Extensor tendons in all zones except digital I and II and thumb TI and TII tolerate controlled active tension by the third post-operative day with precise orthotic positioning and controlled arcs of motion.
The concepts of immediate tension at the repair site(s) are supported by experimental, biomechanical and clinical studies.
Orthotic position, position of exercise, and angles of allowed flexion are dependent on the level of injury and status of adjacent tissues.
Extension should be regained before flexion in all zones.
Early intervention is critical to the ability of the therapist to manage edema, inflammation, impending joint problems, and tendon excursion during the first three weeks of wound healing.
* Author’s note: This chapter provides an update on extensor tendon rehabilitation as an addendum to the previously published chapter authored by myself in edition 5. For the most part my techniques of rehabilitation for early active motion are unchanged, based on clinical experience and updated literature review. New to this chapter are (1) my technique for early short arc motion (SAM) for the closed (boutonnière) zone III, IV injury, (2) a review of the relative extension orthotic positioning technique described by Wendell and Howell, and (3) postoperative management suggestions for sagittal band repairs. Evidence to support the therapist’s interventions following extensor tendon repair with literature review are summarized, and references for all sections are updated.Problems that accompany the complex extensor tendon injury or mismanagement of the simple extensor injury are well known to the hand specialist. Most functional problems in tendon systems are associated with the tendon’s response to injury and repair, and despite decades of research on the subject, the problem of restrictive scar formation remains one of the most unpredictable factors contributing to postoperative morbidity. Over the past three decades, the techniques of early controlled motion initially applied to flexor tendon management have been slowly incorporated into standard postoperative care for the extensor system. We have learned, with clinical studies and experience, that
Extensor tendons in all zones (except zone I) tolerate controlled active motion.
Gapping and rupture are rarely an issue in carefully applied postoperative regimens that control forces and excursion.
More digital joint motion can probably be allowed with injury in zones V through VII than had previously been thought possible.
Wrist position is critical to decreasing resistive forces from the flexor system and is a factor in true tendon excursion gained with digital motion.
These tendons have most likely been moving actively all along within the confines of their dynamic orthoses.
Early referral to therapy with attention to orthosis geometry and applied stress is a critical variable to outcomes.
Although experience and literature review continue to support the use of early postoperative controlled motion, some investigators report similar results with postoperative immobilization protocols. However, for the most part, studies published in the past 5 years on postoperative management technique basically verify earlier studies, which demonstrated that extensor tendons respond well to the application of controlled motion with the use of both dynamic orthoses and exercise. The only difference in protocols is simply some variation in orthosis design and in attempts to incorporate an active component into postoperative regimens. The principles of treatment are the same as for the previous edition of this text: Initiate therapy by postoperative day 3, rest the tendon short (the repair site proximal to its normal resting position), and control excursion with an orthosis design that allows at least 5 mm (or more) of motion at the repair site.
A literature search for extensor tendon rehabilitation (randomized controlled trials) from 2004 to 2009 † produced only one study that systematically reviewed available evidence comparing the effectiveness of different rehabilitation regimens on repaired extensor tendons. The results of this systematic review, which included four randomized controlled trials and one other design study, found that short-term outcomes after immobilization were inferior to early motion programs, and that there was some limited support for early controlled mobilization over early active mobilization at 4 weeks. No conclusive evidence was found, however, regarding the long-term (12 weeks) effectiveness of the different rehabilitation protocols.
† The literature search included the following url and databases www.ncbi.nlm.nih.gov/sites/entrez , Cochrane Library, MEDLINE, PEDro, EMBASE, and CINAHI.The positive biomechanical effects of mechanical stimulation to a repair site continue to be supported with basic science research, and further study of the mechanism for the loss of mechanical properties associated with mechanical unloading (cast immobilization) in the tendon continues to support previous studies that have demonstrated a rapid and significant decrease in material properties associated with immobilization of tendon and ligament. Buckwalter and Grodzinsky point out that one of the most important concepts in orthopedics in this century is the understanding that loading accelerates healing of bone, fibrous tissue, and skeletal muscle, and thus we can surmise that early controlled motion to the healing tendon, as just noted, remains our best current treatment for tendon repair.
The concept of tissue engineering, which is the focus of much current research, is based on the manipulation of cellular and biochemical mediators that will positively affect protein synthesis and improve tissue remodeling. Woo and colleagues have predicted that the combination of cell therapy with growth factor application via gene transfer will offer new opportunities for improving tendon and ligament healing. Although basic science studies continue to explore the healing of connective tissue through tissue engineering and the effects of mechanical stress, early intervention with edema control, precise orthosis geometry, and short arc motion designed to glide the repair site in a controlled range remain our most reliable techniques for producing excellent functional results. Even though these new biologic therapies show much promise for facilitating bone and fibrous tissue healing in the future, others remind us that none has been proven to offer the beneficial effects comparable to those produced by the loading of healing tissues.
Most basic scientific and clinical research on tendons has been focused on the synovial flexor tendon, which has posed the most problems with restoration of functional gliding after repair. Clinicians have perhaps been too liberal in their application of some of these basic science studies to clinical application in the extensor system. Nonetheless, despite the histologic, metabolic, and nutritional differences between the two tendon systems and within the different levels of the extensor system, the fact is that all tendon is a type of dense connective tissue that functions to transmit muscle force to the skeleton, and to perform its work requirement, all tendon must glide relative to its surrounding tissues.
After injury to tendon at any level in either the flexor or the extensor system, the wound response and subsequent problems of maintaining tendon excursion are the same. A blood clot forms, a nonspecific inflammatory response occurs, the clot becomes populated with fibroblasts and advancing capillary buds, and collagen and ground substance are produced that will heal the deficit but that also may limit the tendon’s normal gliding ability. Loss of fiber gliding of tendon and neighboring connective tissues with resulting functional deficits of strength, mobility, and coordination impede the tendon’s ability to transmit muscle force to bone with mechanical efficiency. Rehabilitation of a healing tendon is simply reestablishing its ability to glide and transmit force without creating gapping or rupture at the repair site.
The purpose of this chapter is to define the problems that follow injury at the different extensor tendon levels and to offer solutions to these problems based on current experimental and clinical research in the areas of tendon repair techniques, tendon healing, true tendon excursions, and the application of force to the healing extensor tendon. Anatomy, clinical evaluation, and surgical management of the extensor tendons are addressed in this edition by Pratt (see Chapter 1 ) and Rosenthal and Elhassan (see Chapter 38 ) and elsewhere.
The serious student, inexperienced or not, is encouraged to study the exquisite review of extensor tendon anatomy and management by Rosenthal and Elhassan in Chapter 38 of this text.
The first section, General Considerations , is dedicated to a current literature review of basic science and clinical studies that support early motion for the healing tendon. Although this material may seem complex to the student or inexperienced therapist, it is essential information for any therapist or physician who assumes the responsibility of caring for an injured tendon. Treatment by protocol is appropriate only if there are no variables. Unfortunately, in the real clinical situation we are required to treat tendons with associated injury to bone, joint, nerve, and vessel in patients with associated metabolic, immunosuppressive, or personality problems, often referred by the nonspecializing surgeon, whose surgical technique and timing further complicate the case. This requires the therapist to be able to adjust treatment parameters based on a solid knowledge of wound healing, tendon repair tensile strength, and the effects of immobilization and early motion on the biochemistry and biomechanics of the healing tendon.
The questions once posed to me as a young therapist by Dr. Paul Brand (When? How often? How far? and How much?) when applying stress to healing connective tissue—are addressed in this chapter. These questions should be a part of the hand clinician’s clinical decision-making process with each tendon case if optimal care is to be delivered.
The second section, Clinical Management of Extensor Tendon Injuries , is dedicated to management by level of injury. Within this section, general guidelines for treatment by immobilization, early passive motion, and early active motion are defined in terms of orthosis positioning and stress application.
Characteristics of the extensor tendon vary at each level, dictating variations in treatment. The committee on tendon injuries for the International Federation of the Society for Surgery of the Hand defines extensor tendon injury by delineating seven zones for the extrinsic finger extensors and five zones for the thumb extensors ( Fig. 39-1 ).
Schedules for immobilization, the application of controlled stress, and progressive resistive exercises depend on the tensile strength of the repair technique and the stage of wound healing as they relate to the physiologic and biomechanical differences of the tendon in its different zones or anatomic levels. Guidelines for treatment based on the biochemical and biomechanical requirements of this system must be altered to accommodate the circumstances of the individual patient and injury. The surgeon should apprise the therapist of the quality of the repair, the type of repair, alterations in tendon length, the integrity of the tissue, the status of surrounding tissues, and any additional pathologic conditions that might alter the amount of controlled stress that the healing tendon can accommodate. The patient should be evaluated in terms of anticipated compliance. This information will influence any variation from suggested timing of immobilization and mobilization schedules. Therapeutic management should be considered in terms of the biochemical and biomechanical events of wound healing and the effects that our management techniques have on these events. Management of the inflammatory state; timing of stress application; the judicious application of controlled stress; and the effects of active versus passive motion, orthosis geometry, the position of exercise, external load application with stress application, and the duration of exercise all influence either positively or negatively the healing and remodeling of this fibrous connective tissue.
The importance of early repair, nontraumatic surgery, prevention of infection, and edema control as a means of minimizing the inflammatory response after injury is well documented in the literature. The fibroblastic response and ultimately the amount of collagen produced at the wound site is proportional to the local inflammatory response. Complex dorsal injuries are accompanied by significant edema and may require bulky, compressive dressings between therapy sessions for the first 5 to 7 days, but injuries with less tissue response can be lightly dressed and placed in early motion orthoses by 24 hours after surgery. The digits should be individually wrapped with a single layer of 2-inch Coban to control digital edema for as long as any extra volume of fluid is present about the proximal interphalangeal (PIP) joint, a time that may exceed 8 to 12 weeks after repair for the digital injury. Elevation for 24 hours, motion for the uninvolved joints, and controlled motion for the involved joints help control edema but require precise participation by the patient. Patient education (complete with explanations of anatomy, wound healing, and precautions) is a critical component of rehabilitation. Most people want to get well and will comply if they understand the importance of their actions; however, a recent study examining patient compliance with orthosis use and home program following tendon repairs found through an anonymous self-reported questionnaire of 76 subjects that 67% of patients did not adhere to their orthosis regimen; there was no significant correlation between patient profile and nonadherence. Along those same lines in a study of causative factors to identify the reasons for zones I and II flexor tendon ruptures, it was found that half of the ruptures studied were due to “acts of stupidity.” The experienced clinician knows full well that patient compliance depends on instruction that requires time, patience, knowledge, and firm control on the part of the health care provider.
Proper management of the untidy, open, or cleanly incised and sutured wound is the first important component of preventing infection and controlling the inflammatory response. Careful cleansing, protection of the wound microenvironment, and proper dressing speed epithelialization and enhance macrophage activity so that time in the inflammatory stage is minimized. Exposed wounds tend to be more inflamed and necrotic than occluded wounds. In the later stages, the dermis of exposed wounds is more fibroblastic, fibrotic, and scarred.
Position of Immobilization
The tendon repair site, through the use of an orthosis, should be positioned proximal to the tendon’s normal resting position during the first 3 weeks of healing to minimize stress at the repair site and thus decrease the chances for gap formation. Gap formation has been associated with increased adhesion formation and poorer clinical results ( Fig. 39-2 ). It has been demonstrated (in an experimental flexor tendon study) that gapping of more than 3 mm at a repair site does not increase the prevalence of adhesions or impair range of motion (ROM), but it does prevent the increase in strength and stiffness that normally occurs with time. Any gapping at the repair site can result in elongation of the tendon callus, which is particularly critical in zones I to IV, where the tendon moment arms and excursions are small. Extensor lag in any zone at 3 or 4 weeks is much more difficult to overcome than extensor tightness and can be prevented by precise positions of immobilization.
Effects of Immobilization
Biomechanical and biochemical changes in immobilized connective tissue (tendon, ligament, and cartilage) have been studied primarily in the animal model in various joints. We must interpret the information gained from these experimental studies with caution as we attempt to alter clinical treatment based on basic science studies in the nonhuman model and more often than not on synovial flexor tendon.
The negative effects of total immobilization during the inflammatory and fibroblastic stages of healing on tendon biochemistry are a loss of glycosaminoglycan concentration, loss of water, decreased fibronectin (FN) concentration, and decreased endotenon healing. Biomechanically, the immobilized tendon loses tensile strength in the first 2 weeks after repair, and it loses gliding function by the first 10 days after repair.
Effects of Controlled Stress
Many elegant studies have demonstrated the positive influence of stress on healing tendon, with documented improvement of tensile strength, improved gliding properties, increased repair-site DNA, and accelerated changes in peritendinous vessel density and configuration. Motion may enhance the diffusion of synovial fluid within the tendon in synovial regions. Stress-induced electrical potentials may increase the connective tissue healing potential. Studies have demonstrated that early passive motion in a clinically relevant tendon-repair model increases FN concentration and fibroblast chemotaxis at the tendon repair site. The positive biomechanical effects of mechanical stimulation to a repair site continue to be supported with basic science research. Recent clinical studies continue to demonstrate the benefit of controlled motion over immobilization for the repaired extensor tendon.
Effect of Timing
Time from injury to repair of intrasynovial flexor tendons, considered as an isolated variable, has been demonstrated to have a significant effect on the function of tendon in dogs. Tendons repaired immediately were significantly improved over tendons repaired at 7 or 21 days with respect to angular rotation and linear excursion, but there were no significant differences in total concentration of collagen at the sites of repair or in the levels of reducible Schiff base cross-links in tendons from the three groups.
Gelberman and colleagues demonstrated in the canine model that immobilized tendons become bound by adhesions by the tenth day after repair but that tendons that are immediately mobilized have early restoration of gliding surface without adhesion in-growth.
Preliminary experimental studies indicate that timing in relation to stress during the early inflammatory stage of wound healing is critical. An experimental study on chicken flexor tendons has demonstrated that tendons treated with controlled passive motion have significantly improved tensile strength by 5 days after repair compared with digits treated with immobilization. The magnitude of difference in strength between the two groups increased with time. The authors of that study concluded that immediate constrained digital motion after repair allows progressive tendon healing without the intervening phase of tendon softening or weakening described in the classic study by Mason and Allen in 1941.
In another study of early tensile properties of healing chicken flexor tendons, early controlled passive motion was found to improve healing efficiency. Results of this study indicated that controlled passive-motion tendons had significantly greater values for rupture load, stress, and energy absorbed than immobilized tendons.
FN, which appears to be an important component of the early tendon repair process, has been localized in a clinically relevant tendon repair model. Fibroblast chemotaxis and adherence to the substrate in the days after injury and repair appear to be directly related to FN concentration. Early passive motion has been correlated with an increased FN concentration in the tendon repair model of the previous study. FN concentration in mobilized tendon was found to be twice that of immobilized tendon by 7 days after surgery. Iwuagwu and McGrouther have determined that load applied the first 5 days postoperatively resulted in better orientation and fewer fibroblasts in repaired tendons. Zhao and coworkers have demonstrated in the canine model that starting controlled motion at day 5 following tendon repair is advantageous over day 1 because gliding resistance associated with postoperative surgical edema and other factors is diminished.
We still have no studies on the effect of immediate motion on the healing of the in vivo human tendon, and one must recognize that most of the experimental tendon work is performed on the synovial flexor tendon in animal models. However, these basic science studies offer some documentation that increased cellular activity and strengthening occur with very early motion during the immediate postrepair period and emphasize the critical relationship between the application of force and timing. “Early motion” at 7 to 10 days after surgery may indeed not be early motion at all. By the end of the first week, the window of time may have been lost for the biochemical advantages of immediate motion, and by the tenth day after surgery, the immobilized tendon may be surrounded by dense adhesions.
Duration of Exercise
The duration of the daily controlled passive-motion interval has been determined to be a significant variable in a clinical study of repaired flexor tendons. A prospective multicenter clinical study of 51 patients with flexor tendon repair has determined that greater durations of daily controlled passive motion after flexor tendon repair resulted in increased active interphalangeal (IP) joint motion at a mean time of 6 months after surgery. Two groups of patients treated with continuous passive motion (CPM) and traditional early passive motion were compared in terms of IP motion. The authors concluded that the duration of the daily controlled motion interval is a significant variable in postrepair flexor tendon excursion.
In a related study, designed to determine the effects of frequency and duration of controlled passive motion on the healing flexor tendon after primary repair, adult mongrel dogs were studied as two groups based on the frequency of passive motion. Results indicated that gliding function in both groups was similar, but tensile properties, as represented by linear slope, ultimate load, and energy absorption, were significantly improved in the higher-frequency group. The authors concluded that the frequency of controlled passive motion in postoperative tendon management protocols is a significant factor in accelerating the healing response after tendon repair, and higher-frequency controlled passive motion has a beneficial effect.
These studies offer some proof that “more is better” in the early healing phase. However, in my clinical experience, the patient who exercises excessively may develop inflammation and synovitis within the synovial areas of both flexor and extensor systems if tendon gliding is limited by adhesions that increase friction or if the tendon is still swollen from increased metabolic activity associated with healing. Patient compliance is a significant and often difficult-to-control variable with these postoperative regimens. Dobbe and associates have developed a device to record duration of exercise that is attached to the postoperative orthosis. This concept may prove to have clinical relevance by improving patient compliance, but to date no practical use has been reported. In fact, recent studies have reported that the biggest cause of tendon rupture is “stupidity” and, as noted in a previous section, noncompliance.
Extensor Tendon Excursion
Tendon excursion in the early healing phases of tendon rehabilitation should be limited to a range that is great enough to provide the stress necessary to stimulate biochemical changes at the repair site and to provide some proximal migration of the repair site to control the collagen bonds as they form in the peritendinous region, yet small enough that it does not create gapping or rupture at the repair site. Researchers have raised the question of actual tendon excursion with passive motion. Some tendon researchers believe that a component of controlled active motion may be necessary to create some proximal migration of a tendon repair site and that passive motion may only cause the repair site to fold or buckle instead of glide proximally. This is the rationale for controlled early active motion as opposed to controlled passive motion in postoperative management of the repaired tendon. References to tendon excursion in the next section with cadaveric measurement, measurement by radians, and intraoperative measurement all refer to the relationship of passive joint motion to tendon excursion.
The tendon migration necessary to maintain functional glide and stimulate cellular activity may be in the range of 3 to 5 mm. Duran and Houser recommended this passive excursion range for the flexor tendons in the digital sheath and thought that 3 to 5 mm was sufficient to prevent dense adhesions. Gelberman and colleagues, in an earlier study, suggested that 3 to 4 mm of passive excursion is necessary to stimulate the intrinsic repair process without creating significant repair-site deformation with flexor tendons. More recent studies have indicated that 1.7 mm of tendon excursion is sufficient to prevent adhesion formation in canine tendon and that additional excursion provides little added benefit. Early active and passive motion allowing an estimated 5 mm of excursion has proven to be successful with extensor tendon repairs in zones V through VII, T-IV, and T-V, and approximately 4 mm of active excursion with extensor repair in the digital zones III and IV.
To safely apply stress to a healing tendon, the therapist must understand tendon excursion as it relates to joint motion and understand tendon tensile strengths as they relate to suture techniques and healing schedules. This requires either a general working knowledge of tendon excursions as cited in the literature or the ability to calculate individual tendon excursions with Brand’s technique of using the radian concept and corresponding joint moment arms.
Literature Review of Extensor Tendon Excursions
Reported extensor tendon excursions vary but occur within a consistent range. Differences may exist between the individual extensor digitorum communis (EDC) tendons, as well as from person to person. Variation is also found in the method of study, as well as the size of the hand or joint being examined. Cadaveric studies are limited by the absence of normal biochemistry and muscle tone.
Bunnell’s cadaver studies of excursions provide detailed information and closely correlate with those described by Brand. Bunnell assigned values for individual finger tendons at each metacarpophalangeal (MCP), PIP, and distal interphalangeal (DIP) joint, with the wrist in a neutral position ( Table 39-1 ). The excursions become smaller as the joint size (and thus the tendon moment arm) decreases.
|Extensor Digitorum Communis
|Extensor Pollicis Longus
Excursions for the digital extensor in zones III and IV are small but may be slightly more than those reported by Bunnell, who calculated the EDC excursion to be 2 mm for the index finger, 3 mm for the long finger, 3 mm for the ring finger, and 2 mm for the small finger. In other cadaveric studies, Tubiana cites 8 mm, Valentine cites 7 to 8 mm, Zancolli cites 6 mm, and DeVoll and Saldana cite up to 5 mm of tendon excursion at this level.
An and coworkers calculated tendon excursion and the moment arm of cadaver index finger joints during rotation and found that excursion and joint displacement were not always linear. Excursion of the extensor digitorum at the PIP level with a mean motion of 89.5 degrees was 5.58 mm. Micks and Reswick determined that the extensor moment arm at the PIP joint is not constant and increases with the position of flexion.
Elliot and McGrouther investigated the mathematic relationship between extensor tendon excursion and joint motion in seven cadaver hands and found this relationship to be linear for all joints in all five rays of the hand. They found that excursion of the middle slip over the proximal phalanx is in effect the excursion of the extensor digitorum that accompanies PIP joint movement. They calculate excursion per 10 degrees for each joint with all other joints immobile and all surrounding structures released. The mean motion per 10 degrees was 0.8 mm for the index, long, and ring PIP joints and 0.6 mm for the small PIP joint. This translates to 2.4 mm of excursion for the index, long, and ring fingers; 1.8 mm of excursion for the small finger; and 1.8 mm of excursion for the small finger per 30 degrees of PIP motion. Their findings are contrary to those of An and coworkers.
Calculating Excursions with Radians and Moment Arms
Brand describes a constant relationship between joint motion and tendon excursion at both the MCP and PIP joints. A fairly constant extensor tendon moment arm (the perpendicular distance from the joint axis to the extensor tendon) exists at both joint levels. Although not precisely constant, the extensor tendon moment arm does not change dramatically with joint motion. Brand calculated the mean moment arm for the index MCP joint to be 10 mm in cadaver studies and for the middle finger PIP joint to be 7.5 mm. The moment arm varies with joint size, but these figures give us a working base to calculate approximate extensor tendon excursion.
Tendon excursion can be calculated geometrically with radians. A radian is the angle that is created when the radius laid along the circumference of a circle is joined by a line at each end to the center of the circle ( Fig. 39-3 ). This angle always equals 57.29 degrees (1 radian). This segment of the circumference of the circle is always equal to the radius of the circle when the working angle is 57.29 degrees.
To calculate extensor tendon excursion at the MCP joint level, consider the head of the metacarpal in terms of a circle ( Fig. 39-4A ). The moment arm of the extensor tendon (or the perpendicular distance between the MCP joint axis, or center of the circle, and the extensor tendon) is equal to the excursion of the extensor tendon if the MCP joint is moved through 1 radian, or 57.29 degrees. Using Brand’s figure of 10 mm for an average extensor tendon moment arm in the index finger, MCP joint motion of 57.29 degrees yields 10 mm of extensor tendon excursion (see Fig. 39-4A ). To obtain the 5 mm of excursion suggested by Duran and Houser and Gelberman and associates to minimize extrinsic adhesions, the joint must be moved through 0.5 radian, or 28.64 degrees, of rotation ( Fig. 39-4B ).
Because joint size varies, the therapist must consider that it is the constant relationship of tendon excursion to angular rotation and the length of the moment arm that is important. A smaller joint with a smaller moment arm produces less tendon excursion with the same joint motion. For example, if the MCP joint of the small finger has a moment arm of 7.5 mm, angular change of 0.5 radian, or 28.64 degrees, will produce 3.75 mm of glide. To obtain the necessary excursion to maintain tendon glide in the smaller ulnar joints or in smaller hands, one must move these joints through more than 0.5 radian of rotation.
Calculation by Simple Equation of Safe Parameters for Controlled Motion
Evans and Burkhalter proposed a simple equation for determining excursion of the extrinsic finger extensors in zones V through VII in the initial biomechanical study supporting early passive motion in these zones: Joint motion divided by tendon excursion for that particular joint is equal to the number of degrees of motion required to effect 1 mm of tendon glide.
Joint motion ( degrees ) / Tendon excursion ( mm ) = degrees of motion per millimeter of excursion
Application of this equation is contingent on total joint motion and total tendon excursion for each individual finger at the MCP level and the amount of excursion considered safe and effective for providing controlled stress to the healing tendon.
The suggested equation is applied with these average values for MCP joint motion: 85 degrees index; 88 degrees long; 90 degrees ring; and 92 degrees small finger. Excursions used were those described by Bunnell, because he measured each finger separately (see Table 39-1 ). Controlled stress allowing 5 mm of passive glide, as suggested by Duran and Houser and substantiated by intraoperative measurements in the pilot study, was determined to be a safe and effective excursion ( Box 39-1 ).
Index = 5.66 degrees per mm × 5 mm = 28.3 degrees
Long = 5.5 degrees per mm × 5 mm = 27.5 degrees
Ring = 8.18 degrees per mm × 5 mm = 40.9 degrees
Small = 7.66 degrees per mm × 5 mm = 38.33 degrees
Extensor tendon excursions were investigated in eight fresh cadaveric limbs. The authors found that if the wrist is extended more than 21 degrees, the extensor tendon glides with little or no tension in zones V and VI throughout a full simulated grip to full passive extension. On the basis of this cadaveric study, the authors recommend that up to 6.4 mm of tendon can be safely debrided in these zones and that full grip can be permitted postoperatively if the wrist is positioned in more than 45 degrees of extension. Their study emphasizes the importance of wrist position to tendon excursion, but their conclusions based on cadaveric study should be applied to the clinical situation with caution.
Evans and Burkhalter measured extensor tendon excursion intraoperatively and found by gross measurement that 30 degrees of MCP motion effected 5 mm of extensor glide in zones V through VII, supporting our calculations by radians with the previously described equation. This more limited amount of motion has worked well in my 34 years of clinical experience with early motion of extensor tendons, but others believe that full digital flexion should be considered safe within the confines of orthoses that hold the wrist in extension and fingers controlled in dynamic extension traction.
Cadaveric studies and mathematical equations do not consider biology. No study to date describes extensor tendon excursion after repair in vivo to give us accurate measurements for passive or active motion; it would seem that intraoperative measurement may be more accurate than cadaveric study.
Excursion of the Central Slip Measured in Radians
The same calculation techniques can be applied to the PIP joint and central slip excursion and are used to establish safe parameters for immediate early active motion for central slip repairs (discussed later). Biomechanically, the excursion of the extensor tendon at the level of the PIP joint is proportional to angular changes of the joint. The mean moment arm for the extensor tendon of the long finger PIP joint has been determined to be 7.5 mm. Therefore PIP joint motion of 57.29 degrees, 1 radian, effects 7.5 mm of excursion in the freely gliding tendon. One half (0.5) radian, or 28.64 degrees, effects 3.75 mm of excursion ( Fig. 39-5 ).
There is some disagreement in the literature regarding the extensor moment arm at the PIP joint. Micks and Reswick determined that the extensor moment arm at the PIP joint is not constant and increases with flexion. Brand found the moment arm of the extensor to be fairly constant at this level, unchanging with motion. An and coworkers found that excursion and joint displacement were not always linear at the PIP joint, but Elliot and McGrouther found that relationship to be linear.
The protocol for early motion for the extensor zones III through VII requires only 30 to 40 degrees of joint motion at the respective MCP or PIP joints; therefore the changes in the extensor moment arms are small and unlikely to be significant enough to alter the calculation of excursion.
Excursion of the Extensor Pollicis Longus
Excursions for the extensor pollicis longus (EPL) tendon vary in the literature from 25 to 60 mm. The simple angular arrangement of the flexion–extension axis at the MCP level of the fingers does not exist for the EPL in zones T-IV and T-V. Calculating excursion mathematically is complicated by the oblique course that the tendon takes at Lister’s tubercle, by the moments of adduction and external rotation at the carpometacarpal (CMC) level, and by the fact that alterations in thumb position alter the moment arms at each joint. Evans and Burkhalter measured EPL excursion intraoperatively to determine the amount of joint motion necessary to create 5 mm of glide for the early motion pilot study and found that with the wrist neutral and the thumb MCP joint extended, 60 degrees of IP joint motion effected 5 mm of tendon excursion at Lister’s tubercle.
True Tendon Excursion
The question of actual tendon excursion with passive motion has been raised. Experimentally (intact fresh frozen cadaver specimens), passive motion in flexor tendons has been demonstrated to be almost half that of theoretically predicted values under conditions of low tendon tension; investigators have shown that actual tendon excursion is equal to the predicted tendon excursions of earlier studies only when more than 300 g of tension is applied to the repair site. Similar studies correlating in vivo tendon tension and tendon excursion for the extensor system have not been performed, but cadaver studies suggest that the tendons may buckle only in zones V through VII with digital joint motion from flexion to extension.
The concept of using immediate active motion after tendon repair is simply an attempt to ensure that the repair site does glide proximally. Some clinicians now think that an as yet undefined degree of active tension at a repair site may be necessary to create a proximal migration of the healing tendon and that passive motion may only cause the tendon to buckle, fold, or roll up at the repair site. The use of immediate active motion as a means of restricting the limitation posed by adhesions and improving tendon gliding is neither new nor widely accepted in clinical practice, but it is now the subject of renewed interest in tendon management programs. Stronger suture techniques that are designed for active motion have been developed for flexor tendon repairs and for extensor system repairs. Favorable clinical results with active motion programs have been reported for both the flexor system and the extensor system.
Tendons rehabilitated in dynamic extension orthoses intended for passive motion programs in reality probably move actively throughout the early healing phases, so the discussion of “active versus passive” tendon excursion is most likely a moot point for extensors in zones V through VII when rehabilitated in such orthoses. Newport and Shukla performed electromyographic (EMG) studies on a group of normal volunteers to determine the level of activity present in the EDC within the confines of a dynamic extension orthosis. Their study validates what most therapists observe clinically—that most patients actively move their repaired and positioned extensor tendons within the dynamic extension orthosis and within the exercise regimen that is intended to provide active flexion and passive extension. They found that if the MCP joints were positioned in 0 degrees of extension, the EDC tendons were active within the dynamic extension orthosis; however, if the MCP joints were positioned in modest flexion (20 degrees), the extensor tendons were quiescent during the active flexion exercise. The authors of that study use this information to recommend that the MCP joints be positioned in 20 degrees of flexion to prevent active motion of the repaired tendon. I use their information to support dynamic positioning of the MCP joints at 0 degrees of extension to facilitate some physiologic active motion at the repair site. I believe that the MCP joints should be positioned at 0 degrees, not only to prevent extensor lag but also to create some active tension and true proximal migration of the repair site.
Application of Force
Force application is the sum of muscle contraction and viscoelastic drag of the tissues. Viscoelastic drag is the sum of the antagonistic muscle tension, resistance from the periarticular support systems, edema, and adhesions. Resistance from Coban wraps or bandaging also must be considered as an increased force application in early active motion programs. A safety margin must be established in which the force application (or the stress applied to the tendon) is less than the tensile strength of the tendon with all early motion programs.
Force Application at the Repair Site
Internal tendon forces as they relate to various joint angles and applied external loads are defined for the flexor and extensor tendons in two scientific articles on early active motion for both tendon systems. The results of these biomechanical analyses are presented in a series of mathematical models that negate resistance from the antagonistic muscle–tendon group and any other drag and apply a known external force. The force analysis of the EDC with the wrist in an extended position and no external load is calculated so that conservative estimates of tendon forces can be made ( Fig. 39-6 ). With the digital joints in a neutral position, tensions on the EDC are zero, but as the digits are extended, the forces rise to 1200 g. These forces would drop dramatically if the wrist were placed in 20 degrees of flexion (the position recommended for early active controlled motion) because resistance from the flexor tendons would be reduced by wrist position. The force applied to the extensor tendon at both the MCP and PIP joint levels with active extension of 30 degrees of flexion to 0 degrees of extension (at either joint) has been calculated mathematically to be approximately 300 g if the wrist is positioned at 20 degrees of flexion.
Tensile Strength of Extensor Tendon Repairs
The tensile strength of freshly sutured tendon depends on the strength of the suture material, the suture method, the balance between the strands and knot, the number of strands, the size of the tendon, and the addition of a circumferential suture to a core suture. Several studies have investigated the mechanical strengths of tendon repair techniques and suture materials for the extensor system. Newport and Williams reported on the biomechanical characteristics of extensor tendon suture at 2-mm gapping and at failure. The mattress suture gapped 2 mm at 488 g and failed at 840 g; figure-of-eight suture gapped at 587 g and failed at 696 g; the Kessler suture gapped at 1353 g and failed at 1830 g; and the Bunnell suture gapped at 1425 g and failed at 1985 g. A comparative study of four extensor tendon repair techniques demonstrated that strength to 2-mm gapping for the modified Becker suture technique (56.0 ± 9.2 N) and the modified Kessler technique (48.6 ± 12.6N) was significantly better ( p < 0.05) than for the six-strand double-loop and figure-of-eight techniques.
Most studies on repair tensile strengths report the strength in newtons (N), but as therapists, we usually calculate the forces of dynamic orthoses and torque ROM and the force of motion in grams. Evans and Thompson reviewed a large number of studies on the strengths of the various repairs and translated newtons into grams to assist therapists as they assess the strength of the particular repair they are treating (1 kg = 9.8 N, or 1 g = 0.01 N; conversion of newtons to grams: N divided by 9.8 × 1000 = g). Comparisons of these studies are difficult because of the many variables (subject, material, technique of repair, and method of testing) studied.
The load at which a tendon gaps is the number that we must recognize, particularly with the controlled active motion programs. Gap formation has been associated with increased adhesion formation and poorer clinical results. Although most surgeons believe that gapping above 1 to 3 mm is incompatible with a good result, investigators have demonstrated in an in vivo study that gaps of up to 10 mm in a repaired flexor digitorum profundus (FDP) are compatible with a good functional ROM when passive motion programs are used. Gelberman and colleagues have demonstrated that a gap at a repair site of more than 3 mm does not increase the prevalence of adhesions or impair ROM but does prevent the accrual of strength and stiffness that normally occurs with time.
Adjusting the Equation
The equations for force application and tensile strength just described must be adjusted to consider the increased resistance from postsurgical edema, stiff joints, and bandaging and to allow for a possible drop in tensile strength in the repaired tendon. The estimated tensile strength of the repair may decrease as much as 25% to 50% by postoperative day 5 to 15 in the unstressed tendon; however, tendon subjected to immediate or very early controlled motion may not experience this drop in tensile strength. The estimated force application to the repair site with the early active motion protocols may need to be doubled to account for the resistance from drag.
Effect of Complex Injury
The relationship between the amount of tissue damage and the biological response is a basic phenomenon of wound healing. Increased inflammation associated with severe injury increases the work requirement of the macrophage, and the number of macrophage cells necessary to meet the metabolic demands of the injury determines the number of fibroblasts that are signaled into the wound for repair. Collagen deposition can be expected to be proportional to the number of fibroblasts, or collagen-producing factories, present in the wound bed.
Rothkopf and coworkers studied mechanical trauma and immobilization in the canine flexor tendon model to study adhesion formation associated with complex injury. These researchers defined complex tendon injury as one associated with crush injury, concomitant nerve injury, or tendon injury treated with immobilization. Their experimental model demonstrated significant decreases in tendon excursions and an increase in work requirement to effect tendon excursion in the complex injury. This experimental model in the animal flexor tendon may have implications for the human flexor or extensor tendon.
We all have observed clinically that the more complex injury can be expected to cause more complications associated with increased fibroblastic response and that immobilization of the complex injury adds to those complications. Many authors have endorsed the use of early motion with the complex injury, but there are few clinical reports in the literature on early motion for the complex tendon injury, and most clinical results refer to clean lacerations. Koul and associates recently published the results of 21 complex extensor tendon injuries in zones V through VII treated with single-stage reconstruction, including soft tissue reconstruction rehabilitated with immediate active motion. They reported excellent functional results (total active motion [TAM] 202.6 degrees at 6 weeks and 249.5 degrees at 12 weeks, and 75% grip at 12 weeks) and reduced morbidity and cost of treatment with single-stage reconstruction and early active motion.
Clinical Management of Extensor Tendon Injuries
Zones I and II
A lesion of the terminal extensor tendon results in a flexion deformity of the DIP joint, commonly referred to as the mallet, or baseball, finger. Treatment and prognosis of the mallet finger depend on associated tissue injury and age of the lesion before treatment. These injuries may be open or closed, with or without associated fracture or fracture dislocation. In many cases, conservative treatment with orthotic immobilization is sufficient to restore tendon continuity. However, open injury, associated fracture, or chronic deformity may require direct repair or Kirschner wire (K-wire) fixation (see also Chapter 38 ).
Most authors recommend approximately 6 to 8 weeks of continuous extension positioning with a static orthosis for the middle and distal phalanx only with both conservative and operative treatment. Dagum and Mahoney have recommended that the wrist be positioned with a simple wrist control orthosis in addition to the distal joint orthosis to prevent gapping in zone I; however, in my clinical experience this does not seem to be necessary. Katzman and associates, in a cadaveric study of gap formation in mallet fingers, determined that joint motion proximal to the DIP joint and retraction of the intrinsics did not cause a tendon gap in a finger with a mallet lesion, supporting the concept that positioning one joint is sufficient for these injuries. Honner recommends some limited active flexion at 4 weeks, with continuous positioning between exercise periods for an additional 4 weeks. The DIP joint can be immobilized with commercially available Stack orthoses, aluminum-padded orthoses, or molded thermoplastic orthoses ( Fig. 39-7 ). Orthosis application is most often volar to the level of the PIP joint. A wide plastic tape placed across the dorsal aspect of the DIP joint acts as a counterpressure and holds the DIP joint in complete extension. I prefer to use Transpore tape by 3M with a small square of moleskin lining the portion of the tape that is directly over the DIP joint. Dorsal immobilization permits more freedom of the PIP joint and allows the fingertip its sensory function; however, in the hands of my patients, dorsal positioning ( Fig. 39-8 ) has not been as effective as volar positioning. Clinical experience teaches that early intervention for the zone I extensor lesion with orthotic positioning yields the best results in terms of extensor lag and time required for immobilization; however, Jablecki and Syrko recently made the point in a review article that “the period of time for which conservative treatment can be prolonged and still be effective is still being extended, and the absolute outside time limit remains unknown.”
Orthosis position and skin integrity should be monitored carefully. The distal joint should be immobilized at 0 degrees of extension or slight hyperextension. Extreme hyperextension jeopardizes circulation to the dorsal skin by stretching the volar vasculature, which provides nutrition to the area distal to the termination of the dorsal vessels, and may create skin necrosis. Rayan and Mullins, in a study of skin necrosis complications associated with mallet finger positioning, suggested a position of hyperextension less than the angle that causes skin blanching, a precursor of skin necrosis. They determined the average total passive hyperextension of the distal joint to be 28.3 degrees and found that circulation to the dorsal skin was compromised when the distal joint was positioned at more than 50% of its total hyperextension. Orthosis immobilization that allows even slight flexion results in extensor lag because the tendon callus heals in an elongated position.
Skin maceration is a problem with these injuries. It is difficult for patients to keep the affected hand dry for 6 to 8 weeks, and most patients find it irksome to be so limited by a one-joint injury. Patients must be instructed in proper orthotic application, skin care, technique for maintaining the DIP joint in extension during cleansing (I usually teach them to use the ipsilateral thumb to hold the DIP joint in hyperextension while cleansing with the contralateral hand), orthotic positioning, and orthotic adjustment to make sure that the distal joint always rests in complete extension. The orthoses can be lined with moleskin to absorb perspiration, and patients should be instructed to change the lining if it becomes damp. The DIP joint must be held in hyperextension while the patient changes the orthosis lining. I provide two distal joint orthoses; one can be worn while showering. The orthosis must be adjusted as edema decreases to provide a precise fit.
During the immobilization phase, the patient should be seen weekly for wound care when necessary, for adjustment of the fit of the orthosis, and for maintenance of motion in the unaffected joints. The distal joint must be held in extension continuously during orthotic adjustments to prevent attenuation of the healing tendon.
If the PIP joint develops a posture of slight hyperextension, the PIP joint should be positioned at 30 to 45 degrees of flexion while the DIP joint is held in complete extension ( Fig. 39-9 ). This position advances the lateral bands and may assist in closer approximation of the torn extensor tendon at the DIP joint. Doyle describes a treatment for the mallet finger with plaster casting of the PIP joint at 60 degrees and the DIP joint in slight hyperextension. He points out that in most cases, PIP immobilization is not necessary for these injuries but that casting both the PIP and DIP joints is a workable solution for patients who are unreliable or who are unable to understand or perform the correct application of an orthosis. Bunnell explained the rationale for the 60-degree flexion angle of the PIP. In this position, the lateral bands are advanced a distance of 3 mm. This much flexion could result in flexion contracture of the PIP joint, and clinically 45 to 60 degrees of PIP flexion has worked well for me. Flexion contracture of the PIP joint has not been a problem with PIP flexion positioning at these angles in my practice. Positioning the PIP joint may not be necessary beyond the first few weeks, after which the long orthosis can be exchanged for a shorter one-joint orthosis for the DIP. The most effective combination is a P 2 , P 3 orthosis taped firmly over the DIP joint with an interlocking dorsal orthosis that holds the PIP joint at 45 to 60 degrees of flexion. The patient exercises the PIP joint to full flexion six times per day (15 repetitions) while the DIP joint is held immobile in the volar orthosis to ensure full motion of the PIP joint.
After 6 weeks of uninterrupted positioning in extension, very gentle active flexion exercises are initiated. The opposing FDP is a powerful musculotendinous unit and easily overpowers the more fragile terminal extensor tendon. Brand and Hollister calculated the work capacity of the extensors to be less than one third of that of the flexors; therefore flexion increments should be obtained gradually with the initial emphasis on active extension. Because the moment arm of the extensor tendon at this level is small, so also is extensor tendon excursion.
Instructions to the patient should be very specific. During the first week of mobilization, no more than 20 to 25 degrees of active flexion of the distal joint should be allowed. Exercise duration is empirically prescribed at 10 to 20 repetitions every couple of hours. During the second week, if no lag has developed, distal joint flexion to 35 degrees may be allowed. The overly ambitious patient benefits from a template exercise orthosis with specific angles of motion preset to prevent overstretching of the terminal tendon ( Fig. 39-10 ). If the distal joint is tight in extension, the oblique retinacular ligaments (ORLs) may need to be stretched by manual immobilization with the PIP joint at 0 degrees of extension while the DIP joint is actively or passively flexed.
If an extensor lag develops, repositioning is indicated and exercises are delayed for a few weeks. Positioning between exercise sessions is recommended during the first 2 weeks of mobilization (a total of 8 to 10 weeks after injury), and night positioning with a volar static orthosis should be continued for an additional 4 weeks after intermittent daytime positioning has been discontinued.
Early active or passive motion has not been accepted practice for the tendon at this level, where excursions are small and where the tendon tissue becomes stiffer and more cartilaginous. However, because of inconsistent clinical results with these injuries and problems of loss of both flexion and extension, earlier motion techniques have been investigated. Nakamura and Nanjyo published their experience with surgical intervention with a wire implant and K-wire pinning for 3 weeks in 15 patients with freshly injured mallet fingers (average time from injury to surgery, 19.4 days) without associated fractures. The K-wire is removed at 3 weeks, and distal joint motion in gradually increased increments is allowed. The wire in the tendon stumps is removed at 5 weeks. Nakamura and Nanjyo report improved ROM and fewer complications with this technique, which also reduces immobilization time. Zlatkov reported improved results for the zone I and II extensor tendon operated with “a “special tendon suture” and treated with early active motion (62.5% excellent, 20.8% good, 12.5% satisfactory, and 4.2% unsatisfactory) over those treated with immobilization (45.7% excellent, 22.9% good, 20% satisfactory, and 11.4% unsatisfactory).
Prehension and coordination activities should supplement ROM exercise. Desensitization of a painful fingertip may be necessary with crush injuries or nail bed injuries before the patient incorporates the digit into prehension activities. Exercise may gradually proceed to resistive grasp and pinch activities. Flexion angles should be increased only if complete extension is maintained, and full flexion should not be attempted before 3 months.
Therapy for the zone I and II extensor tendon injury is primarily educational. If the patient understands the nature of the injury and the rationale for treatment, he or she should be able to perform most of the therapy independently. I typically monitor these patients once a week for correct orthosis position, compliance, motion at the PIP joint, and skin issues. This is a very irritating injury for most patients. I find that most patients with zone I and II injury need a great deal of encouragement to stay with the program and to be precise with positioning and motion protocols.
One systematic review was found in a Medline search questioning evidence to support interventions on the zone I and II extensor tendon injury. The results of the four randomized controlled studies, “all of which had methodological flaws,” determined that there was insufficient evidence to support either customized or “off the shelf” orthoses and that there was insufficient evidence to determine when surgery is indicated.
Zones III and IV
Extensor injuries in zones III and IV may result in a boutonnière deformity (see Chapter 38 ). The natural progression of the zone III extensor injury is well defined in the literature. Untreated, the lacerated middle band retracts, allowing the lateral bands to carry the full force of the extrinsic extensor tendon. The lateral bands migrate palmarward, act as flexors of the PIP joint, and with an increase in effort to extend the PIP joint, actually hyperextend the DIP joint. With time, the tendon and the retinacular tissues tighten and accommodate to the change in joint posture, tightening to a point where they resist even passive correction of the deformity. The relationship of the zone III tendon over the PIP joint and the zone IV tendon over the proximal phalanx to the lateral bands does differ, creating different tensions in these two zones as the distal joint is flexed; however, for all practical purposes, I treat both zones with the same positioning and motion techniques and hereafter will refer to this injury as a zone III or central slip injury.
Traditional Management of the Zone III Tendon Injury
The literature contains conflicting opinions concerning direct repair and immobilization with K-wire versus conservative management of the open zone III injury. However, most authors recommend conservative treatment of the acute closed injury at this level with uninterrupted immobilization of the PIP joint at 0 degrees for 6 weeks. Open and repaired injuries are mobilized as early as 3 to 4 weeks by some authors, with protective orthotic use between exercise sessions and graded increments in ROM allowed between 3 and 6 weeks; however, most authors recommend continuous positioning for 6 weeks before motion is initiated. Traditional management of this injury often calls for immobilization of the more proximal joints in extension as well as the PIP joint; however, I have never immobilized more than the digit (PIP and DIP joints) unless the zone IV injury is very proximal, approaching zone V, and have not found this to be a problem clinically. With either operative or nonoperative treatment, it is critical that the orthosis position of the PIP joint be at absolute 0 degrees; otherwise, there is some tension at the repair site, possibly creating some gapping, which may result in tendon healing in an elongated position.
The PIP joint can be immobilized with a volar static thermoplastic orthosis with a counterpressure directly over the PIP joint applied with 1-inch Transpore tape ( Fig. 39-11 ) or with a circumferential finger cast ( Fig. 39-12A ). The finger cast is always the treatment of choice with the closed injury, especially if the PIP joint is tight in flexion or if the digit is swollen. The circumferential pressure decreases edema, and the pressures that the cast imposes serve to elongate the tight volar structures. The finger cast is preferable when noncompliance is possible.