3:84–96 21 Sbernardori MC, et al Histogenesis and morphology of the flexor tendon pulley system in the human embryonic hand J Hand Surg [Br] 2000;25:175–179 22 Doyle JR Anatomy of the finger flexor tendon sheath and pulley system J Hand Surg [Am] 1988; 13: 4 73 484 23 Doyle JR Anatomy of the flexor tendon sheath and pulley system: a current review J Hand Surg [Am] 1989;14 :34 9 35 1 24 Kleinert... Report of the Committee on Tendon Injuries (International Federation of Societies for Surgery of the Hand) J Hand Surg [Am] 19 83; 8:794–798 25 von Schroeder HP, Botte MJ Anatomy and functional significance of the long extensors to the fingers and thumb Clin Orthop 2001 ;38 3:74– 83 26 Schenck R Variations of the extensor tendons of the fingers: Surgical significance J Bone Joint Surg Am 1964;46:1 03 110... Surg Res 19 83; 34: 83 93 91 Bowers WH, Kuzma GR, Bynum DK Closed traumatic rupture of finger flexor pulleys J Hand Surg [Am] 1994;19:782–787 92 Doyle JR, Blythe W Macroscopic and functional anatomy of the flexor sheath Proceedings of the 29th Annual American Society for Surgery of the Hand J Bone Joint Surg Am 1974;56:1094 93 Kwai Ben I, Elliot D “Venting” or partial lateral release of the A2 and A4 pulleys... pulleys after repair of zone II flexor tendon injuries J Hand Surg [Br] 1998; 23: 649–654 94 Mitsionis G, et al Feasibility of partial A2 and A4 pulley excision: residual pulley strength J Hand Surg [Br] 2000;25:90–94 95 Mitsionis G, et al Feasibility of partial A2 and A4 pulley excision: effect on finger flexor tendon biomechanics J Hand Surg [Am] 1999;24 :31 0 31 4 96 Tomaino M, et al The effect of partial . strength of a four-strand repair with the simplicity and tendon han- dling requirements of only a two-stranded repair. Suture material and caliber are also important aspects of tendon repair. . four-stranded repair being twice as strong as a two-stranded repair and, consequently, a six-stranded repair being almost three times stronger. Several authors have even described eight- stranded. without the bulk or tissue handling requirements of six- or eight- stranded repairs (48,67,68,71). Kubota et al. (67) described a four-strand single-knot modified Kessler repair, which provides four