Ankle Fractures Resulting From Rotational Injuries James D. Michelson, MD Abstract Ankle fractures are common muscu- loskeletal injuries, and their incidence and severity are increasing among people older than 65 years. 1 Manage- ment of ankle fractures is governed by the character of the fracture in the context of associated medical condi- tions, such as diabetes or severe os- teoporosis. 2,3 Although surgical pro- cedures for managing ankle fractures are well established, decision making is critical so that patients with stable fractures are not unnecessarily ex- posed to the risks of surgery. Ankle Anatomy and Biomechanics The ankle joint consists of the talus, which articulateswith themalleoli me- dially and laterally and the tibial plafond superiorly. In a neutral po- sition, approximately 90% of the load is transmitted through the tibial plafond, with the remaining load borne by the lateral talofibular articulation. The talus in cross section is a trape- zoid that is wider anteriorly than pos- teriorly. Consequently, when the ta- lus dorsiflexes, the increased talar width introduced into the ankle mor- tise forces the fibula to translate lat- erally and rotate externally. Plantar- flexion is associated with internal rotation of the talus relative to the tibia because of the deltoid ligament, which acts as a checkrein on the talus. The ankle is considered stable when, under physiologic loading, the talus moves in a normal pattern through the full range of motion. Therefore, any ankle injury that re- sults in a stable mechanical configu- ration can potentially be treated non- surgically because biomechanically normal function is not compro- mised. 4 In contrast, an ankle is con- sidered unstable whenthe lossof nor- mal constraints around the ankle permits the talus to move in a non- physiologic pattern. 4 Under such cir- cumstances, the dynamic joint surface contact area withinthe ankle isdimin- ished, which predisposes to articular cartilage damage and premature de- generation. Measurements of contact area over the entire dynamic range of motion more accuratelyreflect altered three-dimensional motion 4,5 than do static measures of contact area ob- tained at a single gait position. Res- toration of normal stability and mo- tion in patients with unstable ankle fractures through open anatomic re- duction and internal fixation yields better long-term outcomes than does closed treatment, which may not ad- equately reconstitute either the ana- tomic constraints or the motion. 6 Determination of Ankle Instability Defining the relationship between specific injury patterns and ankle in- stability has been the focus of much clinical and basic science research in the last decade. Although some early Dr. Michelson is Professor, Orthopaedic Surgery, and Director, Clinical Informatics, George Wash- ington University Hospital, George Washington University Hospital Medical Center Medical Ed- ucation and Simulation Center, Washington, DC. Reprint requests: Dr. Michelson, GWUMC Med- ical Education and SimulationCenter, Room 6200, 900 23rd Street NW, Washington, DC 20037. Neither Dr. Michelson nor the department with which he is affiliated has received anything of val- ue from or owns stock in a commercial company or institution related directly or indirectly to the subject of this article. Copyright 2003 by the American Academy of Orthopaedic Surgeons. Ankle fractures are among the most common skeletal injuries; selection of an op- timal management method depends on ankle stability. Stable fractures (eg, isolated lateral malleolar) generally are managed nonsurgically; unstable fractures (eg, bi- malleolar, bimalleolar equivalent) usually are managed with open reduction and in- ternal fixation. Stress radiographs may aid in the management of incomplete del- toid injury in which there is medial swelling and tenderness without radiographic talar shift. A posterior malleolar fracture should be reduced and stabilized if it com- prises >30% of the articular surface and remains displaced after fibular stabiliza- tion. Ankle fractures with syndesmotic injury have additional tibiofibular instabil- ity that can be controlled by screw fixation. However, the choice between metal and bioabsorbable screws, screw size, number of cortices fixed, and indications for screw removal remain controversial. Conditions such as diabetes or advanced age are no longer contraindications to usual management recommendations. J Am Acad Orthop Surg 2003;11:403-412 Vol 11, No 6, November/December 2003 403 work suggested that the lateral mal- leolus was the key to ankle stability, recent investigations have conclusive- ly demonstrated that it is not. 4,7 The primary stabilizer of the ankle under physiologic loading is the deltoid lig- ament, with contributions from both its deep and superficial components. If thedeltoid is rendered incompetent by either direct rupture or medial malleolar fracture, the motion of the talus is markedly changed. During plantarflexion, the talus externally ro- tates from underneath the tibial plafond, which is the reverse of its normal pattern of movement. Stabi- lization of the fibula only partially corrects this abnormal motion. Addi- tionally, reduction of the fibula can be anatomically accurate only if the ta- lus is precisely located in the mortise at the time of reduction. In that sit- uation, the deltoid is at its resting length during healing, which ulti- mately restores the biomechanical function of the deltoid. In the absence of medial injury, fibular osteotomy or fracture does not result in abnormal motion; the talus cannot become nonanatomic within the mortise un- less the medial structures are rup- tured or fractured. Completely re- moving the fibula will not result in any talar displacement with respect to the tibia. Therefore, if the talus is not anatomically located in the mor- tise, the medial structures must be compromised. Observation of such a displaced talus is de facto evidence of an unstable ankle injury. Numerous short- and long-term clinical studies have borne out this re- lationship between specific injury pat- terns and expected clinical results. In a study of nonsurgically treated iso- lated lateral malleolar fractures fol- lowed for a mean of 20 years (range, 16 to25 years), Kristensen and Hansen 8 found good clinical results in 89 of 94 patients (95%), with no casesultimately requiring salvage surgery for posttrau- matic arthritis. In a comparison be- tween surgical and nonsurgical treat- ment in patients with isolated lateral malleolar fractures, Yde and Kris- tensen 9 found no clinical advantages to surgical intervention. In another study of nonsurgically treated isolat- ed lateral malleolar fractures with a mean follow-up of 29 years, 48 of 49 patients (98%) had clinically satisfac- tory outcomes, which was thought to be equivalent to that expected with surgical intervention. 10 In a study of 82 isolated lateral malleolar fractures in which radiographs were scruti- nized to detect any evidence of sub- sequent fibular or talar displacement after the initial injury, none showed any measurable shift in either the ta- lus or fibula, and none required de- layed surgical intervention for sub- sequent evidence of instability. 11 No studies have demonstrated im- proved clinical results in the treat- ment of isolated lateral malleolarfrac- tures by surgical methods compared with nonsurgical management. In contrast, studies in which the sever- ity of injury has been clearly strati- fied have consistently demonstrated that bimalleolar injuries have supe- rior outcomes with surgical reduction and stabilization. The same authors who showed the advantages of non- surgical management for isolated lateral malleolar fractures reported re- sults of a companion study of bimal- leolar fractures in which surgical reduction and stabilization was asso- ciated with improved results com- pared with nonoperative treatment. 12 Phillips et al 6 demonstrated similar findings in a randomized prospective study of 71 patients. They found that nonsurgical methods did not reliably achieve and maintain anatomic re- duction of bimalleolar fractures, which is why such management yielded satisfactory results in only 60% to65% of patients. However, sur- gically achieved reduction was al- most always possible and was asso- ciated with 90% good or excellent clinical results in short-term (3 years) follow-up. 6 Because the instability of a combined deltoid ligament rupture and fibular fracture is equivalent to a bimalleolar fracture, the results of treatment of these so-called bimalle- olar equivalent injuries also are op- timized by surgical reduction and in- ternal fixation. Radiographic Evaluation Classification schemes have been de- veloped to place frequently occurring fracture patterns into groups corre- sponding to the critical components of ankle stability. The Lauge-Hansen system was devised as a way to un- derstand the injury mechanism, thereby guiding closed reduction by precise reversal of the injury mech- anism. 13 The initial word of the clas- sification (eg, supination, pronation) denotes the position of the foot at the time of injury; the following phrase (eg, external rotation) denotes the di- rection of the deforming force. Rota- tional injury patterns are separated into stages I to IV; translational inju- ry patterns are I or II. The more se- vere the degree of injury, the higher the stage number (Fig. 1). The most common injury pattern is supination- external rotation, which accounts for up to 85% of all ankle fractures. 14 The Weber/AO classification sys- tem was developed to guide surgical treatment of ankle fractures. 15 Be- cause it was developed when the fib- ula was thought to bethe primary sta- bilizer of the ankle, it relies mostly on the level of the fibular fracture (Fig. 2). Unfortunately, Weber type B, which accounts for most ankle frac- tures, does not constitute a homoge- neous group; fractures with medial injury benefit from surgical interven- tion, whereas isolatedlateral fractures do not. Subsequent iterations of this classification scheme have explicitly included subcategories to denote the presence of medial injury, 16 resulting in a system as complex as the orig- inal Lauge-Hansen classification. Discussion of the merits of a clas- sification system must explicitly ad- dress its reproducibility, reliability, Ankle Fractures Resulting From Rotational Injuries 404 Journal of the American Academy of Orthopaedic Surgeons and ability to provide an injury prog- nosis. Because of the complexity of the Lauge-Hansen classification sys- tem, several studies have demonstrat- ed poor interobserver and intraob- server reliability and reproducibility, although results with the somewhat simpler Weber system are no bet- ter. 17 In addition, these classifications have limited prognostic usefulness because of the uncertain link they es- tablish between specific fracture pat- terns and associated soft-tissue inju- ries (eg, deltoid ligament) and because of the inherent limitations of plain radiography of ankle frac- tures. 18 Radiographs are two-dimensional representations of the three-dimen- sional talar external rotation instabil- ity pattern of unstable ankle fractures. This incomplete view has led to the misconception that lateral translation causes instability of the talus, where- as anterolateral rotation is the actual cause ofinstability. The distal fragment of the fibular fracture also was thought to be externally rotated; however, pro- spective studies using computed to- mography (CT) have proved that the distal fibular fragment is anatomical- ly aligned to the talus. 18,19 The appar- ent distal fibular external rotation is actually internal rotation of the prox- imal fibular shaft relative to the tib- ia, which is of no consequence to the mechanical behavior of the ankle (Fig. 3). Therefore, surgical criteria based on so-calleddistal fibular displacement should be skeptically viewed because they are based on what amounts to an optical illusion. Another confounding factor in plain radiography is the lack of stan- dardization for magnification, which makes the critical measurement of displacements unreliable. The most reliable criterion for instability is lat- eral talar displacement relative to the tibia. This displacement is best deter- mined by the presence of a lateral talar shift on the anteroposterior or mortise view. A lateral talar shift is defined as a medial clear space larg- er than the superior clear space. This assessment is internally controlled for magnification (Fig. 4). Although frac- ture classification is equally good with either two or three views of the ankle, three views probably afford greater sensitivity for fracture detec- tion than two (eg, anteroposterior and lateral, mortise and lateral). 20 Management Isolated Lateral Fractures Most ankle fractures are stable iso- lated lateral malleolar injuries. The absolute criterion for diagnosing an isolated lateral malleolar fracture is the radiographic display of a fibular fracture without either medial mal- leolar fracture or disruption of the mortise (as defined by equal medial and superior clear spaces radiograph- ically) and without medial ankle ten- derness or swelling on physical ex- amination.Assuming that thesensory examination is intact, the absence of medial tenderness rules out an acute deltoid ligament tear or medial mal- Figure 1 Transaxial diagrammatic view of the Lauge-Hansen supination-external rota- tion injury. Stage of injury increases from I to IV as the injury progresses in an external rotation (arrow) starting at the anterolateral ankle (disruption of the anterior-inferior tib- iofibular ligament). Stage II constitutes a sta- ble isolated lateral malleolar fracture, while stage IV is an unstable injury that involves combined lateral fracture and medial dam- age (either fracture or deltoid rupture). AITFL = anterior-inferior tibiofibular liga- ment; PITFL = posterior-inferior tibiofibular ligament. Figure 2 Weber/AO fractures. The staging is completely determined by the level of fibular fracture. Type A occurs below the plafond, whereas type C starts above the plafond. James D. Michelson, MD Vol 11, No 6, November/December 2003 405 leolar fracture. As noted, long-term follow-up studies have shown that stable isolated lateral malleolar in- juries can be effectively managed nonsurgically. 8-10 Additionally, the au- thors of a study designed to specif- ically examine the reliability of the ra- diographic and clinical diagnosis of these injuries found no late adverse sequelae related to delayed displace- ment and no need for subsequent sur- gery. 11 The immobilization should be designed to protect the ankle from fur- ther injury; a short leg walking cast, a walking prefabricated cast boot, and a high-top tennis shoe all have shown similar satisfactory results. 21 Surgical management of isolated lateral mal- leolar fractures carries a 1% to 3% chance of serious wound complica- tions or infection and, on average, re- sults in greater long-term swelling about the ankle. 22 Bimalleolar and Bimalleolar Equivalent Fractures Bimalleolar ankle fractures are typ- ically managed with open anatomic reduction and internal fixation. Al- though closed reduction can yield sat- isfactory results in up to 65% of cases, it is generally reserved for patients with severe medical problems that preclude surgery. 3,6 Bimalleolar fractures that are initially dislocated or markedlydis- placed should undergo closed reduc- tion and splinting at initial presen- tation to diminish swelling and associated soft-tissue damage. Al- though some advocate immediate sur- gical intervention before the onset of swelling, it may be safer to allow the initial swelling to recede first. 23 Sur- gical management consists of reduc- tion and stabilization of both the lat- eral and medial malleoli. The lateral malleolus generally is reduced and plated first, followed by reduction and stabilization of the medial malleolus using a combination of interfragmen- tary screws and Kirschner wires. Dif- ficulty with the fibular reduction gen- erally occurs because of the medial malleolar fragment blocking talar re- duction. In this circumstance, the me- dial malleolar fragment should be re- duced and stabilized before the fibula is plated. For bimalleolar equivalent frac- tures, in whichthe deltoidis ruptured and the lateral malleolus is fractured, routine repair of the deltoid does not seem to improve clinical results and may lead to a worse long-term out- Figure 3 A, Mortise view radiograph showing rotational malalignment between the proximal and distal fibular segments. B, Transaxial CT scan proximal to the fracture. The space between the tibia (TI) and fibula should be even from anterior to posterior. Here, the tibiofibular space is larger posteriorly than anteriorly, indicating internal rotation of the distal fibular fragment (DF) relative to the tibia. C, Transaxial CT scan through the distal talofibular articulation at the ankle joint shows the distal fibular fragment is anatomic relative to the talus (TA). Figure 4 Mortise view radiograph showing increased medialtalomalleolarclearspacerel- ative to the superior talotibial space, indic- ative of an unstable ankle fracture. Ankle Fractures Resulting From Rotational Injuries 406 Journal of the American Academy of Orthopaedic Surgeons come. 24,25 Although the medial struc- tures are the primary stabilizers of the ankle, the combination of lateral mal- leolar reconstruction and either cast- ing or bracing provides enough sta- bility while the deltoid is healing to protect the mechanicalintegrity of the ankle. Medial exploration should be undertaken only if the talus does not reduce anatomically beneath the plafond, in which case a medial ar- throtomy is made to extricate the in- carcerated deltoid ligament that is blocking reduction of the talus to the medial malleolus. The most difficult clinical presen- tation is lateral fracture with medial deltoid tenderness. In the presence of any radiographic lateral talar shift, the ankle should be presumed to be unstable and managed accordingly. In the presence of tenderness but the absence of a talar shift, either surgi- cal or nonsurgical management may be appropriate. A recent study has suggested that a gravity stress view (ie, anteroposterior radiograph taken with the leg horizontal [medial side up] without ankle support) (Fig. 5) may be useful in detecting complete deltoid ruptures in the absence of a talar shift on conventional views. 26 In- creased talar tilt (≥15°) or talar shift (≥2 mm) occurred onlywhen bothsu- perficial and deep divisions of the deltoid were ruptured. The most common variants of the gravity stress view include a valgus stress view to evaluate the deltoid, and the external rotation stress view, in which the foot is externally rotated under the tibia while a mortise radiograph is taken. Although widely practiced, the inter- pretation and reliability of these views has not yet been studied. Trimalleolar Fractures The posterior plafond component, or posterior malleolus, is a postero- lateral avulsion fracture resulting from the pull of the posterior-inferior tibiofibular ligament,which also is at- tached inferiorly to the distal fibular fracture fragment. If the posterior malleolar fragment includes >25% to 30% of the articular surface of the plafond and remains displaced >2 mm after lateral malleolar reduction, the tibiotalar joint is rendered unstable. 27-29 Fortunately, most such fractures reduce spontaneously after the fibular fracture is reduced. The need for separate reduction and fix- ation of a posterior malleolar fracture is determined from intraoperative, not preoperative, radiographs. Poste- rior malleolar fractures that remain displaced >2 mm after fibular reduc- tion and plating should be reduced and stabilized if they constitute >30% of the articular surface on a lateral ra- diograph. 27 The posterior fracture fragment usually can be reduced by digital pressure, typically through the lateral incision. It is then stabilized by placing a lag screw from either ante- rior to posterior (through a separate anterior stab incision) or the reverse. Syndesmotic Injuries Syndesmotic injuries constitute a special subgroup of fracture in which the fibular fracture is above the level of the tibial plafond and is associat- ed with disruption of the syndesmot- ic ligament between the plafond and the level of fibular fracture. If the fib- ula isanatomically reduced to the tib- ia, a syndesmotic screw is not re- quired for ankle stability as long as the deltoid is intact and the medial malleolus is either intact or surgical- ly stabilized. 30,31 In a bimalleolar equivalent injury, in which it is not possible to reestablish medial integ- rity, a syndesmotic screw should be placed whenever the fibular fracture is ≥3.5 cm above the plafond. 30 The deep deltoid, which usually is rup- tured in medial malleolar fractures, is generally incompetent after stabi- lization of the medial malleolar frag- Figure 5 A, Optimal positioning to obtain the gravity stress view. B, Gravity stress view of the contralateral normal ankle. C, Gravity stress view of the injured ankle. Note the widened medial clear space compared with the contralateral normal ankle. James D. Michelson, MD Vol 11, No 6, November/December 2003 407 ment (to which the superficial deltoid is attached). Although this combina- tion of medial injury and repair was not tested by Boden et al, 30 a prospec- tive clinical study using their criteria in 21 patients with syndesmotic in- juries showed good results in the 18 patients who did not undergo syn- desmotic fixation. 32 For bimalleolar equivalent injuries with the fracture <3.5 cm above the plafond, as well as any syndesmotic injury in which me- dial integrity is restored, some inves- tigators have advocated placement of a syndesmotic screw if the fibula is unstable on intraoperative manual examination. The most common in- traoperative maneuver for determin- ing syndesmotic stability after fibula fixation is the Cotton test, in which a towel clamp or bone clamp is used to place a direct lateral pulling stress on the fibula. The instability test is positive when the fibula can be lat- erally translated more than 1 cm. In addition, intraoperative fluoroscopy during this maneuver may demon- strate an increase of tibiofibular sep- aration, which also would indicate significant syndesmotic instability. Because no standardized intraoper- ative tests for syndesmotic integrity have been validated with follow-up clinical studies, this is an area of con- troversy. An absoluterequirement for use of a syndesmotic screw, regardless of other considerations, is persistent widening of the syndesmosis on in- traoperative radiographs. The syn- desmosis should be reduced using an external clamp and stabilized bystan- dard techniques. Several studies address fixation in syndesmotic injuries. 31 The usual method iswith oneor two screws (3.5 or 4.5 mm) placed parallel to the tib- ial plafond, traversing the tibiofibu- lar joint between 1 and 2 cm above the plafond. The major controversies in this treatment regimen concern the number of screws used, screw size, number of cortices engaged by the screws (three or four), material used for the screws (metalor reabsorbable), activity status after surgery, and need for subsequent hardware removal. In general, the ankle should be held in full dorsiflexion while the screw is placed to prevent overtightening of the syndesmosis, which would limit ankle dorsiflexion. A recent study raised some doubt about this con- cept, 33 but the nature of the experi- mental technique made its applicabil- ity to normal ankle function hard to assess. An earlier laboratory investi- gation 34 that demonstrated adverse mechanical consequences of syndes- motic overtightening remains the most clinically applicable study of this issue. Occult Pilon Fractures In the context of a high-energy in- jury, such as a motor vehicle accident, a seemingly routine trimalleolar an- kle fracture actually may be a pilon fracture with a posterolateral com- pression fragment (Fig. 6). This most commonly occurs in fracture-dis- locations. Suspicion should be elicit- ed when the lateral plafond has a val- gus alignment. A less common but Figure 6 A, Anteroposterior radiograph made at initial injury of an open ankle fracture dislocation. Note the dissociation of the articular surface from the rest of the tibia and the comminution of the fibula. B, Immediate postoperative radiograph. The posterolateral plafond surface is not visible. The loss of the posterolateral quadrant of the tibial plafond was not noticed at the time of injury and was stabilized with routine ankle fracture stabilization methods. The medial malleolus is incompletely reduced, and the fibula may still be short. This also was a consequence of the complexity of the injury and likely would have contributed to a poor long-term result had not the plafond injury been so profound. C, Immediate postoperative transaxial CT scan at the level of the plafond indicates loss of the lateral 25% of the tibial articular surface. D, Anteroposterior radiograph made 6 weeks postoperative demonstrating posterolateral subluxation of the talus into the tibial plafond defect from the pilon fracture. The only surgical salvage option is arthrodesis. Ankle Fractures Resulting From Rotational Injuries 408 Journal of the American Academy of Orthopaedic Surgeons similar injury is encountered with supination-adduction fractures, in which the talus impacts the medial plafond. In such fractures, the plafond assumes a varus position that should elicit suspicion of a significant intra- articular component of the injury. Confirmation of the pilon configura- tion is by CT. Managing such an in- jury with standard ankle fracture sur- gical techniques will result in an unstable, valgus ankle. Regardless of treatment, the patient should be fore- warned that the articular damage from the initial impaction injury may lead to rapid joint degeneration. Alternative Surgical Techniques Rather than using standard lateral plating of the fibula, a one-third tu- bular plate can be placed as an an- tiglide device in the posterolateral po- sition. 35 Minimal contouring is required, and all of the screws can be bicortical because there is no risk of intra-articular protrusion. This tech- nique is very useful in patients with poor bone stock or in whom there is significant comminution. 35 Also, the use of multiple interfragmentary screws without a plate has been ad- vocated, but that construct hasless ro- tational stability than does a standard plating technique and therefore may fail catastrophically (Fig. 7). Finally, when there is a relatively small me- dial malleolarfragment (at risk of fur- ther comminution by placement of screws), stabilization can be achieved by tension band technique. Absorbable implants (eg, pins, screws) also have been investigated for use in stabilizing ankle fractures, primarily to avoid the need for sub- sequent surgery to remove the hard- ware. There are two main polymer formulations used clinically, polygly- colic acid (PGA) and polylactic acid (PLA). Both ultimately degrade into water and carbon dioxide, with PLA resorbing approximately one third as fast as PGA. 36 Although both have been shown to have sufficient strength for clinical use in ankle frac- tures, 36,37 PGA implants have been as- sociated with notable inflammatory response in as many as 50% of pa- tients. 37 This has led investigators to recommend against the use of PGA implants in ankle fractures. Prelim- inary results with PLA implants have been more encouraging, with little or no inflammatory response noted in short-term follow-up. 36 Postoperative Management Postoperative management usually consists of initial splinting in neutral position followed by casting with progressive weightbearing and range of motion. Protected weight bearing may be started immediately in pa- tients in whom stable medial and lat- eral fixation has been achieved. 38 Al- though there are theoretic advantages to early mobilization in patients with ankle fractures, most studies have found minimal benefit associated with starting either weight bearing or Figure 7 A supination-external rotation injury in a patient with no additional risk factors. A, Mortise view postoperative radiograph show- ing multiple interfragmentary screws used for stabilization. The patient was immobilized non–weight-bearing after surgery. B, Mortise view radiograph 6 weeks after surgery showing early loss of fixation with rotational displacement of the fibula. C, Anteroposterior radiograph 12 weeks after surgery. The displacement of the fibula has increased and is accompanied by subtle lateral displacement of the talus. D, Anteroposterior radiograph 18 months after surgery showing complete disorganization of the ankle architecture. The patient was sub- sequently salvaged with an ankle fusion. James D. Michelson, MD Vol 11, No 6, November/December 2003 409 motion in the first few weeks after surgery. 38-41 However, these studies also have demonstrated the safety of instituting such programs, which gives the surgeon a great deal of lat- itude in determining the postopera- tive regimen for each individual pa- tient. Arthroscopy Ankle fractures can be accompanied by occult osteochondral injury to the talus. In one prospective study, 42 31 of 63 patients (49%) undergoing sur- gical treatment of displaced ankle fractures exhibited cartilage damage to the talar dome when a specific ex- ploration was done. The authorswere able to link such damage to poorer clinical outcomes a mean of 25 months after treatment, but their study suffered from a <50% follow- up rate (25/63). Thordarson et al 43 ex- tended this finding with a prospec- tive randomized study of 19 patients at a mean follow-up of 21 months af- ter ankle fracture surgery. Control pa- tients underwent standard fracture surgery; patients in the experimental group also had arthroscopy at the time of fracture surgery, with dé- bridement done as needed. The ar- throscopy group demonstrated osteo- chondral injuries in eight of nine ankles, with one having débridement of a small fragment. The clinical out- come, measured with the Medical Outcomes Study 36-Item Short Form (SF-36) and Musculoskeletal Out- comes Data Evaluation and Manage- ment System(MODEMS), was no dif- ferent between the arthroscopy and control patients. 43 Hintermann et al 42 arthroscopically examined 288 con- secutive patients with ankle fracture at the time of fracture surgery. They found talar chondral injuries in 69% of patients (200/288), with higher rates in Weber type C than Weber type B fractures. Débridement was done in 14% (41/288) and pinning of osteochondral fragments done in 2% (6/288). The arthroscopic complica- tion rate was 6% (18/288); the authors had no control group for comparison. Based on these studies, routine ar- throscopy of ankle fractures does not seem to be warranted because it does not alter treatment outcome, despite the enhanced appreciation of associ- ated chondral injuries it provides. Ankle arthroscopy for persistent pain after otherwise successful frac- ture healing also has been investigat- ed. Under these circumstances, ar- throscopy for well-localized anterior impingement can be helpful for as many as 75% of patients, while sur- gery for ill-defined symptoms and pain is unlikely to be beneficial. 44,45 Intercurrent Medical Considerations Ankle fractures in diabetics have been a source of concern because of the higher rates of complications. Although surgical infection and wound dehiscence rates are greater in diabetic than nondiabetic pa- tients, attempts to maintain a closed reduction in unstable fractures is as- sociated with a very high rate of skin breakdown and infection. 2,3 This is because of the high contact pres- sures between the skin and cast re- quired to maintain the reduction. Unstable fractures, in which closed reduction is difficult to achieve and maintain, probably should be man- aged surgically to afford greater con- trol over the fracture and may even result in a lower overall complica- tion rate. 2,3 To minimize the risk of fixation failure and Charcot degen- eration in diabetic patients, the post- operative regimen of progressive weight bearing should be delayed until there is radiographic evidence of healing. Ankle fractures are the fourth most common fracture in those older than 65 years; 1 most are the result of sig- nificant trauma. 46 Early reports sug- gested thatsuch patients had increased rates of surgical complications, lead- ing some authors to recommend non- surgical management for all ankle frac- tures in patients older than 50 years. However, more recent studies have not demonstratedany age-related risks to surgery beyond those posed by oth- er comorbidities. 47 Therefore, the cri- teria for surgery should not be differ- ent for elderly patients than for younger individuals. The primary sur- gical concern in people older than 65 years is the increased prevalence of osteoporosis, which may necessitate the use of alternative fixation strat- egies, such as posterolateral fibular an- tiglide plating. 35 Summary Management of ankle fractures is de- termined by the assessment of their mechanical stability. Although cur- rent radiographic classifications pro- vide uncertain guidance in determin- ing ankle stability, the most reliable criterion for instability is the radio- graphic presence of lateral talar shift (ie, increased medial clear space rel- ative to the superior tibiotalar clear space). Stable ankle fractures (eg, iso- lated lateral malleolar) are satisfacto- rily treated by closed methods, whereas unstable fractures (eg, bimal- leolar, trimalleolar) have superior clinical outcomes with surgical reduc- tion and stabilization. The appropri- ate role for routine arthroscopy to manage unstable injuries before sta- bilization, and the use of biodegrad- able implants for stabilization, are both under continuing investigation, with the benefits of either unproven at present. The presence of intercur- rent medical conditions, such as di- abetes andadvanced age, is not a con- traindication to the usual treatment recommendations because surgery can now be relatively safely done in these patients. Ankle Fractures Resulting From Rotational Injuries 410 Journal of the American Academy of Orthopaedic Surgeons References 1. Barrett JA, Baron JA, Karagas MR, Beach ML: Fracture risk in the U.S. Medicare population. J Clin Epidemiol 1999;52:243-249. 2. Flynn JM, Rodriguez-del Río F, Pizá PA: Closed ankle fractures in the diabetic pa- tient. Foot Ankle Int 2000;21:311-319. 3. Blotter RH, Connolly E, Wasan A, Chapman MW: Acute complications in the operative treatment of isolated an- kle fractures in patients with diabetes mellitus. Foot Ankle Int 1999;20:687-694. 4. Michelsen JD,AhnUM,HelgemoSL:Mo- tion of the ankle in a simulated supina- tion-external rotation fracture model. J Bone Joint Surg Am 1996;78:1024-1031. 5. Clarke HJ, Michelson JD, Cox QG, Jin- nah RH: Tibio-talar stability in bimalle- olar ankle fractures: A dynamic in vitro contact area study. Foot Ankle 1991;11: 222-227. 6. Phillips WA, Schwartz HS, Keller CS, et al: A prospective, randomized study of the management of severe ankle frac- tures. J Bone Joint Surg Am 1985;67: 67-78. 7. Earll M, Wayne J, Brodrick C, Vokshoor A, Adelaar R: Contribution of the del- toid ligament to ankle joint contact characteristics: A cadaver study. Foot Ankle Int 1996;17:317-324. 8. Kristensen KD, Hansen T: Closed treat- ment of ankle fractures: Stage II supi- nation-eversion fractures followed for 20 years. Acta Orthop Scand 1985;56:107-109. 9. Yde J, Kristensen KD: Ankle fractures: Supination-eversion fractures stage II. Primary and late results of operative and non-operative treatment. Acta Or- thop Scand 1980;51:695-702. 10. Bauer M, Jonsson K, Nilsson B: Thirty- year follow-up of ankle fractures. Acta Orthop Scand 1985;56:103-106. 11. Michelson JD, Ahn U, Magid D: Eco- nomic analysis of roentgenogram use in the closed treatment of stable ankle fractures. J Trauma 1995;39:1119-1122. 12. Yde J, Kristensen KD: Ankle fractures: Supination-eversion fractures of stage IV. Primary and late results of operative and non-operative treatment. Acta Or- thop Scand 1980;51:981-990. 13. Lauge-Hansen N: Fractures of the ankle: II. Combined experimental-surgical and experimental-roentgenologic investiga- tions. Arch Surg 1950;60:957-985. 14. Burwell HN, Charnley AD: The treat- ment of displaced fractures at the ankle by rigid internal fixation and early joint movement. J Bone Joint Surg Br 1965;47: 634-660. 15. Malleolar fractures, in Müller ME, Allgöwer M, Schneider R, Willenegger H (eds): Manual of Internal Fixation: Techniques Recommended by the AO Group, ed 2. Berlin, Germany: Springer-Verlag, 1979, pp 282-299. 16. Harper MC: Ankle fracture classification systems: A case for integration of the Lauge-Hansen and AO-Danis-Weber schemes. Foot Ankle 1992;13:404-407. 17. Thomsen NO, Overgaard S, Olsen LH, Hansen H, Nielsen ST: Observer varia- tion in the radiographic classification of ankle fractures. J Bone Joint Surg Br 1991;73:676-678. 18. Michelson JD, Magid D, Ney DR, Fish- man EK: Examination of the pathologic anatomy of ankle fractures. J Trauma 1992;32:65-70. 19. Harper MC: The short oblique fracture of the distal fibula without medial inju- ry: An assessment of displacement. Foot Ankle Int 1995;16:181-186. 20. Brandser EA, Berbaum KS, Dorfman DD, et al: Contribution of individual projections alone and in combination for radiographic detection of ankle fractures. AJR Am J Roentgenol 2000;174: 1691-1697. 21. Michelson JD: Fractures about the ankle. J Bone Joint Surg Am 1995;77:142-152. 22. Bauer M, Bergström B, Hemborg A, Sandegard J: Malleolar fractures: Non- operative versus operative treatment. A controlled study. Clin Orthop 1985;199: 17-27. 23. Hoiness P, Engebretsen L, Stromsoe K: The influence of perioperative soft tis- sue complications on the clinical out- come in surgically treated ankle frac- tures. Foot Ankle Int 2001;22:642-648. 24. Baird RA, Jackson ST: Fractures of the distal part of the fibula with associated disruption of the deltoid ligament: Treatment without repair of the deltoid ligament. J Bone Joint Surg Am 1987;69: 1346-1352. 25. Harper MC: The deltoid ligament: An evaluation of need for surgical repair. Clin Orthop 1988;226:156-168. 26. Michelson JD, Varner KE, Checcone M: Diagnosing deltoid injury in ankle frac- tures: The gravity stress view. Clin Or- thop 2001;387:178-182. 27. Harper MC, Hardin G: Posterior malle- olar fractures of the ankle associated with external rotation-abduction inju- ries: Results with and without internal fixation. J Bone Joint Surg Am 1988;70: 1348-1356. 28. McDaniel WJ, Wilson FC: Trimalleolar fractures of the ankle: An end result study. Clin Orthop 1977;122:37-45. 29. Hartford JM, Gorczyca JT, McNamara JL, Mayor MB: Tibiotalar contact area: Contribution of posterior malleolus and deltoid ligament. Clin Orthop 1995; 320:182-187. 30. Boden SD, Labropoulos PA, McCowin P, Lestini WF, Hurwitz SR: Mechanical considerations for the syndesmosis screw:Acadaver study. J Bone Joint Surg Am 1989;71:1548-1555. 31. Wuest TK: Injuries to the distal lower extremity syndesmosis. J Am Acad Or- thop Surg 1997;5:172-181. 32. Yamaguchi K, Martin CH, Boden SD, Labropoulos PA: Operative treatment of syndesmotic disruptions without use of a syndesmotic screw: A prospec- tive clinical study. Foot Ankle Int 1994; 15:407-414. 33. Tornetta P III, Spoo JE, Reynolds FA, Lee C: Overtightening of the ankle syn- desmosis: Is it really possible? J Bone Joint Surg Am 2001;83:489-492. 34. Needleman RL, Skrade DA, Stiehl JB: Effect of the syndesmotic screw on an- kle motion. Foot Ankle 1989;10:17-24. 35. Schaffer JJ, Manoli A II: The antiglide plate for distal fibular fixation: A bio- mechanical comparison with fixation with a lateral plate. J Bone Joint Surg Am 1987;69:596-604. 36. Thordarson DB, Samuelson M, Shepherd LE, Merkle PF, Lee J: Bioabsorbable ver- sus stainless steel screw fixation of the syndesmosis in pronation-lateral rotation ankle fractures: A prospective random- ized trial. Foot Ankle Int 2001;22:335-338. 37. Hovis WD, Bucholz RW: Polyglycolide bioabsorbable screws in the treatment of ankle fractures. Foot Ankle Int 1997; 18:128-131. 38. van Laarhoven CJ, Meeuwis JD, van der Werken C: Postoperative treatment of internally fixed ankle fractures: A prospective randomised study. J Bone Joint Surg Br 1996;78:395-399. 39. Dogra AS, Rangan A: Early mobilisa- tion versus immobilisation of surgical- ly treated ankle fractures: Prospective randomised control trial. Injury 1999; 30:417-419. 40. Egol KA, Dolan R, Koval KJ: Functional outcome of surgery for fractures of the ankle: A prospective, randomised com- parison of management in a cast or a functional brace. J Bone Joint Surg Br 2000;82:246-249. 41. Ahl T, Dalén N, Lundberg A, Bylund C: Early mobilization of operated on ankle James D. Michelson, MD Vol 11, No 6, November/December 2003 411 fractures: Prospective, controlled study of 40 bimalleolar cases. Acta Orthop Scand 1993;64:95-99. 42. Hintermann B, Regazzoni P, Lampert C, Stutz G, Gächter A: Arthroscopic findings in acute fractures of the ankle. J Bone Joint Surg Br 2000;82:345-351. 43. Thordarson DB, Bains R, Shepherd LE: The role of ankle arthroscopy on the surgical management of ankle frac- tures. Foot Ankle Int 2001;22:123-125. 44. van Dijk CN, Verhagen RA, Tol JL: Ar- throscopy for problems after ankle frac- ture. J Bone Joint Surg Br 1997;79:280-284. 45. Bonnin M, Bouysset M: Arthroscopy of the ankle: Analysis of results and indi- cations on a series of 75 cases. Foot An- kle Int 1999;20:744-751. 46. Jensen SL, Andresen BK, Mencke S, Nielsen PT: Epidemiology of ankle frac- tures: A prospective population-based study of 212 cases in Aalborg, Denmark. Acta Orthop Scand 1998;69:48-50. 47. Pagliaro AJ, Michelson JD, Mizel MS: Results of operative fixation of unstable ankle fractures in geriatric patients. Foot Ankle Int 2001;22:399-402. Ankle Fractures Resulting From Rotational Injuries 412 Journal of the American Academy of Orthopaedic Surgeons