A 24-year-old right-hand-dominant laborer presented with a crushing/twisting injury to his dominant hand after he caught his hand in machinery at a sawmill. Among other joint, bone, and tendon injuries to multiple digits, he sustained an open, extra-articular proximal phalanx fracture of the small finger, which was pinned (▶Fig. 17.1). The pins were pulled 3.5 weeks after the operation, and he began hand therapy. Three months after the operation, he had developed a swan neck deformity (SND) of the small finger. He could bend the digit actively but suffered a painful snapping as the extensor mechanism generated enough tension to force the joint past the fulcrum point from hyperextension to flexion.
The SND was mostly correctable passively. However, the small finger could not be passively flexed fully to the distal palmar crease, indicating some extensor tendon adhesions as well. The finger was neither tender nor swollen. Bunnell’s test was negative for intrinsic tightness. The proximal interphalangeal joint (PIPJ) was hyper-extended to 30 degrees and the distal interphalangeal joint (DIPJ) was flexed to 25 to 30 degrees. He had independent flexor digitorum profundus (FDP) and flexor digitorum superficialis (FDS) function. The difficulty initiating flexion of the small finger and the swan neck posture impeded his grasp and fine motor skills. X-rays indicated a malunion of the proximal phalanx fracture of about 10 degrees apex dorsal angulation (▶Fig. 17.2). The PIPJ and DIPJ were congruent and without arthritis.
To understand the problem, one must determine the etiology of the SND and then choose the best of the available treatment options based on the patient’s needs and disabilities.
The term SND refers to a finger with a posture of hyperextension at the PIPJ and concomitant flexion at the DIPJ. Although the appearance of the finger is the same in all instances of SND, the etiology varies. The functional loss associated with this deformity is related to the loss of PIPJ motion. In some patients, the deformity has only minor clinical consequences, and it is mostly a cosmetic issue. Some may only be bothered by a mechanical snapping into flexion. For others, it can adversely affect function, decrease grip strength, and cause pain.
The extensor mechanism is made up of both extrinsic and intrinsic tendons (▶Fig. 17.3). The extrinsic tendons originating from the extensor digitorum communis (EDC) in the forearm travel over the dorsum of the metacarpophalangeal (MCP) joints into the digits. Over the proximal phalanx, each EDC tendon splits into a central tendon, which inserts onto the base of the middle phalanx, and two lateral slips, which join with the lateral bands to form the conjoined tendon just distal to the PIPJ. The conjoined lateral band continues until its insertion onto the base of the distal phalanx, at which point it is known as the terminal tendon. The intrinsic extensor mechanism has contributions from both the lumbricals and the dorsal interossei. The lumbricals originate from the radial aspect of the corresponding FDP tendons. The deep head of the dorsal interossei passes superficially to the sagittal band and contributes to both the transverse fibers of the extensor hood over the middle aspect of the proximal phalanx (to provide a flexion moment arm to the MCP joint) and to the medial band of the interosseous, which blends with the central slip and helps extend the PIPJ. The transverse retinacular ligaments originate from the flexor tendon sheath and insert onto the palmar aspect of the lateral bands to prevent their dorsal subluxation, while the triangular ligament distally binds the lateral bands into the conjoined lateral band and prevents the slips from subluxing palmarly. The extrinsic extension system normally extends the MCP joint (via the extensor hood volar connections), while the intrinsic system flexes the MCP joint and extends the PIPJ and DIPJ. When both systems work in synergy, the digits have coordinated and controlled extension through all three joints.
SND is caused by an imbalance in this extensor mechanism. That imbalance originates in one, or sometimes more, of three areas: (1) extrinsically, at the level of the forearm, wrist, or MCP joint; (2) intrinsically, from the intrinsic extensor tendon system; or (3) at the articular level of the PIPJ or DIPJ themselves. For example, a wrist or MCP joint flexion contracture increases extrinsic EDC tension. This increased tension is transmitted to the base of the middle phalanx via the central slip insertion and can result in hyperextension of the PIPJ. Alternatively, contracture or spasticity of the intrinsic system can flex the MCP joint and hyperextend the PIPJ (via the lateral bands) resulting in the same deformity. A loss of PIP flexion tone from a disruption to the superficialis tendon (which inserts onto the volar base of the middle phalanx) can have the same result though by creating an intrinsic extension/extrinsic flexion force imbalance that allows the PIPJ to hyperextend. Finally, proximal migration of the extensor mechanism due to terminal tendon disruption at the DIPJ (mallet finger) can increase extensor force to the central slip and hyperextend the PIPJ by yet a different mechanism, while laxity (or stretching) of the volar plate of the PIPJ can independently contribute to the SND.
The diagnosis of SND is based just on physical appearance of the digit and functional complaints. The pathophysiology of the tendon imbalance, the presence of arthritis in the joints, and the quality of the soft-tissue envelope and tendency for tendon adhesions must all be considered in designing a successful treatment strategy. A history of known disease states (e.g., rheumatoid arthritis, cerebral palsy) or trauma (e.g., mallet deformity, volar plate avulsion fracture) can direct the workup, and radiographic evaluation of the joints should reveal degenerative and posttraumatic joint conditions.
The passive and active range of motion of the wrist, MCP, and IP joints should be evaluated both with the hand in its resting posture and with the SND passively corrected. This can allow the identification of a chronic mallet injury from either traumatic disruption or gradual attenuation of the terminal slip insertion. Extrinsic tightness can be assessed by evaluating changes in extrinsic extensor tendon tension related to wrist position. Extension of the wrist would allow more finger flexion in the presence of extrinsic tightness. Intrinsic tightness, on the other hand, can be determined utilizing the Bunnell test. The Bunnell test is performed by holding the MCP joint in extension and then in flexion while actively or passively flexing the PIPJ. Decreased passive PIPJ flexion with the MCP joint in extension is compatible with intrinsic tightness (lateral bands run volar to MCP joints and dorsal to the IP joints).
The SND can be classified into one of four types as devised by Nalebuff and Millender:
• Type I: full range of motion of the joints with no significant functional limitations.
• Type II: intrinsic tightness as shown by a positive Bunnel test; the PIPJ can be ranged fully with MCP joint held in flexion only.
• Type III: the PIPJ is stiff both actively and passively irrespective of the position of the MCP joint.
• Type IV: the same as type III, but with radiographic arthritic changes at the PIPJ.
For a type I deformity, only the PIPJ hyperextension needs to be addressed. This can be done by decreasing the amount of skin volarly with a dermodesis or by creating a mechanical restriction to extension with an FDS sling, an oblique retinacular ligament (ORL) reconstruction, or a lateral band rerouting procedure. Excising and closing an ellipse of skin from the volar aspect of the PIPJ is only helpful for very mild deformity (and can stretch out with time). A stronger restraint to hyperextension can be achieved using the FDS to create a restraint sling known as the “sublimis sling.” This was performed on our patient, and is described later under the section “Technique.” These strategies only address the PIP hyperextension and not the DIPJ extension lag. If this appears to be a persistent problem even with the PIP passively corrected than one of the following reconstructions can be used. In the ORL reconstruction, the ulnar lateral band is freed from the extensor mechanism proximally at the level of the MCP joint, passed volar to Cleland’s ligament and volar to the PIPJ axis of rotation, and sutured radially either to the flexor tendon sheath or passed through the proximal phalanx via a bone tunnel. Alternatively, a palmaris free graft can be used, sutured to the terminal tendon, then following the same path as the ulnar lateral band as described earlier. Finally, one or both lateral bands are freed from their dorsal attachments, translocated palmarly, then either sutured to the ipsilateral FDS slip and volar plate or placed into a flap made in the flexor tendon sheath at the PIPJ level.
For a type II deformity, the same procedures as in type I are available for the PIPJ deformity, but the MCP joint needs to be addressed as well. Often, this is done with just an intrinsic release. However, any MCP joint or wrist disorder that aggravates the imbalance of the finger such as subluxation or deviation must be corrected as well.
For a type III deformity, in addition to the intrinsic release, the PIPJ needs to be released. There is often a paucity of skin dorsally that may need to be addressed. A formal joint release with transection of the collateral ligaments, volar translocation of the lateral bands, and possible step-cut lengthening of the central slip may be necessary. Also, due to the prolonged lack of range of motion, there may be flexor tendon adhesions, which need to be freed simultaneously. This approach is obviously more complex and the results tend to be less predictable.
For a type IV deformity, the PIPJ needs to be addressed most commonly with an arthrodesis.
The Presented Case
Our patient had a type 1 SND, complicated by the proximal phalanx malunion and extensor tendon adhesions. His biggest problems were initiating flexion and keeping his small finger from catching on objects in its resting swan posture. Dermodesis was unlikely to be sufficient or lasting in a young laborer. This left the superficialis sling surgery, the lateral band translocation procedure, and the ORL reconstruction. The final option also helps correct DIP flexion deformity. His DIP flexion deformity was not severe enough to warrant correction especially in the small finger and because of the crush nature of his initial injury, there were concerns about the propensity for adhesions and stiffness with a more complex procedure.
The patient was taken to the operating room after administration of regional anesthesia and placed in the supine position on the operating table. An upper extremity tourniquet was inflated to 250 mm Hg. Wide-awake surgery local anesthesia without tourniquet (WALANT) is also an option and is normally the author’s preference as it allows for visualization of the range of motion in real time and allows for any necessary modifications at the time of surgery. However, this was not possible in this case due to the necessity of performing several concomitant procedures.
There are multiple approaches to exposing the flexor tendon sheath, and many variations in technique for creating the sling (e.g., using one or both slips of FDS, securing the tendon through the A1 or the A2 pulley or to the proximal phalanx).
In this case, a Brunner incision was made from the level of the A1 pulley to the level the A3 pulley. Skin flaps were created and held to the side with 4–0 silk retention sutures. A beaver blade and freer were slid under the extensor mechanism from both the radial and ulnar sides, freeing the extensor mechanism from the proximal phalanx where it was adherent in the region of the fracture. The A3 pulley was then incised, exposing the distal end of the flexor superficialis tendon toward its insertion at the base of the middle phalanx (▶Fig. 17.4a). To make sure there were no flexor tendon adhesions, an Allis was placed around the profundus and superficialis tendons proximal to the A1 pulley and used to gently twirl the tendons around it. This simulated active flexion of the tendons. Each was tested independently. There was a bit of catching of the profundus and a freer was placed between the profundus and the bone, releasing the adhesion.