Arthroscopic Ankle Arthrodesis

KEITH W. CHAN

RICHARD D. FERKEL

JAMES M. GLICK

Arthrodesis is the procedure of choice for salvaging debilitating conditions of the ankle associated with pain and instability. With a successful fusion, the patient usually can return to work (including heavy labor) and some sports with close to a normal gait. Open arthrodesis of the ankle joint was first described by Albert in 1882.¹ Numerous open surgical procedures have been described for fusion of the tibiotalar joint. In reviewing the literature, postoperative discomfort is often prolonged, and the complication rate may be as high as 60%, with an average rate of pseudarthrosis of about 20% and an infection rate of 5% to 25%.², ³, ⁴, ⁵, ⁶, ⁷ In contrast, Morgan and colleagues⁸ in 1985 reported a 96% rate of successful fusion, with excellent or good functional clinical results in 90% of 101 fusions, followed for an average of 10 years. The complication rate in this series was 6% including pseudarthrosis. His fusion method differed from others in that it included a debridement that maintained the normal bony contour of the talar dome and tibial plafond instead of squaring it off. In addition, the ankle was placed in a neutral position, and transmalleolar cross-screw internal fixation was used.

With the advent of improvements in arthroscopic technique and instrumentation of the ankle, it has become possible to apply open arthrodesis techniques using minimal incisions. The use of invasive or noninvasive distraction has permitted easier access to both the anterior and posterior aspects of the tibiotalar joint to facilitate the arthrodesis procedure. Schneider⁹ was the first to describe an arthroscopic technique for ankle arthrodesis, and Morgan¹⁰ published the first report on this method. Since that time, numerous authors have reported their techniques and results.¹¹, ¹², ¹³, ¹⁴, ¹⁵, ¹⁶, ¹⁷, ¹⁸, ¹⁹, ²⁰

ADVANTAGES AND DISADVANTAGES

The advantages of arthroscopic ankle arthrodesis include reduced morbidity, reduced hospitalization, a rapid fusion rate, better cosmesis, decreased complications, and an optional tourniquet. Disadvantages include a difficult learning curve for the surgeon, the need for expensive arthroscopic equipment, and the inability to correct significant varus, valgus, or rotational problems.

INDICATIONS AND CONTRAINDICATIONS

Indications for arthroscopic ankle arthrodesis include moderate-to-severe, unrelenting pain at the tibiotalar joint that does not respond to conservative measures. Etiology of the pain can include traumatic arthritis, hemophilic arthropathy, congenital deformity, rheumatoid arthritis, old osteochondritis dissecans, and previous ankle infection now eradicated. Relative contraindications include significant varus or valgus malalignment >15° although the surgeon may proceed if he or she feels that the deformity is flexible enough to be correctable. Strict contraindications include coronal plane deformity >45°, malrotation of the ankle, significant bone loss, active infection, previous failed fusion, complex regional pain syndrome, a neuropathic destructive process in the tibiotalar joint, and anteroposterior translation of the tibiotalar joint requiring correction to planar joint surfaces (Fig. 13-1).

FIGURE 13-1. Mortise x-ray of the right ankle demonstrating significant talar tilt and varus angulation. This ankle was not able to be manipulated into a neutral position and should not be fused arthroscopically.

In some cases where the deformity is severe, the authors will take the patient to the operating room to attempt manipulation of the ankle into a neutral position with fluoroscopy while the patient is under anesthesia or milder cases can be attempted in the office. Arthroscopic fusion can then be done at a later setting if the manipulation is successful.

FUSION POSITION AND BIOMECHANICS

The optimal position for arthrodesis of the ankle is neutral for dorsiflexion and plantar flexion.²¹ Equinus, especially more than 10°, should be avoided unless the patient has poliomyelitis, in which case 10° to 15° of equinus helps stabilize the knee joint (Fig. 13-2). Normally, fusions done in more than 10° of equinus produce a significant loss of dorsiflexion that is compensated for by external rotation of the limb and a back-knee thrust into hyperextension that is uncomfortable for the patient.²¹ The calcaneus should be in about 5° of valgus, and the transverse plane rotation should be equivalent to that of the uninvolved side, usually 5° to 7° of external rotation.²¹

Case studies demonstrate that when the ankle is fused in neutral position, patients can walk with good velocity and without unusual movements of the limbs or trunk.²² However, after ankle fusion, most patients show some minor gait irregularities while walking, running, climbing stairs, and particularly walking on inclines.²³ Morgan and associates demonstrated through motion studies that there is an average 58° ± 7° of combined tibiotalar-tarsal motion in normal controls and 18° ± 3° in securely fused ankles, a loss of 70% of the total motion arc with a fused ankle.⁸ In the same study, the authors demonstrated that compensatory tarsal hypermobility was increased 85%.

FIGURE 13-2. Fusion position. (A) In most cases, more than 10° of equinus should be avoided with ankle arthrodesis. (B) The optimal position for ankle arthrodesis is neutral dorsiflexion and plantar flexion. (Illustration by Susan Brust, Copyright Richard D. Ferkel.)

Trouillier et al.²⁴ performed gait analysis on 17 patients who underwent ankle arthrodesis on average 15 years earlier. The gait pattern revealed no significant pathology for the operated foot. However, on analysis of contact pressures, there was an increase in forefoot strain due to a decrease in the push-off phase of gait as compared to the healthy foot. In a standard sport shoe, the fused foot has ground contact as a whole for only a minimum of time (2.7% of overall ground contact time) thus resulting in higher loads in the heel-toe cycle. When adapted orthopedic shoes were worn, gait disturbances were minimized. The authors concluded that a rocker bottom sole as well as shoe lifts that minimize limb length discrepancies are crucial for gait and minimizing stress in adjacent joints following ankle arthrodesis.

Adjacent joint arthritis following ankle arthrodesis is fairly common. However, Sheridan et al.²⁵ reported that 68 out of 71 ankles had preexisting concomitant subtalar or midfoot joint arthritis, thus suggesting that such adjacent joint arthritis was not necessarily a result of tibiotalar arthrodesis. Moreover, development of adjacent joint arthritis did not correlate with the clinical outcome score or quality of life measurements (SF-36) in a study by Fuchs et al.²⁶

OPERATIVE TECHNIQUE

Arthroscopic ankle arthrodesis is performed using the same instrumentation and techniques as described in Chaps. 3 and 7. It is done with or without a tourniquet in a supine position with a thigh support and distraction (usually noninvasive).

Like the open method, the arthroscopic procedure includes three basic steps: (1) removal of all hyaline articular cartilage, avascular subchondral bone, and subchondral cysts that communicate with the joint, (2) fusion reduction in the neutral position, and (3) internal fixation with two (occasionally three) transmalleolar screws.

Setup and Instrumentation

The equipment commonly used includes a 30° and 70° oblique 4-mm or 2.7-mm arthroscope with camera, monitor, and video equipment; high-speed motorized suction shaver and burr; cup curettes, small-joint osteotomes, and small-joint drill guide; fluoroscope; invasive or noninvasive distraction systems; and cannulated compression screws. This is one of the few indications where we feel that the large joint shavers and burrs prevent instrument clogging and expedite the procedure. Our preferred method is to use the 2.7-mm arthroscope and the large 4.5-mm shaver and 4.0-mm ball burr along with the other standard instrumentation.

The patient is positioned supine on a standard operating table. The procedure can be done under general with
a popliteal block or regional anesthesia. There are two methods for patient positioning. In the first, the patient is placed in the supine position on a radiolucent operating table with the knee bent over a well-padded support and the end of the table bent 90° (Fig. 13-3). In the second, our preferred method, the patient is placed in the supine position. The leg is supported with a well-padded thigh support and side post, and the pad is removed from the foot of the bed (Fig. 13-4).

FIGURE 13-3. With this technique, the patient is positioned supine with the knee bent over a padded support and the end of the table bent 90°. (Illustration by Susan Brust.)

Distraction

Distraction can be applied by an invasive or noninvasive device. With the knee flexed over the end of the table, a distraction strap is placed around the foot and ankle, and a hole is cut in the drape so that 20 to 25 pounds of weight can be applied to distract the tibiotalar surfaces (Fig. 13-5). If the patient is in the supine position, a similar strap is applied but is attached to a distraction device that generates a force by pulling against a loop in the strap, which is our preferred method (Fig. 13-4).

If noninvasive distraction does not provide adequate separation of the joint surfaces, an invasive distractor can be applied (see Chap. 3). The invasive distractor can be placed either medially with 5-mm (³/₁₆″) pins in the tibia and talus, as described by Morgan, or laterally with 5-mm (³/₁₆″) pins placed in the tibia and os calcis, as described by Guhl (Fig. 13-6).

The advantages of medially based distraction include easier dorsiflexion and plantar flexion of the ankle while maintaining parallel separation of the joint surfaces rather than lateral talar tilting, which occasionally occurs with laterally based distraction, and easier access to the posterolateral and anterolateral portals. The advantages of lateral pin distraction include avoidance of injury to the calcaneal branches of the posterior tibial nerve, easier access to the anteromedial portal of the ankle, and less difficulty inserting the guide pins for screw insertion.

Pin placement can also be used to correct varus or valgus angulation prior to screw insertion. Over the last 15 years, the authors have not found invasive distraction necessary to achieve adequate visualization of the working area. We utilize noninvasive distraction only.

FIGURE 13-4. Patient positioning can also be done using a thigh support with a noninvasive strap attached to a distraction device (see Chap. 3). (Illustration by Susan Brust.)

FIGURE 13-5. When the patient is positioned as in Figure 13-3, a hole in the drape is made and nonsterile weights are applied to the end of the sterile distraction strap.

Portals

Three arthroscopic portals are used: anterolateral, anteromedial, and posterolateral. Initially, a complete 21-point arthroscopic examination is done and the entire anterior portion of the ankle is visualized. With increased distraction, the arthroscope can be maneuvered through the medial notch of the tibia (notch of Harty) and the posterolateral portal established under direct vision. Fluid inflow can be delivered by gravity or a mechanical pump that functions through the sheath that houses the arthroscope or through a posterolateral cannula. If a pump is used, the amount of fluid extravasation must be monitored at all times to avoid complications. Alternatively, good inflow and outflow can usually be maintained by using a large-bore cannula for fluid inflow through the posterolateral portal, which is the author’s choice.

Intraoperative Imaging

Intraoperative imaging is a crucial component to the case in order to assure proper alignment and positioning of the tibiotalar fusion. As well, it is often useful prior to sterile surgical preparation to image the operative ankle under C-arm fluoroscopy to obtain a baseline image. Ankle alignment can be confirmed, and successful manual correction of any valgus or varus deformity can be assessed under live fluoroscopy at this time (Fig 13-7). If the tibiotalar joint alignment proves resistant to manual correction, it may be necessary to consider more invasive methods to correction,
as described below, during reduction and internal fixation of the tibiotalar arthrodesis or, alternatively, converting the operation to an open arthrodesis.

FIGURE 13-6. Lateral pin distraction is done through the distal tibia and calcaneus. Extreme caution is needed to avoid injury to neurovascular structures. (Illustration by Susan Brust.)

Procedure

The key factor in establishing a complete, rapid union is to remove as little subchondral bone as possible. Before beginning surgery, the preoperative x-rays, MRIs, and CT scans should be studied to assist in surgical planning (Fig. 13-8A-D). Most cases demonstrate at least partial loss of the articular cartilage, osteophytes, and synovitis. The shape of the talus and tibial surfaces should be maintained.

FIGURE 13-7. Fluoroscopic images of the ankle (A) before and (B) after manual correction in the operating room.

Cartilage Removal

Initially, the soft tissues obstructing visualization are removed with the intra-articular shaver. All remaining hyaline cartilage is removed systematically from the
articular surfaces of the talus, tibial plafond, and medial and lateral gutters, thus exposing the subchondral plate in these areas. Handheld ring and cup curettes are best for removing the hyaline cartilage (Fig. 13-9A-C). Periodically, the floating loose pieces of cartilage are removed using a grasper or shaver or by suctioning through the arthroscopic cannula. All cysts are curetted out until good bone is seen.

FIGURE 13-8. Prearthrodesis evaluation. (A) AP x-ray demonstrating significant loss of joint space. (B) Lateral x-ray demonstrating virtually no joint space and loss of articular surfaces. (C) Sagittal image from CT scan demonstrating loss of joint space and a loose anterior osteophyte. (D) Coronal image from CT scan demonstrating cystic changes in the lateral joint.

FIGURE 13-9. Preparing the joint surfaces for fusion. (A) The arthroscope is inserted anteromedially and instruments are inserted anterolaterally. (B) Arthroscopic image demonstrating diffuse loss of articular cartilage. (C) Strong ring curettes are used to remove the remaining diseased articular cartilage. (D) Arthroscopic view demonstrating the use of a curved curette on the distal tibia. (E) The subchondral plate is denuded to a bleeding surface using a motorized burr. (F) Arthroscopic view using 4.0-mm burr in medial gutter of left ankle. (Illustrations by Susan Brust, A, C & E—Copyright Richard D. Ferkel.)

Only gold members can continue reading. Log In or Register to continue