- Email: [email protected]
Early active motion flexor tendon protocol using One splint
Early Active Motion Flexor Tendon Protocol Using One Splint Linda Klein, OTR, CHT Hand Surgery, Ltd. Milwaukee, Wisconsin
For decades, surgeons and therapists have been attempting to attain better results after primary flexor tendon repair. The most recent advances in surgical technique have centered on stronger repairs that tolerate active motion to ensure proximal gliding of the repaired flexor tendon. Although research is ongoing, it has been determined that the strength of a flexor tendon repair is related roughly to the number of suture strands crossing the repair site.1–5 Traditional repairs, such as the Kessler,6 consist of two strands of suture material crossing the repair. Fourstrand repairs are roughly twice as strong, better able to tolerate controlled active flexion immediately after surgery.4,5,7–11 Six-strand and eight-strand repairs require even more force to break; however, they have disadvantages of increased tendon trauma during repair and increased bulk of repair.1 The addition of a peripheral circumferential suture has been shown to add 10% to 50% to the strength of the core suture.3,12,13 After a four-strand repair of a lacerated flexor tendon, early active motion is becoming an accepted standard of treatment. The process of developing an early active postoperative therapy and splinting protocol at our facility was enhanced by consideration of current protocols in the literature. Many studies have shown good results.14–20 The splints used in each protocol vary widely, including materiThis article is adapted from a clinical study presented at the ASHT 2000 Conference in Seattle, Washington. Correspondence and reprint requests to Linda Klein, OTR, CHT, Hand Surgery, Ltd., 2500 North Mayfair Road, Suite 570, Milwaukee, WI 53226; e-mail:.
ABSTRACT: This article describes an early active motion protocol for use after a four-strand flexor tendon repair. The protocol uses a simple dorsal blocking splint with the wrist in neutral and four fingers in rubber band traction for the first five weeks, then gradually advances the patient over the next seven weeks. The patient is able to perform the exercises without changing the splint at home during the first five weeks of the protocol. The results of the retrospective chart review are promising. Of 40 digits, 95% experienced excellent and good results in zone II, and 87.5% experienced excellent and good results in zones I, II, and III. One rupture (2.5%) occurred in a noncompliant patient. The DASH scale was used to determine functional outcome, with results of 7.82 on the physical function/symptoms category, 16.07 in sports/ performing arts, and 10.23 in the work category. J HAND THER. 2003;16:199–206.
als, position of the wrist, and whether rubber-band traction and wrist tenodesis exercises are used. Wrist position has been studied as to the effects on grip strength and the amount of tension developed within the flexor tendons with the wrist in various positions.21,22 Savage,22 in studying four positions of the wrist and metacarpophalangeal (MCP) joints, concluded that of the four positions studied, 458 wrist extension with 908 MCP flexion was associated with the least flexion force of the flexor tendons. Wrist neutral was the position that developed the next least amount of flexion force. Rubber-band traction has been used by some clinicians to increase passive flexion of the joints, prevent inadvertent active motion of the repaired flexor tendon, reduce tension on the suture line, and, when applied to all four fingers, increase flexor tendon excursion and the power applied during proximal interphalangeal (PIP) extension by the intrinsics.23–27 Rubber-band traction has been linked, however, to a larger tendency for flexion contracture of the interphalangeal (IP) joints,27–29 and other clinicians prefer to eliminate the rubber-band traction from their postoperative protocols. May et al.27 modified their protocol to minimize flexion contractures by shortening the cast to end at the PIP joint and including all four fingers in rubber-band traction to facilitate a stronger IP extension contraction against the rubber bands. Studies of wrist tenodesis motion early in the postoperative regimen have shown that wrist synergistic motion provides a larger tendon excursion for the flexor tendons in zones II and III compared with July–September 2003 199
the Kleinert or modified Kleinert splints.30,31 Partial wrist extension reduces the antagonistic muscletendon tension of the extensors during active flexion.17,18 Of the studies done on early active flexor tendon protocols, the Indiana protocol5,14 uses a position of partial wrist extension (308) during the active hold portion of the exercises, consistent with minimizing tension in the repaired flexor according to the abovenoted studies. This protocol necessitates the use of two splints: a traditional dorsal blocking splint to protect the repaired flexor tendon at rest and a wrist hinge splint that blocks wrist extension at 308 when applying the place/active flexion hold component of the exercises. This requires that the patient be trusted to change the splints independently at home. The Indiana protocol incorporates controlled passive flexion rather than rubber-band traction, which may minimize IP flexion contractures.32 It takes advantage of the wrist tenodesis exercise by use of the wrist hinge during exercise. Silfverskiold and May15,16 developed a protocol that immobilizes the wrist and MCPs in a cast, with the wrist in neutral, that ends at the level of the PIP joint, with four fingers in rubber-band traction. A night IP extension addition to the cast assists in minimizing IP flexion contractures. Use of the cast for four weeks does not allow use of wrist tenodesis, and the bulk of the cast may limit the end range of flexion, although compliance for positioning is expected to be better with use of a cast. This study showed superior results (96% excellent and good) but included only simple tendon lacerations. Additionally the patients were hospitalized for five days after surgery, receiving more frequent and close supervision of their exercises than patients seen on an outpatient basis. Evans17,18 designed a protocol to be used after a two-strand flexor tendon repair that uses the traditional flexor tendon splint, with the wrist and MCPs flexed and rubber-band traction to all four fingers. The splint is removed in therapy for exercises, including wrist tenodesis and place/active hold exercises with the wrist in 208 extension. At home, the patients are allowed to perform passive flexion and active IP extension only, as in the traditional flexor tendon protocols, due to the weaker repair. By limiting the active hold portion of the exercises to the clinic, less frequent active tendon gliding is obtained. This protocol shows significant improvement in end results compared with passive flexion only for twostrand repairs. Additional early active motion protocols after a two-strand repair technique include those reported by Gratton19 and Elliott,20 in which a small amount of PIP and distal interphalangeal (DIP) active motion were allowed initially, increasing the amount of active flexion weekly. The position of the wrist in these studies is 208 to 308 flexion and, although not consistent with minimizing tension in the repaired 200
JOURNAL OF HAND THERAPY
flexor tendon during active use, was chosen to decrease the tendency to use the fingers actively when not exercising. With these studies in mind, the goal of developing this protocol was to design a splint that would not be changed by the patient at home, would allow immediate active controlled motion, yet protect the gliding tendon from rupture in the early and the later stages of tendon healing. Minimizing rupture potential was the primary concern when initiating use of an active motion flexor tendon protocol. To prevent the patient from changing the splint at home and to allow the patient to use the same splint for protection at rest and during the active phase of the exercises, the position of wrist neutral was chosen. Silfverskiold and May’s study,15,16 using a cast with the wrist in a neutral position, did not show an increased rupture rate. This is not the ideal position for minimizing flexor tendon tension during an active flexion contraction according to Savage’s study22; however, the risk of changing splints at home with possible inadvertent motion or use of the hand is avoided. The wrist tenodesis exercise was included, when in therapy, to maximize tendon gliding, while minimizing tendon force. Use of four-finger rubberband traction also was chosen to minimize risk of rupture by inadvertent use of the fingers between exercises, maximize passive IP flexion, and facilitate flexor tendon gliding by inclusion of all fingers in the traction.27 Emphasis is placed on achieving full IP extension during exercises by passively flexing the MCPs fully and using a padded aluminum wedge or the patient’s contralateral hand as a fulcrum behind the PIP joint to increase PIP and DIP extension potential.18,19 At night, the rubber-band traction is removed, and the fingers are strapped into extension to the hood of the dorsal blocking splint. This protocol varies from others in its extended protection of the well gliding tendon. Much has been learned regarding tendon healing since the 1970s33–36; however, the mechanism of tendon healing and development of tensile strength are still being explored. Studies have shown that flexor tendons allowed early motion have increased tensile strength compared with immobilized tendons.37,38 A study by Goldfarb et al.39 showed, however, that increasing the force of early passive motion after a flexor tendon repair does not increase the amount or maturation level of collagen at the repair site. Goldfarb et al.39 concluded that increasing the force application to a healing tendon may not accelerate the biologic or biomechanical aspects of healing in the first six weeks after intrasynovial tendon repair. In our experience, most patients allowed early motion after flexor tendon repair experience excellent to good motion and have an increased tendency to use their hand functionally, even when instructed to avoid resistance. Without the added support of adhesions, the
patients may be at more risk of rupture during resisted functional use than patients experiencing limitations due to adhesions. For this reason, a well gliding flexor tendon (patient experiencing a good or excellent result) is provided a more structured, gradually decreasing level of splinting until 12 weeks postoperatively within this protocol. When the fulllength dorsal blocking splint is discontinued, a smaller, hand-based dorsal blocking splint is applied. It is removed for exercise, bathing, rest periods, or light use. Longer protection of a flexor tendon that is gliding well after the early stage of flexor tendon healing has been supported by Silfverskiold (personal communication), who stated that the main goal during the first 12 weeks is to lay the basis for an excellent result. When an excellent result is obtained, the repair is relatively weak compared with a repair splinted by adhesion to surrounding tissue. The patient treated with an early active motion program has to be slowed down between 4 and 12 weeks.
METHODS A retrospective chart review was performed of all four-strand flexor tendon repairs performed by five hand surgeons and treated by five hand therapists in a private practice between 1997 and 1999. The surgery technique varied slightly with each surgeon, consisting of two Tajima40 or modified Kessler sutures6 or a modified Kessler suture and mattress or locking mattress suture using 3–0 or 4–0 braided synthetic material. All added a simple running epitenon repair of 6–0 polypropylene. A total of 40 digits in 35 patients were included. The flexor digitorum profundus tendon was repaired in all digits. Fifteen digits required additional flexor digitorum superficialis repairs, and 17 digits required one or both digital nerve repairs. Of the 40 repairs, 9 were of complex injuries. These included volar plate repairs at the DIP (four digits) or PIP (one digit), collateral ligament repair (one digit), digital artery repair (one digit), one rotation flap for soft tissue coverage, and one case in which the flexor digitorum profundus was cut and repaired in two places along its length. One additional case developed reflex sympathetic dystrophy. All patients who completed the 12-week protocol were included for results of the program. Results were calculated using the Strickland/Glogovac formula41: active PIP þ DIP flexion extensor loss at PIP and DIP 3 100 1758 ¼ % of normal active IP motion
The results are classified as excellent (85% to 100% of normal), good (70% to 84% of normal), fair (50% to 69% of normal), or poor (\50% of normal IP motion).
Criteria for inclusion into use of this protocol were the presence of a four-strand repair, patient ability to demonstrate the exercises as instructed, verbalization of understanding of the precautions and risks, and physician direction. Patients were not excluded from the protocol specifically for stiffness or swelling. Stiffness and swelling increase the work of flexion42 when attempting to flex into the restricted range. If stiffness or swelling limited passive motion at the initial visit, the place–active hold position was performed in the position the joints could attain with light placement by the therapist (not into the restricted range). Passive limitations were addressed as part of the protocol, and the place–active hold position of flexion gradually increased as passive motion improved. The average time postoperatively that the patient was initiated in therapy was 3.5 days. The average number of visits over the 12-week period was 16 (range 8 to 28 visits). Exclusions from the study included nine patients who were lost to follow-up and did not complete the 12-week protocol, seven patients who were seen at other facilities due to insurance restraints, and six patients who were set up with modified protocols because of age (4 to 13 years) or severity of injury (associated fracture and adjacent amputation, severe crush), as requested by the surgeon. The youngest patient to which this protocol was applied without modification during this period was 16 years old.
Protocol Description A dorsal blocking splint is fabricated using thermoplastic material, with the wrist in neutral, MCP’s 508 to 708 flexion, and IPs allowed full extension. In the presence of a digital nerve repair, the IPs were blocked at 158 flexion or as specified by the surgeon. The hood of the dorsal blocking splint extends to the fingertips, allowing the IPs to be strapped loosely in extension at night. Rubber-band traction is applied to all four fingers, run under a palmar pulley (safety pin or other attachment at distal palmar crease), and attached at the proximal forearm strap (Figure 1). Exercises are performed hourly for 10 repetitions. The patient is instructed in initial edema control consisting of elevation and neck, shoulder, and elbow motion, with light wound bandaging. If needed, a self-adhesive circumferential wrap to the injured digit to decrease swelling is applied at night, if the patient or significant other is able to apply it safely. It is not used during the day because it may add resistance to the active hold exercises in flexion. Instruction is given for passive flexion of all joints of the fingers and full active IP extension to the dorsal hood of the blocking splint. Maximal passive MCP flexion using the contralateral hand, padded July–September 2003 201
FIGURE 1. Dorsal blocking splint with wrist neutral, MCPs 508 to 708 flexion, and rubber-band traction to all four fingers for the first five weeks. At night, the rubber-band traction is disconnected, and the fingers are strapped gently to the dorsal hood.
FIGURE 2. Use of padded aluminum placed behind the finger
resistance to the repaired flexor tendon before performing the gentle place–active hold in flexion. In therapy, the splint is removed for cleansing, wrist tenodesis exercises, passive flexion, active IP extension, and place–active hold with the wrist in 208 to 308 extension (Figure 5). Patients are seen in therapy one to three times per week, depending on their level of swelling, pain, passive range of motion, ability to hold in flexion with minimal effort, and ability to perform the exercises independently. At five weeks postoperatively, the rubber-band traction is discontinued. The dorsal blocking splint is continued, with fingers strapped to the dorsal hood between exercises. The patient is instructed to remove the splint at home for the exercises already described with the addition of active wrist extension and flexion with fingers relaxed (wrist tenodesis) and active finger flexion with the wrist at 308 extension. The hand is assessed for intrinsic tightness or MCP volar tightness, with intrinsic stretch performed pas-
during active IP extension to facilitate full extension.
FIGURE 3. Passive flexion is performed before the place–active
aluminum splint, or other method is instructed while performing the active IP extension against the resistance of the rubber bands, to minimize PIP flexion contractures (Figure 2). Rubber bands are allowed to be disconnected proximally during the active IP extension to attain the full amount of extension allowed when needed. Place–active hold in flexion is then performed, with passive flexion of all digits by the contralateral hand (Figure 3) followed by gentle active hold of the fingers in the flexed position for 2 to 3 seconds when released by the contralateral hand (Figure 4). No pressure of the fingertips on the palm or palmar strap is allowed. The rubber-band traction is detached from the forearm attachment for the place–active hold exercises. If passive flexion is limited by swelling, stiffness, or pain, the patient is instructed to perform stronger passive flexion to tolerance, using the contralateral hand, to decrease joint stiffness and
FIGURE 4. Active hold in flexion of all fingers by a patient four weeks post repair of the flexor digitorum profundus and radial digital nerve in the small finger.
202
JOURNAL OF HAND THERAPY
hold exercises. The fingers are placed gently into flexion by the patient’s other hand.
FIGURE 5. In therapy, the splint is removed to allow wrist tenodesis motion to 208 to 308 extension during the place–active hold finger flexion exercises.
sively in therapy only, with the wrist flexed to protect the repaired tendon, if needed. At six weeks postoperatively, these exercises are continued, with the addition of composite extension of all fingers with the wrist neutral. At eight weeks postoperatively, passive IP extension exercises are added to the home program if flexion contractures are present, as determined appropriate by surgeon and therapist. The dorsal blocking splint is cut to free the wrist. The hand-based dorsal blocking splint is worn when active or at work, to prevent inadvertent resistance to the flexors (Figure 6). At 12 weeks postoperatively, all protective splinting is discontinued. The hand is allowed to be used normally, avoiding strong tip resistance for another 2 weeks.
RESULTS Thirty-five subjects, with 40 flexor tendon repairs in zones I, II, and III, were evaluated with the
Strickland/Glogovac formula41 with active range of motion at the end of treatment, 12 to 14 weeks postoperatively. The results show 29% excellent, 43% good, 21% fair, and 1% poor results in zone I; 53% excellent, 42% good, 0% fair, and 0% poor results in zone II; and 100% excellent, 0% good, 0% fair, and 0% poor results in zone III (Table 1). The only rupture occurred in a patient performing an activity not allowed by the protocol at 4.5 weeks postoperatively (using the repaired index finger on a knife to scoop peanut butter out of a jar). The tendon was immediately repaired again, and the patient went on to an excellent result. Flexion and extension averages were calculated, to determine more specifically the motion recovery tendencies. The average flexion of the PIP and DIP joints for zone I was 1538; zone II, 1628; and zone III, 161.58. Average loss of IP extension was 208 in zone I, 118 in zone II, and 4.58 in zone III. Total active motion averages of the IP joints were 1338 for zone I, 150.58 for zone II, and 157.58 for zone III (Table 2). Results of zones I and II were calculated separately for comparison with other studies (Table 3). All studies reported their results using the Strickland/ Glogovac criteria,41 with the exception of the 1989 Belfast study, which used ASSH criteria.19 Silfverskiold and May,15,16 including repairs in zone II, showed 71% excellent, 25% good, 0% fair, and 0% poor results, with one rupture. No complex injuries were included. Strickland and Gettle,14 including repairs in zones I and II, showed 24% excellent, 46% good, 24% fair and 5% poor results, with three ruptures. Evans,17,18 including repairs in zone II applied to two-strand repairs, showed 43% excellent, 32% good, 14% fair, and 11% poor results, with one rupture. Complex injuries were included. Of the studies reported by Gratton19 (also applied after twostrand repairs), results of the Belfast study in 1989 showed 46% excellent, 31% good, 14% fair, and 9% poor results, with 11 ruptures; the Sheffield study in 1989 showed 71% excellent, 6.5% good, 3.5% fair, and 19% poor results, with 2 ruptures; and the Sheffield study in 1990 showed 49% excellent, 36% good, 11% fair, and 4% poor results, with 10 ruptures. This study showed 42% excellent, 42% good, 9% fair, and 3%
TABLE 1. Results of Flexor Tendon Repair by Zone of Injury Using the Strickland/Glogovac41 Formula for Grading Flexor Tendon Results* No. Zone Digits Excellent I II III
FIGURE 6. Hand-based dorsal blocking splint worn intermittently between weeks 8 and 12 during work or heavy activity to remind the patient to avoid strong resistance to the fingers.
14 19 7
Good
Fair
Poor Ruptures
4 (29%) 6 (43%) 3 (21%) 1 (7%) 0 10 (53%) 8 (42%) 0 0 1 (2.5%)y 7 (100%) 0 0 0 0
*See text for description of formula and classification system. yRupture calculated as 2.5% of all digits and 3% of digits in zones I and II. The rupture occurred in a noncompliant patient.
July–September 2003 203
TABLE 2. Range of Motion Averages at Interphalangeal Joints Zone I II III
Flexion Average (8)
Extension Loss Average (8)
PIP 100 DIP 53 PIP 98.5 DIP 63 PIP 102 DIP 60
PIP 10.5 DIP 9.5 PIP 8 DIP 3 PIP 4.5 DIP 0
Total Active IP Motion Average (8) 133 150.5 157.5
poor results in zones I and II. Zone II calculated alone showed 53% excellent, 42% good, 0% fair, and 0% poor results, with one rupture. Complex injuries were included. Functional results and limitations were assessed using the DASH (Disabilities of the Arm, Shoulder, and Hand) survey.43 The DASH survey was mailed to the 35 patients in the study, and 13 (37%) surveys were returned. In the physical function and symptoms category, the average DASH score was 7.82 on a scale of 0 to 100, with 0 being no disability and 100 being severe disability. In the sports/performing arts category, the average DASH score was 16.07. In the work category, the average DASH score was 10.23.
DISCUSSION The protocol described in this article was developed after consideration of protocols already designed with good results and has modified various components to fit the needs of our facility. After a four-strand repair, we wanted to attempt an active motion protocol with as many of the patients as possible, while minimizing the potential for rupture. The wrist neutral position was chosen based on Silfverskiold and May’s study15,16 to allow the patient to remain in the same splint for rest and exercise. The splint was designed as a thermoplastic, removable splint that extends the full length of the fingers. Rubber-band traction has been used by Dovelle and Chow et al.,23,24 Lister et al.,25 and May et al.26 Four-finger rubber-band traction, included to
prevent active use of the fingers between exercises, increase passive flexion, and maximize flexor digitorum profundus excursion and IP extension, has been used by Evans and Thompson,18 Silfverskiold and May,15,16 and Karlander et al.44 Inclusion of wrist tenodesis exercises is based on studies by Cooney et al.30 and Horii et al.31 The longer protection period with a hand-based splint was added by this group, after two ruptures in the earlier design of this protocol that occurred between 6.5 and 10 weeks postoperatively, at which time the splint was allowed to be removed completely at six weeks. The ruptures occurred in two patients with excellent results when they used their fingers with DIP resistance without prior thought. There have been no ruptures in patients compliant with the extended splint use in the protocol as described in this article. Patients experiencing limited motion due to adhesions may be advanced more quickly through the protocol. The problem of later ruptures and the need for a graduated level of protection through the full length of tendon healing are supported by other authors. Strickland and Gettle14 noted the occurrence of ruptures in ‘‘patients who were doing so well with their rehabilitation program that they elected to use their hands in an extremely strong manner (such as heavy lifting) well before the tendons had returned to sufficient strength to tolerate the extremely high tensile demands of such activity.’’ Silfverskiold (personal communication) suggested that in cases with a good or excellent result he would not recommend significant resisted exercises until 11 to 12 weeks postoperatively. When comparing results of studies, it is important to be aware of whether complex injuries were included within the results because one can expect the results of a study of simple tendon lacerations to be better than those that include crush injuries or injuries and repair to other structures. Gault28 discussed the tendency of flexor tendon repairs with volar plate injuries to develop flexion contractures. Elliot et al.20 and Gerbino et al.45 discussed difficulty in zone I in obtaining a gliding tendon repair under the A4 pulley. In the present study, zone I repairs
TABLE 3. Comparison with Other Early Active Motion Protocols Protocol
No. Digits
Excellent
Good
Fair
Poor
Ruptures
Silfverskiold/May* (zone II) Strickland/Cannon/Gettley (zone I,II) Evansz (zone II) Belfast 1989z (zone II) Sheffield 1989z (zone II) Sheffield 1990z (zone II) Klein (this study)y (zone I, II) (zone II alone)
55 40 44 117 31 47 33 19
39 (71%) 9 (24%) 19 (43%) 46% 71% 49% 14 (42%) 10 (53%)
14 (25%) 17 (46%) 14 (32%) 31% 6.5% 36% 14 (42%) 8 (42%)
0 9 (24%) 6 (14%) 14% 3.5% 11% 3 (9%) 0
0 2 (5%) 5 (11%) 9% 19% 4% 1 (3%) 0
2 3 1 11 2 10 1 1
*Two-strand plus cross-stitch repair. zTwo-strand repair. yFour-strand repair.
204
JOURNAL OF HAND THERAPY
showed a smaller number of excellent results compared with other zones; however, 35% of the injuries in zone I were complex, including volar plate or collateral ligament repairs. Most had good flexion but developed IP flexion contractures between 58 and 308, resulting in less overall motion. With this in mind, the author will consider modifying the use of rubber-band traction with this protocol with patients who have had a volar plate or collateral ligament injury in addition to flexor tendon injury. Earlier intervention with passive IP extension in therapy with the flexor tendon in a protected position with proximal joints held flexed also is recommended and now is being performed by the author within the protocol as soon as limited IP extension is noted. Intermittent positioning during the day, alternating between rubber-band traction and strapping the IPs in extension in the dorsal blocking splint, also may be helpful for patients showing early IP flexion contracture development. There are many limitations of this study. The retrospective format has inherent limitations, including the possibility of unrecognized factors affecting outcome, lower response rate on the DASH scale mail survey than if completed in the clinic, possible patient recall bias in completing the DASH survey, and potential reviewer bias. With this in mind, the charts were reviewed with strict application of the Strickland/Glogovac formula according to the final measurements in the chart for all patients completing the 12-week protocol. The amount of tension applied by the patient during the active hold in flexion was not measured, as in the Evans study.17,18 No ruptures were noted, however, during exercises or at rest using this protocol, with the only rupture occurring due to noncompliant use against resistance. Work of flexion, as described by Halikis et al,41 describes the sum of forces that resist flexion of the digit. Work of flexion is affected by the resistance offered to the flexor tendon by edema, joint stiffness, and adhesions. Early active motion protocols, such as this one, encounter the resistance of edema and joint stiffness when initially moving the digits due to trauma related to the injury and surgery. The study by Halikis et al41 showed that there is increased work of flexion developed when a tendon is mobilized immediately after surgery compared with a short delay of three to five days, with three days showing the development of the least amount of work of force. This protocol relied on education provided to the patient to apply a light active hold and the experience of the therapists in applying gentle passive flexion during the place–active hold, incorporating patient feedback as to how difficult it felt to hold the fingers in flexion. Practice with the contralateral hand, showing the light pull needed to hold the fingers gently in flexion and the way in which the wrist
position affects the amount of pull needed to achieve flexion (wrist tenodesis), is provided. Strong emphasis is placed on use of a light pull by the patient when applying the active hold component, with education on potential for tendon rupture with excessive internal tension by the patient. When passive motion was limited initially, emphasis was placed on the passive flexion exercises done before the place–active hold, to decrease joint stiffness gradually. Although not formally assessed, these methods are helpful in decreasing the force applied to a newly repaired flexor tendon during active flexion. There were no cases of rupture using this method in compliant patients. The results of this study reflect not only the protocol used, but experienced surgeons using atraumatic tendon handling techniques. Surgical technique was not controlled, and a variety of fourstrand techniques were used.
CONCLUSIONS The results of the retrospective chart review of this protocol for the management of four-strand flexor tendon repair show that it is safe (only one rupture), effective (87.5% good and excellent results in zones I, II, and III, including complex injuries, and in zone II alone, 95% good and excellent results, yielding high functional scores on the DASH survey), and simple.
REFERENCES 1. Savage R, Risitano G. Flexor tendon repair using a ‘‘six strand’’ method of repair and early mobilisation. J Hand Surg Br. 1989;14:396–9. 2. Komanduri M, et al. Tensile strength of flexor tendon repairs in a dynamic cadaver model. J Hand Surg Am. 1996;21:605–11. 3. Silfverskiold K, Andersson C. Two new methods of tendon repair: an in vitro evaluation of tensile strength and gap formation. J Hand Surg Am. 1993;18:58–65. 4. Shaieb MD, Singer DI. Tensile strengths of various suture techniques. J Hand Surg Br. 1997;22:764–7. 5. Strickland JM, Cannon NM. Flexor tendon repair—Indiana method. Indiana Hand Center Newsletter. 1993;1:1–19. 6. Kessler I. The ‘‘grasping’’ technique for tendon repair. Hand. 1973;5:253–5. 7. Trail IA, Powell ES, Noble J. The mechanical strength of various suture techniques. J Hand Surg Br. 1992;17:89–91. 8. Robertson GA, Al-Quattan MM. A biomechanical analysis of a new interlock suture technique for flexor tendon repair. J Hand Surg Br. 1992;17:92–3. 9. 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 Am. 1990; 15:953–8. 10. McLarney E, Hoffman H, Wolfe S. Biomechanical analysis of the cruciate four-strand flexor tendon repair. J Hand Surg Am. 1999;24:295–301. 11. Savage R. In vitro studies of a new method of flexor tendon repair. J Hand Surg Br. 1985;10:135–41. 12. Diao E, et al. Effect of periperal suture depth on strength of tendon repairs. J Hand Surg Am. 1996;21:234–9. 13. Wade PJF, et al. Flexor tendon repair: significant gain in strength from the Halsted peripheral suture technique. J Hand Surg Br. 1989;14:232–5.
July–September 2003 205
14. Strickland JM, Gettle KH. Flexor tendon repair: the Indianapolis method. In: Hunter JM, Schneider LH, Mackin EJ (eds). Tendon and Nerve Surgery in the Hand: A Third Decade. St. Louis, Mo.: Mosby, 1997:353–61. 15. Silfverskiold KL, May EJ. Flexor tendon repair in zone II with a new suture technique and an early mobilization program combining passive and active flexion. J Hand Surg Am. 1994;19:53–60. 16. Silfverskiold KL, May EJ. Flexor tendon repair with active mobilization: the Gothenburg experience. In: Hunter JM, Schneider LH, Mackin EJ (eds). Tendon and Nerve Surgery in the Hand: A Third Decade. St. Louis, Mo.: Mosby, 1997: 342–52. 17. Evans RB. The application of force to the healing tendon. J Hand Ther. 1993;6:266–84. 18. Evans RB, Thompson DE. Immediate active short arc motion following tendon repair. In: Hunter JM, Schneider LH, Mackin EJ (eds). Tendon and Nerve Surgery in the Hand: A Third Decade. St. Louis, Mo.: Mosby, 1997:362–93. 19. Gratton P. Early active mobilization after flexor tendon repairs. J Hand Ther. 1993;6:285–9. 20. Elliott D, et al. The rupture rate of acute flexor tendon repairs mobilized by the controlled active motion regimen. J Hand Surg Br. 1994;19:607–12. 21. O’Driscoll SW, Horii E. The relationship between wrist position, grasp size, and grip strength. J Hand Surg Am. 1992; 17:169–77. 22. Savage R. The influence of wrist position on the minimum force required for active movement of the interphalangeal joint. J Hand Surg Br. 1988;13:252–68. 23. Dovelle S, Kulis Heeter P. The Washington regimen: rehabilitation of the hand following flexor tendon injuries. Phy Ther. 1989;69:42–8. 24. Chow JA, Linwood JT, Dovelle S, Monsivais J, Milnor WH, Jackson JP. Controlled motion rehabilitation after flexor tendon repair and grafting. J Bone Joint Surg Br. 1988;70:591–5. 25. Lister G, Kleinert H, Kutz J, Atasoy E. Primary flexor tendon repair followed by immediate controlled mobilization. J Hand Surg. 1977;2:441–51. 26. May EJ, Silfverskiold KL, Sollerman CJ. The correlation between controlled range of motion with dynamic traction and results after flexor tendon repair in zone II. J Hand Surg Am. 1992;17:1133–9. 27. May EJ, Silfverskiold KL, Sollerman CJ. Controlled mobilization after flexor tendon repair in zone. II: a prospective comparison of three methods. J Hand Surg Am. 1992;17:942–53. 28. Gault DT. A review of repaired flexor tendons. J Hand Surg Br. 1987;12:321–5. 29. Burge PD, Brown NB. Elastic band mobilisation after flexor tendon repair; splint design and risk of flexion contracture. J Hand Surg Br. 1990;15:443–8.
206
JOURNAL OF HAND THERAPY
30. Cooney WP, Lin GT, An KN. Improved tendon excursion following flexor tendon repair. J Hand Ther. 1989;2:102–13. 31. Horii E, et al. Comparative flexor tendon excursion after passive mobilization: an in vitro study. J Hand Surg Am. 1992;17:559–66. 32. Strickland JM. Development of flexor tendon surgery: twentyfive years of progress. J Hand Surg Am. 2000;25:214–35. 33. Becker H, et al. Intrinsic tendon cell proliferation in tissue culture. J Hand Surg. 1981;6:616–9. 34. Lundborg G, Rank F. Experimental intrinsic healing of flexor tendons based upon synovial fluid nutrition. J Hand Surg. 1978;3:21–31. 35. Manske PR, Lesker PA. Biochemical evidence of flexor tendon participation in the repair process—an in vitro study. J Hand Surg Br. 1984;9:117–20. 36. Gelberman RH, et al. The early stages of flexor tendon healing: a morphologic study of the first fourteen days. J Hand Surg Am. 1985;10:776–84. 37. Gelberman RH, Botte MJ, Spiegelman JJ, et al. The excursion and deformation of repaired flexor tendons treated with protected early motion. J Hand Surg Am. 1986;11:106–10. 38. Hitchcock TF, Light TR, Bunch WH, et al. The effect of immediate constrained digital motion on the strength of flexor tendon repairs in chickens. J Hand Surg Am. 1987;12:590–5. 39. Goldfarb CA, et al. The effect of variations in applied rehabilitation force on collagen concentration and maturation at the intrasynovial flexor tendon repair site. J Hand Surg Am. 2001;26:841–6. 40. Lister GD, et al. Primary flexor tendon repair followed by immediate controlled mobilization. J Hand Surg. 1977;2: 441–451. 41. Strickland JW, Glogovac SV. Digital function following flexor tendon repair in zone 2: a comparison study of immobilization and controlled passive motion. J Hand Surg Am. 1980; 5:537–43. 42. Halikis MN, Manske PR, Kubota H, Aoki M. Effect of immobilization, immediate mobilization, and delayed mobilization on the resistance to digital flexion using a tendon injury model. J Hand Surg Am. 1997;22:464–72. 43. McConnell S, Beaton DE, Bombardier C. The DASH Outcome User’s Manual. Toronto, Ontario: Institute for Work & Health, 1999. 44. Karlander LE, et al. Improved results in zone 2 flexor tendon injuries with a modified technique of immediate controlled mobilization. J Hand Surg Br. 1993;18:26–30. 45. Gerbino PG, Saldana MJ, Westerbeck P, Schacherer TG. Complications experienced in the rehabilitation of zone 1 flexor tendon injuries with dynamic traction splinting. J Hand Surg Am. 1991;16:680–6.
For decades, surgeons and therapists have been attempting to attain better results after primary flexor tendon repair. The most recent advances in surgical technique have centered on stronger repairs that tolerate active motion to ensure proximal gliding of the repaired flexor tendon. Although research is ongoing, it has been determined that the strength of a flexor tendon repair is related roughly to the number of suture strands crossing the repair site.1–5 Traditional repairs, such as the Kessler,6 consist of two strands of suture material crossing the repair. Fourstrand repairs are roughly twice as strong, better able to tolerate controlled active flexion immediately after surgery.4,5,7–11 Six-strand and eight-strand repairs require even more force to break; however, they have disadvantages of increased tendon trauma during repair and increased bulk of repair.1 The addition of a peripheral circumferential suture has been shown to add 10% to 50% to the strength of the core suture.3,12,13 After a four-strand repair of a lacerated flexor tendon, early active motion is becoming an accepted standard of treatment. The process of developing an early active postoperative therapy and splinting protocol at our facility was enhanced by consideration of current protocols in the literature. Many studies have shown good results.14–20 The splints used in each protocol vary widely, including materiThis article is adapted from a clinical study presented at the ASHT 2000 Conference in Seattle, Washington. Correspondence and reprint requests to Linda Klein, OTR, CHT, Hand Surgery, Ltd., 2500 North Mayfair Road, Suite 570, Milwaukee, WI 53226; e-mail:
ABSTRACT: This article describes an early active motion protocol for use after a four-strand flexor tendon repair. The protocol uses a simple dorsal blocking splint with the wrist in neutral and four fingers in rubber band traction for the first five weeks, then gradually advances the patient over the next seven weeks. The patient is able to perform the exercises without changing the splint at home during the first five weeks of the protocol. The results of the retrospective chart review are promising. Of 40 digits, 95% experienced excellent and good results in zone II, and 87.5% experienced excellent and good results in zones I, II, and III. One rupture (2.5%) occurred in a noncompliant patient. The DASH scale was used to determine functional outcome, with results of 7.82 on the physical function/symptoms category, 16.07 in sports/ performing arts, and 10.23 in the work category. J HAND THER. 2003;16:199–206.
als, position of the wrist, and whether rubber-band traction and wrist tenodesis exercises are used. Wrist position has been studied as to the effects on grip strength and the amount of tension developed within the flexor tendons with the wrist in various positions.21,22 Savage,22 in studying four positions of the wrist and metacarpophalangeal (MCP) joints, concluded that of the four positions studied, 458 wrist extension with 908 MCP flexion was associated with the least flexion force of the flexor tendons. Wrist neutral was the position that developed the next least amount of flexion force. Rubber-band traction has been used by some clinicians to increase passive flexion of the joints, prevent inadvertent active motion of the repaired flexor tendon, reduce tension on the suture line, and, when applied to all four fingers, increase flexor tendon excursion and the power applied during proximal interphalangeal (PIP) extension by the intrinsics.23–27 Rubber-band traction has been linked, however, to a larger tendency for flexion contracture of the interphalangeal (IP) joints,27–29 and other clinicians prefer to eliminate the rubber-band traction from their postoperative protocols. May et al.27 modified their protocol to minimize flexion contractures by shortening the cast to end at the PIP joint and including all four fingers in rubber-band traction to facilitate a stronger IP extension contraction against the rubber bands. Studies of wrist tenodesis motion early in the postoperative regimen have shown that wrist synergistic motion provides a larger tendon excursion for the flexor tendons in zones II and III compared with July–September 2003 199
the Kleinert or modified Kleinert splints.30,31 Partial wrist extension reduces the antagonistic muscletendon tension of the extensors during active flexion.17,18 Of the studies done on early active flexor tendon protocols, the Indiana protocol5,14 uses a position of partial wrist extension (308) during the active hold portion of the exercises, consistent with minimizing tension in the repaired flexor according to the abovenoted studies. This protocol necessitates the use of two splints: a traditional dorsal blocking splint to protect the repaired flexor tendon at rest and a wrist hinge splint that blocks wrist extension at 308 when applying the place/active flexion hold component of the exercises. This requires that the patient be trusted to change the splints independently at home. The Indiana protocol incorporates controlled passive flexion rather than rubber-band traction, which may minimize IP flexion contractures.32 It takes advantage of the wrist tenodesis exercise by use of the wrist hinge during exercise. Silfverskiold and May15,16 developed a protocol that immobilizes the wrist and MCPs in a cast, with the wrist in neutral, that ends at the level of the PIP joint, with four fingers in rubber-band traction. A night IP extension addition to the cast assists in minimizing IP flexion contractures. Use of the cast for four weeks does not allow use of wrist tenodesis, and the bulk of the cast may limit the end range of flexion, although compliance for positioning is expected to be better with use of a cast. This study showed superior results (96% excellent and good) but included only simple tendon lacerations. Additionally the patients were hospitalized for five days after surgery, receiving more frequent and close supervision of their exercises than patients seen on an outpatient basis. Evans17,18 designed a protocol to be used after a two-strand flexor tendon repair that uses the traditional flexor tendon splint, with the wrist and MCPs flexed and rubber-band traction to all four fingers. The splint is removed in therapy for exercises, including wrist tenodesis and place/active hold exercises with the wrist in 208 extension. At home, the patients are allowed to perform passive flexion and active IP extension only, as in the traditional flexor tendon protocols, due to the weaker repair. By limiting the active hold portion of the exercises to the clinic, less frequent active tendon gliding is obtained. This protocol shows significant improvement in end results compared with passive flexion only for twostrand repairs. Additional early active motion protocols after a two-strand repair technique include those reported by Gratton19 and Elliott,20 in which a small amount of PIP and distal interphalangeal (DIP) active motion were allowed initially, increasing the amount of active flexion weekly. The position of the wrist in these studies is 208 to 308 flexion and, although not consistent with minimizing tension in the repaired 200
JOURNAL OF HAND THERAPY
flexor tendon during active use, was chosen to decrease the tendency to use the fingers actively when not exercising. With these studies in mind, the goal of developing this protocol was to design a splint that would not be changed by the patient at home, would allow immediate active controlled motion, yet protect the gliding tendon from rupture in the early and the later stages of tendon healing. Minimizing rupture potential was the primary concern when initiating use of an active motion flexor tendon protocol. To prevent the patient from changing the splint at home and to allow the patient to use the same splint for protection at rest and during the active phase of the exercises, the position of wrist neutral was chosen. Silfverskiold and May’s study,15,16 using a cast with the wrist in a neutral position, did not show an increased rupture rate. This is not the ideal position for minimizing flexor tendon tension during an active flexion contraction according to Savage’s study22; however, the risk of changing splints at home with possible inadvertent motion or use of the hand is avoided. The wrist tenodesis exercise was included, when in therapy, to maximize tendon gliding, while minimizing tendon force. Use of four-finger rubberband traction also was chosen to minimize risk of rupture by inadvertent use of the fingers between exercises, maximize passive IP flexion, and facilitate flexor tendon gliding by inclusion of all fingers in the traction.27 Emphasis is placed on achieving full IP extension during exercises by passively flexing the MCPs fully and using a padded aluminum wedge or the patient’s contralateral hand as a fulcrum behind the PIP joint to increase PIP and DIP extension potential.18,19 At night, the rubber-band traction is removed, and the fingers are strapped into extension to the hood of the dorsal blocking splint. This protocol varies from others in its extended protection of the well gliding tendon. Much has been learned regarding tendon healing since the 1970s33–36; however, the mechanism of tendon healing and development of tensile strength are still being explored. Studies have shown that flexor tendons allowed early motion have increased tensile strength compared with immobilized tendons.37,38 A study by Goldfarb et al.39 showed, however, that increasing the force of early passive motion after a flexor tendon repair does not increase the amount or maturation level of collagen at the repair site. Goldfarb et al.39 concluded that increasing the force application to a healing tendon may not accelerate the biologic or biomechanical aspects of healing in the first six weeks after intrasynovial tendon repair. In our experience, most patients allowed early motion after flexor tendon repair experience excellent to good motion and have an increased tendency to use their hand functionally, even when instructed to avoid resistance. Without the added support of adhesions, the
patients may be at more risk of rupture during resisted functional use than patients experiencing limitations due to adhesions. For this reason, a well gliding flexor tendon (patient experiencing a good or excellent result) is provided a more structured, gradually decreasing level of splinting until 12 weeks postoperatively within this protocol. When the fulllength dorsal blocking splint is discontinued, a smaller, hand-based dorsal blocking splint is applied. It is removed for exercise, bathing, rest periods, or light use. Longer protection of a flexor tendon that is gliding well after the early stage of flexor tendon healing has been supported by Silfverskiold (personal communication), who stated that the main goal during the first 12 weeks is to lay the basis for an excellent result. When an excellent result is obtained, the repair is relatively weak compared with a repair splinted by adhesion to surrounding tissue. The patient treated with an early active motion program has to be slowed down between 4 and 12 weeks.
METHODS A retrospective chart review was performed of all four-strand flexor tendon repairs performed by five hand surgeons and treated by five hand therapists in a private practice between 1997 and 1999. The surgery technique varied slightly with each surgeon, consisting of two Tajima40 or modified Kessler sutures6 or a modified Kessler suture and mattress or locking mattress suture using 3–0 or 4–0 braided synthetic material. All added a simple running epitenon repair of 6–0 polypropylene. A total of 40 digits in 35 patients were included. The flexor digitorum profundus tendon was repaired in all digits. Fifteen digits required additional flexor digitorum superficialis repairs, and 17 digits required one or both digital nerve repairs. Of the 40 repairs, 9 were of complex injuries. These included volar plate repairs at the DIP (four digits) or PIP (one digit), collateral ligament repair (one digit), digital artery repair (one digit), one rotation flap for soft tissue coverage, and one case in which the flexor digitorum profundus was cut and repaired in two places along its length. One additional case developed reflex sympathetic dystrophy. All patients who completed the 12-week protocol were included for results of the program. Results were calculated using the Strickland/Glogovac formula41: active PIP þ DIP flexion extensor loss at PIP and DIP 3 100 1758 ¼ % of normal active IP motion
The results are classified as excellent (85% to 100% of normal), good (70% to 84% of normal), fair (50% to 69% of normal), or poor (\50% of normal IP motion).
Criteria for inclusion into use of this protocol were the presence of a four-strand repair, patient ability to demonstrate the exercises as instructed, verbalization of understanding of the precautions and risks, and physician direction. Patients were not excluded from the protocol specifically for stiffness or swelling. Stiffness and swelling increase the work of flexion42 when attempting to flex into the restricted range. If stiffness or swelling limited passive motion at the initial visit, the place–active hold position was performed in the position the joints could attain with light placement by the therapist (not into the restricted range). Passive limitations were addressed as part of the protocol, and the place–active hold position of flexion gradually increased as passive motion improved. The average time postoperatively that the patient was initiated in therapy was 3.5 days. The average number of visits over the 12-week period was 16 (range 8 to 28 visits). Exclusions from the study included nine patients who were lost to follow-up and did not complete the 12-week protocol, seven patients who were seen at other facilities due to insurance restraints, and six patients who were set up with modified protocols because of age (4 to 13 years) or severity of injury (associated fracture and adjacent amputation, severe crush), as requested by the surgeon. The youngest patient to which this protocol was applied without modification during this period was 16 years old.
Protocol Description A dorsal blocking splint is fabricated using thermoplastic material, with the wrist in neutral, MCP’s 508 to 708 flexion, and IPs allowed full extension. In the presence of a digital nerve repair, the IPs were blocked at 158 flexion or as specified by the surgeon. The hood of the dorsal blocking splint extends to the fingertips, allowing the IPs to be strapped loosely in extension at night. Rubber-band traction is applied to all four fingers, run under a palmar pulley (safety pin or other attachment at distal palmar crease), and attached at the proximal forearm strap (Figure 1). Exercises are performed hourly for 10 repetitions. The patient is instructed in initial edema control consisting of elevation and neck, shoulder, and elbow motion, with light wound bandaging. If needed, a self-adhesive circumferential wrap to the injured digit to decrease swelling is applied at night, if the patient or significant other is able to apply it safely. It is not used during the day because it may add resistance to the active hold exercises in flexion. Instruction is given for passive flexion of all joints of the fingers and full active IP extension to the dorsal hood of the blocking splint. Maximal passive MCP flexion using the contralateral hand, padded July–September 2003 201
FIGURE 1. Dorsal blocking splint with wrist neutral, MCPs 508 to 708 flexion, and rubber-band traction to all four fingers for the first five weeks. At night, the rubber-band traction is disconnected, and the fingers are strapped gently to the dorsal hood.
FIGURE 2. Use of padded aluminum placed behind the finger
resistance to the repaired flexor tendon before performing the gentle place–active hold in flexion. In therapy, the splint is removed for cleansing, wrist tenodesis exercises, passive flexion, active IP extension, and place–active hold with the wrist in 208 to 308 extension (Figure 5). Patients are seen in therapy one to three times per week, depending on their level of swelling, pain, passive range of motion, ability to hold in flexion with minimal effort, and ability to perform the exercises independently. At five weeks postoperatively, the rubber-band traction is discontinued. The dorsal blocking splint is continued, with fingers strapped to the dorsal hood between exercises. The patient is instructed to remove the splint at home for the exercises already described with the addition of active wrist extension and flexion with fingers relaxed (wrist tenodesis) and active finger flexion with the wrist at 308 extension. The hand is assessed for intrinsic tightness or MCP volar tightness, with intrinsic stretch performed pas-
during active IP extension to facilitate full extension.
FIGURE 3. Passive flexion is performed before the place–active
aluminum splint, or other method is instructed while performing the active IP extension against the resistance of the rubber bands, to minimize PIP flexion contractures (Figure 2). Rubber bands are allowed to be disconnected proximally during the active IP extension to attain the full amount of extension allowed when needed. Place–active hold in flexion is then performed, with passive flexion of all digits by the contralateral hand (Figure 3) followed by gentle active hold of the fingers in the flexed position for 2 to 3 seconds when released by the contralateral hand (Figure 4). No pressure of the fingertips on the palm or palmar strap is allowed. The rubber-band traction is detached from the forearm attachment for the place–active hold exercises. If passive flexion is limited by swelling, stiffness, or pain, the patient is instructed to perform stronger passive flexion to tolerance, using the contralateral hand, to decrease joint stiffness and
FIGURE 4. Active hold in flexion of all fingers by a patient four weeks post repair of the flexor digitorum profundus and radial digital nerve in the small finger.
202
JOURNAL OF HAND THERAPY
hold exercises. The fingers are placed gently into flexion by the patient’s other hand.
FIGURE 5. In therapy, the splint is removed to allow wrist tenodesis motion to 208 to 308 extension during the place–active hold finger flexion exercises.
sively in therapy only, with the wrist flexed to protect the repaired tendon, if needed. At six weeks postoperatively, these exercises are continued, with the addition of composite extension of all fingers with the wrist neutral. At eight weeks postoperatively, passive IP extension exercises are added to the home program if flexion contractures are present, as determined appropriate by surgeon and therapist. The dorsal blocking splint is cut to free the wrist. The hand-based dorsal blocking splint is worn when active or at work, to prevent inadvertent resistance to the flexors (Figure 6). At 12 weeks postoperatively, all protective splinting is discontinued. The hand is allowed to be used normally, avoiding strong tip resistance for another 2 weeks.
RESULTS Thirty-five subjects, with 40 flexor tendon repairs in zones I, II, and III, were evaluated with the
Strickland/Glogovac formula41 with active range of motion at the end of treatment, 12 to 14 weeks postoperatively. The results show 29% excellent, 43% good, 21% fair, and 1% poor results in zone I; 53% excellent, 42% good, 0% fair, and 0% poor results in zone II; and 100% excellent, 0% good, 0% fair, and 0% poor results in zone III (Table 1). The only rupture occurred in a patient performing an activity not allowed by the protocol at 4.5 weeks postoperatively (using the repaired index finger on a knife to scoop peanut butter out of a jar). The tendon was immediately repaired again, and the patient went on to an excellent result. Flexion and extension averages were calculated, to determine more specifically the motion recovery tendencies. The average flexion of the PIP and DIP joints for zone I was 1538; zone II, 1628; and zone III, 161.58. Average loss of IP extension was 208 in zone I, 118 in zone II, and 4.58 in zone III. Total active motion averages of the IP joints were 1338 for zone I, 150.58 for zone II, and 157.58 for zone III (Table 2). Results of zones I and II were calculated separately for comparison with other studies (Table 3). All studies reported their results using the Strickland/ Glogovac criteria,41 with the exception of the 1989 Belfast study, which used ASSH criteria.19 Silfverskiold and May,15,16 including repairs in zone II, showed 71% excellent, 25% good, 0% fair, and 0% poor results, with one rupture. No complex injuries were included. Strickland and Gettle,14 including repairs in zones I and II, showed 24% excellent, 46% good, 24% fair and 5% poor results, with three ruptures. Evans,17,18 including repairs in zone II applied to two-strand repairs, showed 43% excellent, 32% good, 14% fair, and 11% poor results, with one rupture. Complex injuries were included. Of the studies reported by Gratton19 (also applied after twostrand repairs), results of the Belfast study in 1989 showed 46% excellent, 31% good, 14% fair, and 9% poor results, with 11 ruptures; the Sheffield study in 1989 showed 71% excellent, 6.5% good, 3.5% fair, and 19% poor results, with 2 ruptures; and the Sheffield study in 1990 showed 49% excellent, 36% good, 11% fair, and 4% poor results, with 10 ruptures. This study showed 42% excellent, 42% good, 9% fair, and 3%
TABLE 1. Results of Flexor Tendon Repair by Zone of Injury Using the Strickland/Glogovac41 Formula for Grading Flexor Tendon Results* No. Zone Digits Excellent I II III
FIGURE 6. Hand-based dorsal blocking splint worn intermittently between weeks 8 and 12 during work or heavy activity to remind the patient to avoid strong resistance to the fingers.
14 19 7
Good
Fair
Poor Ruptures
4 (29%) 6 (43%) 3 (21%) 1 (7%) 0 10 (53%) 8 (42%) 0 0 1 (2.5%)y 7 (100%) 0 0 0 0
*See text for description of formula and classification system. yRupture calculated as 2.5% of all digits and 3% of digits in zones I and II. The rupture occurred in a noncompliant patient.
July–September 2003 203
TABLE 2. Range of Motion Averages at Interphalangeal Joints Zone I II III
Flexion Average (8)
Extension Loss Average (8)
PIP 100 DIP 53 PIP 98.5 DIP 63 PIP 102 DIP 60
PIP 10.5 DIP 9.5 PIP 8 DIP 3 PIP 4.5 DIP 0
Total Active IP Motion Average (8) 133 150.5 157.5
poor results in zones I and II. Zone II calculated alone showed 53% excellent, 42% good, 0% fair, and 0% poor results, with one rupture. Complex injuries were included. Functional results and limitations were assessed using the DASH (Disabilities of the Arm, Shoulder, and Hand) survey.43 The DASH survey was mailed to the 35 patients in the study, and 13 (37%) surveys were returned. In the physical function and symptoms category, the average DASH score was 7.82 on a scale of 0 to 100, with 0 being no disability and 100 being severe disability. In the sports/performing arts category, the average DASH score was 16.07. In the work category, the average DASH score was 10.23.
DISCUSSION The protocol described in this article was developed after consideration of protocols already designed with good results and has modified various components to fit the needs of our facility. After a four-strand repair, we wanted to attempt an active motion protocol with as many of the patients as possible, while minimizing the potential for rupture. The wrist neutral position was chosen based on Silfverskiold and May’s study15,16 to allow the patient to remain in the same splint for rest and exercise. The splint was designed as a thermoplastic, removable splint that extends the full length of the fingers. Rubber-band traction has been used by Dovelle and Chow et al.,23,24 Lister et al.,25 and May et al.26 Four-finger rubber-band traction, included to
prevent active use of the fingers between exercises, increase passive flexion, and maximize flexor digitorum profundus excursion and IP extension, has been used by Evans and Thompson,18 Silfverskiold and May,15,16 and Karlander et al.44 Inclusion of wrist tenodesis exercises is based on studies by Cooney et al.30 and Horii et al.31 The longer protection period with a hand-based splint was added by this group, after two ruptures in the earlier design of this protocol that occurred between 6.5 and 10 weeks postoperatively, at which time the splint was allowed to be removed completely at six weeks. The ruptures occurred in two patients with excellent results when they used their fingers with DIP resistance without prior thought. There have been no ruptures in patients compliant with the extended splint use in the protocol as described in this article. Patients experiencing limited motion due to adhesions may be advanced more quickly through the protocol. The problem of later ruptures and the need for a graduated level of protection through the full length of tendon healing are supported by other authors. Strickland and Gettle14 noted the occurrence of ruptures in ‘‘patients who were doing so well with their rehabilitation program that they elected to use their hands in an extremely strong manner (such as heavy lifting) well before the tendons had returned to sufficient strength to tolerate the extremely high tensile demands of such activity.’’ Silfverskiold (personal communication) suggested that in cases with a good or excellent result he would not recommend significant resisted exercises until 11 to 12 weeks postoperatively. When comparing results of studies, it is important to be aware of whether complex injuries were included within the results because one can expect the results of a study of simple tendon lacerations to be better than those that include crush injuries or injuries and repair to other structures. Gault28 discussed the tendency of flexor tendon repairs with volar plate injuries to develop flexion contractures. Elliot et al.20 and Gerbino et al.45 discussed difficulty in zone I in obtaining a gliding tendon repair under the A4 pulley. In the present study, zone I repairs
TABLE 3. Comparison with Other Early Active Motion Protocols Protocol
No. Digits
Excellent
Good
Fair
Poor
Ruptures
Silfverskiold/May* (zone II) Strickland/Cannon/Gettley (zone I,II) Evansz (zone II) Belfast 1989z (zone II) Sheffield 1989z (zone II) Sheffield 1990z (zone II) Klein (this study)y (zone I, II) (zone II alone)
55 40 44 117 31 47 33 19
39 (71%) 9 (24%) 19 (43%) 46% 71% 49% 14 (42%) 10 (53%)
14 (25%) 17 (46%) 14 (32%) 31% 6.5% 36% 14 (42%) 8 (42%)
0 9 (24%) 6 (14%) 14% 3.5% 11% 3 (9%) 0
0 2 (5%) 5 (11%) 9% 19% 4% 1 (3%) 0
2 3 1 11 2 10 1 1
*Two-strand plus cross-stitch repair. zTwo-strand repair. yFour-strand repair.
204
JOURNAL OF HAND THERAPY
showed a smaller number of excellent results compared with other zones; however, 35% of the injuries in zone I were complex, including volar plate or collateral ligament repairs. Most had good flexion but developed IP flexion contractures between 58 and 308, resulting in less overall motion. With this in mind, the author will consider modifying the use of rubber-band traction with this protocol with patients who have had a volar plate or collateral ligament injury in addition to flexor tendon injury. Earlier intervention with passive IP extension in therapy with the flexor tendon in a protected position with proximal joints held flexed also is recommended and now is being performed by the author within the protocol as soon as limited IP extension is noted. Intermittent positioning during the day, alternating between rubber-band traction and strapping the IPs in extension in the dorsal blocking splint, also may be helpful for patients showing early IP flexion contracture development. There are many limitations of this study. The retrospective format has inherent limitations, including the possibility of unrecognized factors affecting outcome, lower response rate on the DASH scale mail survey than if completed in the clinic, possible patient recall bias in completing the DASH survey, and potential reviewer bias. With this in mind, the charts were reviewed with strict application of the Strickland/Glogovac formula according to the final measurements in the chart for all patients completing the 12-week protocol. The amount of tension applied by the patient during the active hold in flexion was not measured, as in the Evans study.17,18 No ruptures were noted, however, during exercises or at rest using this protocol, with the only rupture occurring due to noncompliant use against resistance. Work of flexion, as described by Halikis et al,41 describes the sum of forces that resist flexion of the digit. Work of flexion is affected by the resistance offered to the flexor tendon by edema, joint stiffness, and adhesions. Early active motion protocols, such as this one, encounter the resistance of edema and joint stiffness when initially moving the digits due to trauma related to the injury and surgery. The study by Halikis et al41 showed that there is increased work of flexion developed when a tendon is mobilized immediately after surgery compared with a short delay of three to five days, with three days showing the development of the least amount of work of force. This protocol relied on education provided to the patient to apply a light active hold and the experience of the therapists in applying gentle passive flexion during the place–active hold, incorporating patient feedback as to how difficult it felt to hold the fingers in flexion. Practice with the contralateral hand, showing the light pull needed to hold the fingers gently in flexion and the way in which the wrist
position affects the amount of pull needed to achieve flexion (wrist tenodesis), is provided. Strong emphasis is placed on use of a light pull by the patient when applying the active hold component, with education on potential for tendon rupture with excessive internal tension by the patient. When passive motion was limited initially, emphasis was placed on the passive flexion exercises done before the place–active hold, to decrease joint stiffness gradually. Although not formally assessed, these methods are helpful in decreasing the force applied to a newly repaired flexor tendon during active flexion. There were no cases of rupture using this method in compliant patients. The results of this study reflect not only the protocol used, but experienced surgeons using atraumatic tendon handling techniques. Surgical technique was not controlled, and a variety of fourstrand techniques were used.
CONCLUSIONS The results of the retrospective chart review of this protocol for the management of four-strand flexor tendon repair show that it is safe (only one rupture), effective (87.5% good and excellent results in zones I, II, and III, including complex injuries, and in zone II alone, 95% good and excellent results, yielding high functional scores on the DASH survey), and simple.
REFERENCES 1. Savage R, Risitano G. Flexor tendon repair using a ‘‘six strand’’ method of repair and early mobilisation. J Hand Surg Br. 1989;14:396–9. 2. Komanduri M, et al. Tensile strength of flexor tendon repairs in a dynamic cadaver model. J Hand Surg Am. 1996;21:605–11. 3. Silfverskiold K, Andersson C. Two new methods of tendon repair: an in vitro evaluation of tensile strength and gap formation. J Hand Surg Am. 1993;18:58–65. 4. Shaieb MD, Singer DI. Tensile strengths of various suture techniques. J Hand Surg Br. 1997;22:764–7. 5. Strickland JM, Cannon NM. Flexor tendon repair—Indiana method. Indiana Hand Center Newsletter. 1993;1:1–19. 6. Kessler I. The ‘‘grasping’’ technique for tendon repair. Hand. 1973;5:253–5. 7. Trail IA, Powell ES, Noble J. The mechanical strength of various suture techniques. J Hand Surg Br. 1992;17:89–91. 8. Robertson GA, Al-Quattan MM. A biomechanical analysis of a new interlock suture technique for flexor tendon repair. J Hand Surg Br. 1992;17:92–3. 9. 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 Am. 1990; 15:953–8. 10. McLarney E, Hoffman H, Wolfe S. Biomechanical analysis of the cruciate four-strand flexor tendon repair. J Hand Surg Am. 1999;24:295–301. 11. Savage R. In vitro studies of a new method of flexor tendon repair. J Hand Surg Br. 1985;10:135–41. 12. Diao E, et al. Effect of periperal suture depth on strength of tendon repairs. J Hand Surg Am. 1996;21:234–9. 13. Wade PJF, et al. Flexor tendon repair: significant gain in strength from the Halsted peripheral suture technique. J Hand Surg Br. 1989;14:232–5.
July–September 2003 205
14. Strickland JM, Gettle KH. Flexor tendon repair: the Indianapolis method. In: Hunter JM, Schneider LH, Mackin EJ (eds). Tendon and Nerve Surgery in the Hand: A Third Decade. St. Louis, Mo.: Mosby, 1997:353–61. 15. Silfverskiold KL, May EJ. Flexor tendon repair in zone II with a new suture technique and an early mobilization program combining passive and active flexion. J Hand Surg Am. 1994;19:53–60. 16. Silfverskiold KL, May EJ. Flexor tendon repair with active mobilization: the Gothenburg experience. In: Hunter JM, Schneider LH, Mackin EJ (eds). Tendon and Nerve Surgery in the Hand: A Third Decade. St. Louis, Mo.: Mosby, 1997: 342–52. 17. Evans RB. The application of force to the healing tendon. J Hand Ther. 1993;6:266–84. 18. Evans RB, Thompson DE. Immediate active short arc motion following tendon repair. In: Hunter JM, Schneider LH, Mackin EJ (eds). Tendon and Nerve Surgery in the Hand: A Third Decade. St. Louis, Mo.: Mosby, 1997:362–93. 19. Gratton P. Early active mobilization after flexor tendon repairs. J Hand Ther. 1993;6:285–9. 20. Elliott D, et al. The rupture rate of acute flexor tendon repairs mobilized by the controlled active motion regimen. J Hand Surg Br. 1994;19:607–12. 21. O’Driscoll SW, Horii E. The relationship between wrist position, grasp size, and grip strength. J Hand Surg Am. 1992; 17:169–77. 22. Savage R. The influence of wrist position on the minimum force required for active movement of the interphalangeal joint. J Hand Surg Br. 1988;13:252–68. 23. Dovelle S, Kulis Heeter P. The Washington regimen: rehabilitation of the hand following flexor tendon injuries. Phy Ther. 1989;69:42–8. 24. Chow JA, Linwood JT, Dovelle S, Monsivais J, Milnor WH, Jackson JP. Controlled motion rehabilitation after flexor tendon repair and grafting. J Bone Joint Surg Br. 1988;70:591–5. 25. Lister G, Kleinert H, Kutz J, Atasoy E. Primary flexor tendon repair followed by immediate controlled mobilization. J Hand Surg. 1977;2:441–51. 26. May EJ, Silfverskiold KL, Sollerman CJ. The correlation between controlled range of motion with dynamic traction and results after flexor tendon repair in zone II. J Hand Surg Am. 1992;17:1133–9. 27. May EJ, Silfverskiold KL, Sollerman CJ. Controlled mobilization after flexor tendon repair in zone. II: a prospective comparison of three methods. J Hand Surg Am. 1992;17:942–53. 28. Gault DT. A review of repaired flexor tendons. J Hand Surg Br. 1987;12:321–5. 29. Burge PD, Brown NB. Elastic band mobilisation after flexor tendon repair; splint design and risk of flexion contracture. J Hand Surg Br. 1990;15:443–8.
206
JOURNAL OF HAND THERAPY
30. Cooney WP, Lin GT, An KN. Improved tendon excursion following flexor tendon repair. J Hand Ther. 1989;2:102–13. 31. Horii E, et al. Comparative flexor tendon excursion after passive mobilization: an in vitro study. J Hand Surg Am. 1992;17:559–66. 32. Strickland JM. Development of flexor tendon surgery: twentyfive years of progress. J Hand Surg Am. 2000;25:214–35. 33. Becker H, et al. Intrinsic tendon cell proliferation in tissue culture. J Hand Surg. 1981;6:616–9. 34. Lundborg G, Rank F. Experimental intrinsic healing of flexor tendons based upon synovial fluid nutrition. J Hand Surg. 1978;3:21–31. 35. Manske PR, Lesker PA. Biochemical evidence of flexor tendon participation in the repair process—an in vitro study. J Hand Surg Br. 1984;9:117–20. 36. Gelberman RH, et al. The early stages of flexor tendon healing: a morphologic study of the first fourteen days. J Hand Surg Am. 1985;10:776–84. 37. Gelberman RH, Botte MJ, Spiegelman JJ, et al. The excursion and deformation of repaired flexor tendons treated with protected early motion. J Hand Surg Am. 1986;11:106–10. 38. Hitchcock TF, Light TR, Bunch WH, et al. The effect of immediate constrained digital motion on the strength of flexor tendon repairs in chickens. J Hand Surg Am. 1987;12:590–5. 39. Goldfarb CA, et al. The effect of variations in applied rehabilitation force on collagen concentration and maturation at the intrasynovial flexor tendon repair site. J Hand Surg Am. 2001;26:841–6. 40. Lister GD, et al. Primary flexor tendon repair followed by immediate controlled mobilization. J Hand Surg. 1977;2: 441–451. 41. Strickland JW, Glogovac SV. Digital function following flexor tendon repair in zone 2: a comparison study of immobilization and controlled passive motion. J Hand Surg Am. 1980; 5:537–43. 42. Halikis MN, Manske PR, Kubota H, Aoki M. Effect of immobilization, immediate mobilization, and delayed mobilization on the resistance to digital flexion using a tendon injury model. J Hand Surg Am. 1997;22:464–72. 43. McConnell S, Beaton DE, Bombardier C. The DASH Outcome User’s Manual. Toronto, Ontario: Institute for Work & Health, 1999. 44. Karlander LE, et al. Improved results in zone 2 flexor tendon injuries with a modified technique of immediate controlled mobilization. J Hand Surg Br. 1993;18:26–30. 45. Gerbino PG, Saldana MJ, Westerbeck P, Schacherer TG. Complications experienced in the rehabilitation of zone 1 flexor tendon injuries with dynamic traction splinting. J Hand Surg Am. 1991;16:680–6.