ARTHROSCOPY OF THE ANKLE
July 5th, 1995
Mark S. Myerson, MD
Introduction
Over the past decade, our understanding of the pathophysiology of disorders of the ankle has expanded. Intraarticular anatomy and soft-tissue, chondral, and bony defects of the tibia and talus have been clarified. Substantial advances in ankle arthroscopy have evolved through improved instrumentation and the use of mechanical distraction, without which, the field of vision is limited to the anterior third of the joint. Distraction has enabled us to perform a global evaluation of the joint, including the anterior and posterior compartments, without causing scuffing and damage to the articular surfaces. Our understanding of soft-tissue pathology and its application to functional instability of the ankle is increasing. Newer innovative techniques, such as arthroscopic ankle arthrodesis, combine the advantages inherent in arthroscopy with current methodologies of fracture fixation. Arthroscopic examination of the ankle joint allows us to directly examine the articular cartilage and soft tissues, all of which are discussed below.
Arthroscopic Anatomy
Integral to the success of ankle arthroscopy is a thorough understanding of the surface and intraarticular anatomy. The surface landmarks of the ankle are demarcated by the malleoli. The tip of the fibula is approximately 2 cm distal and posterior to the medial malleolus. The joint line is 2.5 cm proximal to the tip of the fibula and approximately 1 cm proximal to the medial edge of the medial malleolus. The surface structures from medial to lateral are the thick anterior tibial tendon medially and the extensor hallucis and extensor digitorum communis tendons lying centrally over the joint. More laterally, the peroneus tertius tendon runs obliquely across the ankle to its insertion on the base of the fifth metatarsal. These tendons should be identified and marked before commencing the procedure.
There is minimal subcutaneous tissue separating the joint cavity from the skin. Neurovascular structures are therefore at particular risk, but are easily palpable and must be identified preoperatively. The deep peroneal nerve and dorsalis pedis artery lie in the sulcus between the extensor digitorum and extensor hallucis tendons and deep to the extensor retinaculum. This deeper location is vulnerable when performing a synovectomy as little separates the neurovascular bundle from the capsule. The superficial peroneal nerve bifurcates 6 cm proximal to the level of the fibula into the dorsal medial and the dorsal intermediate cutaneous nerves (Adkison et al 1991). The dorsal intermediate cutaneous nerve is more lateral, lying immediately lateral to the peroneus tertius tendon, and is in a more vulnerable location. The dorsal intermediate cutaneous nerve can be easily identified preoperatively, particularly in thin patients by plantar flexing the fourth toe. The nerve becomes prominent, and its location should be marked.
The posterolateral aspect of the ankle is contained by the fibula and the peroneal tendon immediately posterior and adjacent to it. The sural nerve lies almost midway between the peroneal and Achilles tendons and should be avoided with the posterolateral portal.
The anterior articular surface of the joint is not horizontal, and undulations along its surface provide the basis for the anterior portals. Just lateral to the medial malleolus, the articular surface is recessed by approximately 4 mm, creating a notch forming the anteromedial portal between the medial malleolus and the anterior tibial tendon. There is a slightly rounded distal projection from the tibia centrally, where it juts out to correspond to a similar concavity on the dorsal surface of the talus. It is not easy, therefore, to sweep the arthroscope under the tibia across the joint from side to side, which has to be performed anterior to the tibia or across the talar neck. The fibula is situated 1.5 to 2 cm posterior to the anterior articular edge of the tibia. When viewing the lateral aspect of the joint from the anteromedial portal, the fibula is therefore not visible. To see the anterior aspect of the fibula, the arthroscope is gently introduced through the anteromedial portal and then carefully swept across the tibiotalar surface.
The posterior articular surface of the tibia lies more distally than its anterior edge. This distal projection is increased by the condensation of soft-tissue structures posteriorly, including the transverse ligament, which is adherent to the inferior tibiofibular ligament and easily visible with mechanical distraction. The ligament is covered with synovium; with dorsiflexion of the ankle, it remains attached to the tibia and, as the talus moves distally, a synovial-lined recess is exposed. This recess should also be carefully examined, particularly for loose bodies. An additional posterior structure that is occasionally seen from an anterior portal is the tibial slip, which extends from the edge of the posterior talofibular ligament obliquely to the medial edge of the transverse ligament.
The anterior talofibular ligament (ATFL) should be routinely inspected to differentiate it from pathological conditions, as this is a frequent site for soft-tissue lesions. The ATFL is partly intracapsular and is intraarticular. When torn, the hemorrhagic cellular exudate occasionally does not resolve. During the healing process, the inflammatory tissue hypertrophies and forms a variable fibrous mass over the anterior talofibular recess. When debriding this hypertrophic tissue, care must be taken not to resect too deeply, as the remnant of the ligament is likely to be debrided. Similarly, on the medial aspect of the joint, the deep portion of the deltoid ligament is easily visible and may also be involved in a pathologic inflammatory process. The tibiofibular joint is synovial lined, and proliferative synovitis, which occurs in inflammatory, degenerative, and posttraumatic conditions, commonly affects this joint. The transverse tibiofibular ligament and posterior talofibular ligament are extraarticular and not normally seen arthroscopically. However, the posterior talofibular ligament may become hypertrophied as a result of trauma, in which case it becomes visible as a posterior bulge. Patients with this condition present with posterior ankle pain; the pathologic tissue is amenable to resection.
Almost the entire visible surface of the talus is covered by articular cartilage. Most of the dome of the talus is visible through either the anterior portals. Approximately 50% of the dome is visible without distraction; the amount visualized can be increased to about 80% with distraction. If further visualization posteriorly is required from an anterior portal, then the 70o arthroscope can be used. Anteriorly, the neck of the talus is visible from either the anteromedial or anterolateral portals. The talar neck is not often involved in a pathologic process; however, it is commonly covered with loose or adherent osteophytes in posttraumatic degenerative conditions. The arthroscopic management of these degenerative arthritides may be difficult, as many of these osteophytes are intracapsular and not intraarticular.
Arthroscopic Technique and Positioning
Anesthesia
Although general anesthesia is preferred, local anesthesia may be used occasionally for arthroscopic examination of the joint, provided one is certain that only anterior joint pathology is present. However, general anesthesia is preferred for a thorough evaluation of a joint that may require mechanical distraction. If local anesthesia is used, 10 to 15 cc of a mixture of l% xylocaine and 0.5% bupivicaine with admixed epinephrine is instilled into the joint and 2 cc are instilled into each portal 30 minutes before commencing the procedure.
Patient Positioning
The patient is placed in a supine position with the hip and knee flexed and the ankle supported with a sterile holder, depending on the method of distraction used. With the patient in the supine position and the ankle free, the anterior portals are easily accessible. If the posterolateral portal is used, the limb must be internally rotated or a sterile ankle holder used. A tourniquet is always applied to the thigh, although it is not always inflated.
Initial Procedures
The joint is inflated first with 10 to 15 cc of saline or the local anesthetic mixture. I use small incisions for the portals with a #11 blade to puncture the skin only, and a hemostat is then used to longitudinally spread the subcutaneous tissues. This moves the subcutaneous structures, such as tendons and nerves, out of the way and also minimizes the portal size, which prevents fluid extravasation. A blunt obturator is then used to enter the joint and is twisted around in a circular motion to free up the tissues. The arthroscope is then inserted.
It is important in ankle arthroscopy to maintain high inflow and outflow, which I accomplish with gravity, using continuous overhead saline inflow. An arthroscopy pump may be used to maintain high volume and pressure, but it can be quite dangerous in the ankle due to fluid extravasation, and I do not recommend its use.
Portals
Before induction of anesthesia, all of the anterior and posterior soft-tissue structures are carefully marked out. The anteromedial portal is located in the sulcus in between the medial malleolus and the anterior tibial tendon and is confirmed by insertion of a #22 needle. The anterolateral portal is located immediately lateral to the tendon of the peroneus tertius and medial to the dorsal intermediate cutaneous branch of the superficial peroneal nerve at the level of the joint. Although this may be located before insertion of the arthroscope, it is easier to find by turning the arthroscope toward the skin so that the light source highlights the subcutaneous soft tissues.
I use the anteromedial portal for insertion of the arthroscope, preferably directly over the joint line. If it is more than 1 cm inferior or below the joint line, difficulty will be experienced in visualizing the lateral aspect of the joint, particularly more posteriorly. If this portal is made directly over the joint line, there is enough room inferiorly for an accessory anteromedial portal. The joint is quickly inspected anteriorly and the anterolateral portal is selected as described above.
The anterocentral and posteromedial portals are fraught with potential complications, and are not recommended under any circumstances. The posterolateral portal is particularly useful for fluid inflow, and is also used to evaluate posteromedial lesions of the talar dome. With the arthroscope anteromedially, the portal is located with a #22 needle, 5 mm medial to the peroneal tendon and l cm proximal to the tip of the fibula. The posterior articular margin of the tibia is 5 mm lower than the anterior margin, and this overhang must be taken into consideration with this portal.
Distraction
If improved visualization is required, then some form of distraction of the joint is necessary and may be accomplished through invasive or noninvasive means. The simplest form of noninvasive distraction is through the use of a sterile stocking bandage wrapped around the midfoot and then hitched onto the surgeon's foot to apply distraction. A commercially available noninvasive soft-tissue distractor is available and may be the preferred method for distraction.
Mechanical distraction may be performed invasively using a calcaneal and tibial pin. The pins for the latter system are threaded and are inserted after predrilling with a 3.2- or 3.5-mm drill bit. Alternatively, a threaded trochar tipped Steinmann pin may be used. The tibial pin is inserted 6 cm proximal to the joint line from lateral to medial, slightly posterior to the anterior tibial crest. The anterior tibial muscle is pushed posteriorly with a clamp or trochar and the pin is inserted using a power tool. Although it has been recommended that only one cortex be used, I have inserted the pin across both cortices as this method inhibits the angulation tendency of the pins. A #22 needle is used to locate he inferior surface of the calcaneus, and then the inferior pin is inserted 1 to 2 cm proximal to the its inferior edge, engaging both the lateral and medial cortices of the calcaneus. It is important to angulate the distal pin approximately 20 to 30o caudally (toward the tibial pin) to prevent varus angulation of the ankle joint during distraction. The distraction should be initiated slowly and should never exceed 50 pounds of force (as read on the distractor). During the course of the arthroscopic distraction, some relaxation of the ligaments occur due to inherent elastic deformation; in this case, more distraction can be applied. The gradual application of distraction is particularly important in patients with arthrofibrosis.
Most authors recommend insertion of these pins laterally in the tibia and calcaneus. However, there is a tendency when using this unilaterally applied distractor for the ankle to tilt into varus, which may decrease visualization medially. If this occurs, the fixator pins may be inserted medially in the tibia and talus directly inferior to the medial malleolus. This is particularly relevant when attempting to visualize the medial and posterior surfaces of the talus.
Osteochondral Lesions of the Talus
Etiology and Diagnosis
The terminology for osteochondral defects of the talus can be quite confusing, and whether this represents a traumatic or idiopathic etiology, the generic term of transcondylar talar dome lesions is applicable (Baker et al 1986; Berndt & Harty 1959; Flick & Gould 1985; Guhl et al 1993; O'Farrell & Costello 1982; Ferkel & Scranton, Jr. 1993). Whether these lesions are idiopathic in nature and represent a pathologic fracture of necrotic bone as a result of ischemia or whether they are traumatic in etiology, it is my preference to use the term osteochondral lesions of the talus (OLT) to describe these lesions. The etiology of the transchondral lesion is certainly controversial, yet most authors currently recognize the predominant effect of trauma in producing these lesions. The medial OLT is typically produced by an inversion and plantar flexion injury with external rotation of the tibia on the talus. Lateral lesions are caused by inversion with simultaneous dorsiflexion of the ankle.
Berndt and Harty (1959) classified these fractures into four stages based upon the radiographic appearance: stage 1, a small subchondral compression fracture; stage 2, a partially detached osteochondral fragment; stage 3, a completely detached osteochondral fracture without displacement; and stage 4, a displaced osteochondral fracture. Although this classification scheme is still used today, it is outdated. Plain radiographs may underestimate the extent of the defect and, particularly in stage 1 and stage 2 defects, they do not adequately demonstrate discontinuity. Most surgeons currently use a classification system based on a computed tomography (CT) scan or the arthroscopic appearance of the lesion. With an arthroscopic staging system, irregularities and subtle defects of the chondral surface are clearly identified and can be appropriately treated.
Surgical Treatment
Any patient with a symptomatic OLT should be treated surgically, regardless of the stage. However, since stage 1 lesions are rarely symptomatic and are theoretically quiescent, they should probably be treated initially with physical therapy and rehabilitative modalities.
One may be able to treat a stage 2 lesion with restriction of activity and physical therapy, but these patients rarely obtain permanent relief. If these patients are prepared to modify their activities, it is possible that, with time, the chondral surface may stabilize. Most patients, however, will experience repetitive bouts of discomfort with exercise over a period of years until the lesion is treated surgically. Given the propensity for further softening and fibrillation of the chondral surface with potential for loose body formation, these symptomatic lesions should therefore be treated surgically.
These patients will describe a variety of symptoms (including aching, a sensation of clicking or snapping, or pain and swelling of the ankle after exercise), none of which seem to be specific. A common presentation is an ankle sprain that does not get better with time or therapy. Some athletes can manage their regular day-to-day activities but become symptomatic after exercise.
The defect is usually visible on routine radiographs. The medial lesions are typically posterior and the lateral lesions are anterior; these locations appear to be consistent with the proposed pathogenesis. In addition to these radiographs, I further clarify the position of the OLT with lateral radiographs in maximum plantarflexion and dorsiflexion. The OLT is then further defined with CT. CT (but not magnetic resonance imaging) is useful to plan the extent of surgery since it defines precisely the exact size and depth of the lesion as well as the location of any loose bodies.
The lateral lesions are usually anterior on the talus and are accessible through an arthrotomy. However, the medial lesion is typically more posterior and can seldom be reached through an anteromedial arthrotomy. Some clinicians have therefore advocated an osteotomy of the medial malleolus, whereas others have described grooving of the anterior surface of the tibia over the lesion to satisfactorily expose the talus. These approaches involve substantial soft-tissue and bony trauma, with the added risk of malunion or nonunion of the medial malleolus. Arthroscopy is currently the surgical method of choice for approaching OLT.
The arthroscopic procedure is performed under general anesthesia. Mechanical distraction is used to improve visualization, although anterolateral lesions of the talus are more readily accessible and distraction may not always be required. I am able to determine the need for distraction preoperatively with accurate localization of the lesion and, in particular, the accessibility of the talus with the foot in maximum plantar flexion.
After a synovectomy, the arthroscope is reintroduced into the anterolateral portal for access to the medial lesion. A thorough inspection of the joint has to be made, particularly for evidence of loose bodies. The cartilage is carefully probed to determine softening and to accurately stage the lesion. The posterior compartment is best inspected through a posterolateral portal. The OLT is debrided with an angled curette, a ring curette, grasper, and a banana knife. The OLT often fragments and cannot be excised in one piece; therefore, care must be taken to avoid leaving small fragments behind. All degenerated cartilage and necrotic subchondral bone has to be removed. The final steps in the procedure are to prepare the bed of the defect by a combination of drilling and curettage. Before drilling the crater, the hard necrotic subchondral bone has to be removed. This is best achieved with a straight and angled curette, depending on the size and location of the lesion.
For the anterolateral lesions, I use a #20 spinal needle to ensure that the pin enters the talus perpendicular to the OLT. The needle is inserted along the anterior edge of the tibia with the foot in maximum plantar flexion. The needle is removed and exchanged for a 0.062-inch diameter Kirschner wire, which is then used to drill four to six holes, depending on the size of the OLT.
It is more difficult to reach the medial lesions. The posteromedial lesions are best visualized through a posterolateral portal, and the instruments are inserted anteriorly with drilling through the medial malleolus. The transmalleolar approach is particularly useful in posteromedial OLT. Although the talar defect can be visualized from the anteromedial or posterolateral portals, there is no room for passage of the operating instruments from either of the anterior portals. With distraction, the anterolateral or posterolateral portal for visualization and the anteromedial portal for the operating instruments is occasionally sufficient. The initial debridement and removal of the defect may be accomplished with an angled curette introduced from the anteromedial portal but, if the lesion is in the anterior third of the talus, the final preparation of the crater with drilling can be satisfactorily accomplished only from the medial portal. This approach is contraindicated in a child with open epiphyses. I use a 0.062-inch diameter Kirschner Wire introduced from the anteromedial cortex of the distal tibia. Although blind triangulation with the transmalleolar pin is difficult, this has been my preferred method. More recently, a mini-drill guide (Micro Vector) has been developed with a retractable tip and offset guide that precisely positions the drill pin over the talar lesion.
Without the drill guide, once in position with the pin, movement of the ankle will allow one to insert three or four holes in one plane. If the lesion is large, the drill has to be repositioned for another set of holes.
Anterior Soft Tissue Pathology
Soft-tissue pathology in the ankle is generally synovial in origin and includes various inflammatory synovitides and the more common forms of posttraumatic synovitis (Bassett et al 1990; Drez, Jr. et al 1982; Ferkel et al 1991; Guhl 1986; Martin et al 1989; Wolin et al 1950). Although rheumatoid arthritis frequently involves the ankle, it has been my experience that arthroscopic synovectomy of the ankle in these patients has a limited role. Ideally, a synovectomy should be performed before the development of radiographic changes of joint space narrowing. However, by the time the patients are symptomatic, these changes have already occurred, and the beneficial effect of the synovectomy is minimal. As with other joints, it is almost impossible to perform a complete synovectomy of the ankle, even using distraction. Nevertheless, there are patients who present early enough in the course of the disease who may benefit from a synovectomy.
Synovial chondromatosis and pigmented villonodular synovitis are two other conditions that are amenable to arthroscopic debridement.
Synovial osteochondromatosis has a variable presentation. The condition evolves slowly over a period of years, and the diagnosis and treatment differ for each phase of its development. During stage 1, there is involvement only of synovium, and the exact diagnosis can only be made arthroscopically.
Pigmented villonodular synovitis is quite nonspecific and should be treated by as total a synovectomy as possible. If synovial thickening is identified posteriorly, then the distractor should be used to obtain adequate exposure for a more extensive debridement. The diagnosis is confirmed if chondral elements are found among areas of acute and chronic synovitis. During the second stage of activity, multiple chondral bodies form within the synovium, may break off into the joint, and may be diagnosed on arthrography or magnetic resonance imaging. The inflammatory synovitis becomes quiescent during the third stage, but many, even hundreds of osteochondral fragments remain in both intraarticular and intrasynovial locations. Since a complete synovectomy cannot be performed, the recurrence rate is high. During the stage 2, there are literally hundreds of these small and relatively soft fragments in the joint that are difficult to retrieve. During stage 3, the small intraarticular osteochondral fragments are easily removed; however, there are many intracapsular fragments, and some form an osseous union with the adjacent talus, usually necessitating an arthrotomy to complete the debridement.
Localized forms of synovitis occur more commonly. These usually are located anterolaterally and follow an episode of trauma, typically an ankle sprain. After an ankle sprain or even a minimally displaced fracture, a hemarthrosis develops and is followed by a cellular reaction around the ATFL. Instead of this resolving, a proliferative fibrosis and finally synovitis and adhesion formation occurs. These lesions are usually anterolateral and correspond to tears of the ATFL. Although less common, chronic posttraumatic anteromedial synovitis from a tear of the deep portion of the deltoid ligament may also occur.
The symptoms are all nonspecific. However, patients will describe aching discomfort, a sensation of fullness, pressure build-up, and swelling. Occasionally, patients report a clicking, particularly with twisting motions. Athletes are unable to perform cutting motions. Marked local tenderness and focal synovitis is present. Pain is exacerbated by plantarflexion inversion movements for lateral lesions and dorsiflexion eversion for the medial lesions. Localized tenderness is present anterior to the lateral malleolus directly over the joint line and is associated with palpable synovitis. There is no ankle instability, but the term "functional ankle instability," anterior capsular impingement syndrome, or, more commonly, anterolateral impingement syndrome of the ankle is used to describe the condition. Wolin et al (1950) were the first to describe a thickened band of tissue between the fibula and the talus that occurred after trauma, calling it "meniscoid." Although the tissue differs pathologically from a true meniscus, the hyalinized connective tissue from the talofibular joint capsule is probably a more advanced histologic form of the more common posttraumatic anterolateral synovitis. If symptomatic, I treat this posttraumatic focal synovitis with renewed physical therapy, emphasizing isokinetic exercise, diathermy ultrasound, and nonsteroidal antiinflammatory medication. If these fail, I use a local cortisone injection, which is effective in approximately 50% of these patients. Arthroscopy used for those who remain symptomatic.
If one is certain of the extent of the anterior pathology, the arthroscopy may be performed under local anesthesia. This is followed by instillation of 10 cc of a local anesthetic. An initial evaluation is performed with the arthroscope anteromedial and the operating instruments anterolateral. The pathologic tissue is consistently found in the talofibular recess in the region of the ATFL. Debridement is performed with a motorized 3.5-mm full-radius resector shaver, followed by use of a basket punch forceps to abrade the lesion. Care should be taken to avoid debriding any remaining functional ATFL.
There appears to be histologic gradation of pathologic tissue, particularly in the anterolateral ankle, from flimsy adhesions to tougher fibrous tissue, and finally to the so-called meniscoid lesion, which by all accounts appears to be fibrocartilaginous tissue similar to the meniscus of the knee joint. These focal lesions are generally anterior. Occasionally, the entire anterior and anterolateral joint is filled with dense adhesions and not much can be seen. Typically, I commence the debridement with a motorized shaver and, as debridement progresses, more of the joint surfaces can be seen. I have found the 3.5-mm full-radius resector shaver to be useful for the initial debridement. A basket forceps and rongeur are then used to complete the debridement.
Posterior Soft-Tissue Lesions
The incidence of posterior soft-tissue impingement is not as high as that anterolaterally. Focal synovitis and adhesions occur, but additional pathology may be associated with hypertrophy of the transverse tibiofibular ligament or a posterior meniscus of the ankle. The transverse tibiofibular ligament was described by Chen and Ikeuchi. The ligament extends from its origin on the distal fibula adjacent to the posterior talofibular ligament. It extends around the posterior margin of the joint to the edge of the medial malleolus. It lies directly below the posterior tibiofibular ligament. Before the use of mechanical distraction, lesions of the transverse tibiofibular ligament were not commonly recognized. However, over the past few years since I have been using distraction more frequently, it has been easy to recognize the normal and pathologic forms of this ligament. Hypertrophy of the ligament with or without fibrillation and synovitis may occur. A true posterior meniscus is rare and has been previously documented only in marsupials.
Posterolateral impingement is common in dancers, particularly during pointe maneuvers, who present in a similar manner to other forms of posterior impingement, but not associated with a bony block to plantarflexion, such as an os trigonum. Ankle pain is diffuse, and although vague, is limited to the posterior aspect of the joint. This is worsened with exercise and stair climbing, and descent is particularly difficult. Dorsiflexion of the foot is painless. Clinically, it is difficult to differentiate the lesion from other posterior ankle pathology, but it is anterior to the more common retrocalcaneal bursitis. Tenderness is located slightly laterally, which differentiates it from the lesion of the os trigonum that occurs on the medial aspect of the talus. These patients are best treated by conservative modalities, much as those described for the anterior lesions. Achilles tendon stretching is particularly useful. If symptoms persist after physical therapy treatments, the posterior ankle is injected with cortisone and patients are placed in a short leg walking cast with the foot positioned in 5o of dorsiflexion. Magnetic resonance imaging is particularly useful to differentiate posterior ankle pathology and, in rare circumstances, has demonstrated a meniscus.
Arthroscopic Ankle Arthrodesis
As an alternative to more traditional techniques of ankle fusion, arthroscopic arthrodesis is an innovative approach that combines current technologies of arthroscopy with compression fixation of the joint. It is likely that, with improvements in instrumentation and technique, the arthroscopic procedure will become more appealing. The indications for arthroscopic arthrodesis are the same as for conventional fusion and include debilitating arthritis of the joint refractory to conservative treatment modalities. The arthroscopic technique can be used in most patients, provided both sagittal and horizontal plane alignment of the tibiotalar joint is normal and no substantial bone loss has occurred.
I have previously reported on the arthroscopic approach to ankle arthrodesis with a significantly more rapid rate of arthrodesis when compared to a traditional open method (Myerson & Allon 1989; Myerson & Quill 1991). I attributed this to decreased periosteal stripping and devascularization compared to open arthrodesis. The rapid and high rate of fusion makes this approach an appealing alternative, particularly where the need for early ambulation is imperative. Patients with vascular disease, dermatologic disorders, and those with other general medical contraindications to a major surgical procedure would also benefit from the arthroscopic arthrodesis.
The arthroscopic procedure is tedious. Despite an improving learning curve, it still takes almost twice as long to perform than an open arthrodesis. Adequate visualization of the joint is sometimes difficult. Because of the narrow field of operative exposure, arthritis visualization in the setting of severe posttraumatic may be difficult. I routinely use distraction and recommend intermittent fluoroscopic imaging throughout the procedure. Anterolateral and anteromedial portals are used.
Distraction should be performed cautiously due to severe scarring, contracture, and arthrofibrosis, and the distraction pins should not be allowed to bend. During the course of the debridement, some soft-tissue laxity occurs, and more distraction can be applied. Once debridement commences, the available space increases, further facilitating full posterior access and visualization. The initial debridement and synovectomy is performed with a 4.5-mm shaver carefully debriding the talofibular and medial malleolar recess. Curettage and burring are alternately used to denude the articular surfaces, followed by a high-speed cannulated burr, which results in smooth bleeding bone surfaces. The contour of the ankle should be preserved, and squaring off of the joint should be avoided. A chisel is not generally necessary. Optimally, the talus should fit snugly into the rounded off mortise.
The foot is then positioned in neutral dorsiflexion, 5o of valgus and 5o of external rotation, and guide pins from a cannulated large screw system are inserted fluoroscopically. The distractor is then dismantled, and the screws inserted. The first screw is inserted through the posterior tibial cortex into the talus anteromedially, and the second screw is inserted from the medial malleolus into the lateral talus, just anterior to the posterior facet. More recently, I have used a third screw introduced from the fibula directed medially into the talus. Although this screw does not provide compression, it controls rotation.
Postoperative Care
I routinely suture the portal punctures to prevent egress of any remaining inflow fluid, which may result in a synovial fistula. A compression bandage is applied; it may be removed and reapplied after three days. Swelling is a serious problem if the foot is left in a dependent position, and limb elevation and non-weight-bearing on crutches should be enforced until asymptomatic. During this period of time, patients are encouraged to elevate the extremity and perform range of motion exercises as soon as comfort permits. Generally, patients are non-weight-bearing for approximately 10 to 14 days. After the more extensive debridements for OLT, patients are kept non-weight-bearing for approximately four weeks. I am not certain that this makes a difference, but the prolonged period of non-weight-bearing may enhance the fibrocartilage "filling in" of the defect. This depends on the size of the lesion, and smaller shallow defects probably do just as well with earlier weight-bearing. Occasionally, a postoperative hemarthrosis may occur that is extremely painful; it may be aspirated.
Complications of Ankle Arthroscopy
Complications of ankle arthroscopy are relatively rare and are generally minor problems that resolve in time (Barber et al 1990; Drez, Jr. et al 1982; Ferkel et al 1993). True infection is also rare; when it occurs, it is superficial and limited to the portals. I have encountered mild skin problems on occasion, but these all appeared superficial and resolved with local wound care with peroxide or Betadine dressing changes and five days of oral antibiotic therapy. Sinus formation has been reported by Guhl (1986) with four cases in his first series of 69 patients. This was related to problems with triangulation when he used accessory portals that were too close to each other. One of the problems peculiar to the ankle is that there is minimal subcutaneous tissue and, with repeated passage of instruments through the portals, the tissues are prone to break down. This has occurred when switching frequently from medial to lateral portals and should be minimized by using small plastic canulas for the arthroscope and instruments. I have experienced this problem on occasion when performing arthroscopic ankle arthrodesis or other procedures that involve extensive manipulation of instruments. It is probably caused by the size of the instruments as I use the large (4.8- and 5.5-mm) motorized shavers. None of these patients developed chronic drainage nor infection of these sinuses, as has been described by Guhl (1986). In my experience, attention to wound care with frequent dressing changes resolved this problem within a few days.
Neurovascular complications have been described. These should really be avoided with careful placement of portals. Before the procedure, the structures over the anterior ankle are carefully marked out. Particular attention should be paid to palpation of the deep peroneal nerve, dorsalis pedis artery, saphenous vein and nerve, and the branches of the superficial peroneal nerve. Although this works well medially and anteriorly, it is sometimes difficult to palpate the superficial peroneal nerve. In thin ankles, this nerve is easily palpated and can be rolled under the fingers. However, in obese patients or those with swollen ankles, this nerve has to be identified indirectly by the superficial skin and bone markings, or by its presumed course adjacent to the extensor digitorum and peroneus tertius tendons.
When introducing the arthroscope or instruments, a very small skin incision is made with a #11 blade. The tissues are spread longitudinally with a small hemostat, and a blunt probe is then used to enter the joint. Under no circumstances should the knife blade be introduced too deeply. Not only can this cause inadvertent laceration of nerves, but the portal created by this technique is too large, and fluid egress may become a problem.
The arthroscope is introduced initially anteromedially and then turned anteriorly to visualize the soft-tissue structures and skin anterolaterally before creating the anterolateral portal. This avoids puncturing the lesser saphenous vein and possibly the superficial peroneal nerve. Neuromas of the deep and superficial peroneal nerve may be permanent. I have seen two unfortunate patients who now have permanent disability from injury to the deep peroneal nerve as a result of imprecise portal placement. One patient has a severe neuritis of the deep peroneal nerve that triggered a reflex sympathetic dystrophy, which has not resolved with treatment. From the other patient's description, it would appear that a motorized shaver was used to perform an anterior synovectomy that was carried too superficially, debriding the entire deep peroneal nerve in the process. There is a tendency to continue debridement with the shaver at times knowing that you are in a "safe" area. I make it a rule to always keep the tips of the shavers and operating instruments in full view while performing the debridement. Doing so will not only avoid inadvertent damage to soft-tissue structures, it will also prevent articular cartilage scuffing. The latter problem is difficult to quantify, but is a very real problem. This is a complication that should no longer occur now that satisfactory modes of distraction are available.
Occasional massive swelling intraoperatively causes problems with visualization. Although this may not be construed as a complication, I had to abandon a procedure midway due to excessive fluid extravasation during the procedure. This used to be a common problem years ago, but no longer seems so since I have been using mechanical distraction more frequently. Furthermore, the size of the portals should be kept as small as possible, minimizing potential extravasation between the arthrotomy portal and the instrument. Some clinicians maintain that the use of an arthroscopy pump will also minimize extravasation. If a tourniquet is being used, this should be deflated to avoid further extravasation of fluid into the anterior compartment of the leg.
References
Adkison D P, Bosse M J, Gaccione D R, et al 1991 Anatomical variations in the course of the superficial peroneal nerve. J Bone Joint Surg 73A(1): 112-114
Baker C L, Andrews J R, Ryan J B 1986 Anthroscopic treatment of transchondral talar dome fractures. 2(2): 82-87
Barber F A, Click J, Britt B T 1990 Complications of ankle arthroscopy. 10(5): 263-266
Bassett F H, Gates H S, Billys J B, et al 1990 Talar impingement by the anteroinferior tibiofibular ligament. A cause of chronic pain in the ankle after inversion sprain. J Bone Joint Surg 72A(1): 55-59
Berndt A L, Harty M 1959 Transchondral fractures (osteochondritis dissecans) of the talus. J Bone Joint Surg 41A: 988-1020
Drez D Jr., Guhl J F, Gollehon D L 1982 Ankle arthroscopy. Technique and indications. Clin Sports Med 1: 35-45
Ferkel R D, Guhl J F, Van Buecken K, et al 1993 Complications in ankle arthroscopy: analysis of the first 518 cases [abstr]. Orthop Trans 16: 726-727
Ferkel R D, Karzel R P, Del Pizzo W, et al 1991 Arthroscopic treatment of anterolateral impingement of the ankle. Am J Sports Med 19(5): 440-446
Ferkel R D, Scranton P E Jr. 1993 Arthroscopy of the ankle and foot. J Bone Joint Surg 75A: 1233-1242
Flick A B, Gould N 1985 Osteochondritis dissecans of the talus (transchondral fractures of the talus): review of the literature and new surgical approach for medial dome lesions. 5(4): 165-185
Guhl J F 1986 New techniques for arthroscopic surgery of the ankle: preliminary report. 9: 261-269
Guhl J F, Ferkel R D, Stone J W 1993 Other osteochondral pathology -- fractures and frature defects. In: Guhl J F (ed) Foot and Ankle Arthroscopy. Slack Inc, Thorofare (NJ). pp 131-139
Martin D F, Curl W W, Baker C L 1989 Arthroscopic treatment of chronic synovitis of the ankle. 5(2): 110-114
Myerson M S, Allon S M 1989 Arthroscopic ankle arthrodesis. Contemp Orthop 19(1): 21-27
Myerson M S, Quill G 1991 Ankle arthrodesis. A comparison of an arthroscopic and an open method of treatment. Clin Orthop 268(Jul): 84-95
O'Farrell T A, Costello B G 1982 Osteochondritis dissecans of the talus. The late results of surgical treatment. J Bone Joint Surg 64B: 494-497
Wolin I, Glassman F, Sideman S, et al 1950 Internal derangement of the talofibular component of the ankle. Surg Gynecol Obstet 91: 193-200
|
 |
