ORIGINAL ARTICLE

Treatment of the Stress Positive Ligamentous SE4 AnkleFracture: Incidence of Syndesmotic Injury and Clinical

Decision Making

Paul Tornetta, III, MD,* Thomas W. Axelrad, MD, PhD, Tarek A. Sibai, MD,*and William R. Creevy, MD*

Objectives: The objective of the study was to review our experiencewith the treatment of stress positive (+) supination and external patterninjuries using shared decision making with the patients.

Design: Retrospective case review.

Setting: Level 1 trauma center.

Patients: Over a 9-year period, we treated 114 patients (aged 1976 years, average 43 years) with stress (+) supination and externalrotation type bula fractures, who were included in the present study.

Intervention: X-rays were reviewed, and the medial clear space(MCS) measured on the presentation, stress, and nal united radio-graphs. The decision for surgical or nonsurgical management wasmade by the patient and surgeon after a discussion of risks/benetsof both. Syndesmotic instability for the operative cases was diag-nosed by medial widening and talar subluxation on abduction/external rotation stress after bular xation.

Main Outcome Measure: MCS measurement at union.

Results: Of the 114 cases, 54 were denitively treated in a cast, and60 were treated operatively. Twenty-seven (45%) of the operativecases demonstrated syndesmotic instability on radiographic exami-nation. The MCS on stress examination was statistically different,with greater widening seen for operatively treated patients (4.8 0.5vs. 6.9 0.86) (P , 0.001). No patient healed with any subluxationon weight bearing x-rays.

Conclusions: Stress (+) SE pattern bular fractures with minimalMCS widening on stress examination may be treated in a cast to unionwith predictable healing. In those patients treated operatively, the treat-ing surgeon should be aware of the high rate of syndesmotic injury.

Key Words: stress positive, SE4 ankle fracture, cast treatment,conservative management

Level of Evidence: Therapeutic Level III. See Instructionsfor Authors for a complete description of levels of evidence.

(J Orthop Trauma 2012;26:659661)

INTRODUCTIONAnkle fractures are common injuries that are the result

of varying mechanisms. Still, the high degree of variability inpatterns and unpredictable extent of soft tissue injury havemade the treatment for many of these injuries controversial.One such area of debate is in the treatment of supination andexternal (SE) rotation pattern, Weber B ankle fractures thatpresent with a well-aligned mortise but demonstrate instabil-ity on stress examination, by either manual stress or gravitystress radiographs. Specically, Weber B injuries are bularfractures at the level of the joint line, and the SE pattern refersto direction of the force and the position of the foot at the timeof injury. We refer to these injuries as stress positive (+) SE4fractures. The decision to operate may be based on patientfactors as well as the degree of displacement or degree ofinstability or widening on the stress examination. Additionally,although other ligamentous equivalent SE4 Weber B anklefractures have a reported incidence of syndesmotic injury of19%40%, the rate of syndesmotic incompetence in stress (+)SE4 ankle fractures has not been evaluated. Our objective wasto report our experience with the treatment of stress (+) SEpattern injuries using shared decision making with the patientsand determine if the patients healed in a reduced position.

PATIENTS AND METHODSAll patients treated over a 9-year period were identied

from a prospectively collected trauma database. A total of 114patients (aged 1976 years, mean 43 years) with stress (+)isolated SE type Weber B bula fractures were identied.Men comprised 65% of the patients. The mechanism of injurywas a twist and fall in 83% (94) of the patients, motor vehicleaccident in 6% (7), a fall from heights in 4% (5), and unable to

Accepted for publication March 16, 2012.From the *Department of Orthopaedics, Boston University Medical Center,

Boston, MA; and Department of Orthopaedics, Lake Charles MemorialHospital, Lake Charles, LA.

Dr Thomas W. Axelrad is salaried and supported by publication royalties fromWolters Kluwer/Lippincott Williams & Wilkins (www.rockwoodadultsfractures.com). Dr Paul Tornetta III is salaried and supported by publicationroyalties from Smith & Nephew and Wolters Kluwer/Lippincott Williams &Wilkins; holds the intellectual property rights/patent and royalties from Smith &Nephew; receives consulting fees from Smith & Nephew; and is supported bycontracted research with Smith & Nephew. Dr Paul Tornetta III shares owner-ship interest (stocks, stock options, or other ownership interest excludingdiversied mutual funds) with Exploramed. Dr Tarek A. Sibai and Dr WilliamR. Creevy have no nancial disclosures or conicts of interest to declare.

Presented in part at the Annual Meeting of the Orthopaedic Trauma Association,2010, Baltimore, MD; at the Annual Meeting of the American Academy ofOrthopaedic Surgeons, 2011, San Diego, CA; and at the Annual Meeting ofthe American Orthopaedic Association, June 22-25, 2011, Boston, MA.

Reprints: Thomas W. Axelrad, MD, PhD, 1717 Oak Park Boulevard, 3rdFloor, Lake Charles, LA 70601 (e-mail: lc.ortho.trauma@gmail.com).

Copyright 2012 by Lippincott Williams & Wilkins

J Orthop Trauma Volume 26, Number 11, November 2012 www.jorthotrauma.com | 659

be determined in the remaining 8 patients. In those patients inwhom an SE type Weber B bula fracture was identied andthe talus was reduced within the ankle mortise, a manual stressexamination was performed as described previously.1 Briey,the examination was performed with the ankle in neutral ex-ion with a manual abduction-external rotational force of 810lbs (3.64.5 kg). The orthopaedic resident on call, who wastrained on the techniques of the stress examination by theattending trauma surgeon, performed all stress examinations.A fracture was considered stress (+) if there was widening ofthe medial clear space (MCS) greater than 4 mm and at least 1mm greater than the superior joint space, or if talar subluxationoccurred. A senior trauma surgeon made all classications pro-spectively and a single author reviewed all images specicallyfor this study. MCS was measured on the presentation, stress,casted, and nal follow-up radiographs. Additionally, the lat-eral injury and casted lms were used to assess the amount offracture displacement of the distal bula in relation to theproximal bula. All measurements were made on 2 separateoccasions to ensure consistency of the values.

All patients were casted after the nding of a positivestress radiograph, with post casting lms taken to assessreduction. The treatment plan for stress (+) fractures was madeby one of the two full-time orthopaedic trauma surgeons aftera discussion of surgical versus nonsurgical options with eachpatient. Syndesmotic instability for the operative cases wasdiagnosed by medial widening and/or talar subluxation on stressexamination after bular xation.2 Syndesmotic xation wasperformed for any elicited subluxation on Cotton and/or exter-nal rotation stress radiography in the operating room.3 The Cot-ton test is a direct test for syndesmotic instability and involvesmanual lateral displacement of the reduced bula, whereas theexternal rotation stress examination is an indirect test that placesa force on the talus that will demonstrate widening with syn-desmotic injury. All patients were followed until radiographicand clinical healing had occurred. All patients had nal lmsobtained after union with the ankle in a neutral position.

RESULTSPatients identied from the trauma database were fol-

lowed until radiographic union, with no patients lost to follow-up before the achievement of union. Fifty-four (47%) patientswere denitively treated in a cast (6 weeks Non-Weight Bearing,then Weight Bearing As Tolerated in brace), while the remaining60 patients were treated operatively. Twenty-seven (45%) of theoperative cases demonstrated syndesmotic instability afterxation of the bular fracture and were treated with syndesmoticscrews. Twenty of these patients received a single syndesmoticscrew, with the remaining patients receiving 2 or 3 screws.

Factors contributing to the use of multiple screws includedpatient factors, such as diabetes mellitus (1 patient), concern forpatient noncompliance (1 patient), and bone quality (3 patients).The fracture pattern was the reason for multiple syndesmoticscrews in the remaining 3 patients.

The radiographic measurements are reported for the3 treatment groups: cast, Open Reduction Internal Fixationwith a stable syndesmosis, and ORIF with unstable syndesmo-sis requiring one or more syndesmotic screws. Measurementsof the radiographs were done on the PACS system, and thereviewer was blinded to the nal treatment. The same reviewerrepeated measurements on 2 separate occasions with nostatistically signicant differences noted (P = 0.95). Therewas no difference in the MCS measurements among the 3groups on the presentation radiographs. There was a signicantdifference in the measurements of the MCS on the stress radio-graph between the casted and operative patients (P , 0.001)(Table 1). Additionally, the range of widening for the patientstreated operatively was 5.1515.3, whereas the range for thosetreated in a cast was 3.925.9. The postreduction casted lmsalso demonstrated that patients treated nonoperatively had sig-nicantly less MCS displacement than patients treated surgi-cally with or without syndesmotic injuries (P = 0.007) and (P =0.044), respectively. Although these differences where small asa whole, they suggest that greater widening on stress examina-tion results in a decreased ability to maintain the mortisereduced in a cast. We further found that the difference inMCS widening from presentation and stress examination wasgreater in the operatively treated group compared with thecasted group (2.6 vs 4.3). The nding of an increasing MCSafter stress examination was supported by similar ndingsof an increased MCS after casting. All patients went on toheal without evidence of subluxation, and there was no dif-ference in the MCS among the groups at union.

DISCUSSIONThe stress examination requires radiation exposure to

the examiner and can be painful to the patient. Some haveproposed a gravity stress test, however, others argue that thefoot is not controlled during this procedure and can falselyincrease the MCS as a result of plantar exion and thevariations in talar anatomy.4,5 Regardless, treatment of theseinjuries can vary from closed reduction and casting to openreduction and internal xation, with the ideal treatment beingunknown for fractures that are well reduced in a cast.

Several reasons exist for the variations in treatments,including concerns about the competency of the ligamentsabout the medial and lateral aspects of the ankle. It is thoughtthat a stress (+) ankle fracture must have an injury to the deep

TABLE 1. Radiographic Findings

TreatmentInitial MCS

(mm)Stress MCS

(mm)MCS Casted

(mm)Initial Fibular Displacement

(mm)United MCS

(mm)

Cast (54) 2.6 0.4 4.8 0.5 2.8 0.7 1.7 1 2.5 0.5

ORIF, syndesmosis stable (33) 2.6 0.4 6.3 0.9 3.1 0.7 2.6 1.2 2.4 0.3

ORIF, syndesmosis unstable(27)

2.9 0.4 7.6 2.2 3.2 0.6 3.1 1.7 2.5 0.4

Tornetta et al J Orthop Trauma Volume 26, Number 11, November 2012

660 | www.jorthotrauma.com 2012 Lippincott Williams & Wilkins

portion of deltoid ligament, however, studies using magneticresonance imaging (MRI) have found that these injuries maybe only partial in nature.6,7 Koval et al. reviewed the resultsof 19 patients with widening of the MCS, who were treatednonoperatively after MRI showed only partial deltoid injuries.All patients healed without evidence of widening or arthrosis.Fifteen patients were seen at 1-year follow-up, with 14 of the15 patients stating they would have the same treatment again.One concern with this study is that patients without medialtenderness were not subjected to a stress examination andwere excluded from the study. Tenderness has also beenshown to be a poor predictor of deltoid injury, with regardto both positive and negative predictive values.1,8

With regard to the syndesmotic ligament, it has beensuggested that measurements of plain lms and the level of thebular fracture are poor at predicting these injuries.9,10 Stressexamination of Weber B fractures would not be able to assessthe competency of the syndesmosis given that the attachmentsare above the level of the fracture. MRI studies have shown thatthere is a large variation in the amount of disruption that occurs.We have previously reported on incidence of syndesmotic injuryin displaced ligamentous SE4 type fractures to be 39%.2 Inter-estingly, we found a similar incidence in this group of stress (+)injuries in patients who went on to have surgical xation. Thepatients who demonstrated syndesmotic instability after bularxation had a greater degree of ankle instability based on theMCS on initial stress testing, such that these patients hada greater displacement (6.3 9 vs 7.9 2.2, P , 0.05). Thisis a previously unreported nding, and its implications areunknown. None of the patients who were treated in a castdeveloped MCS widening or syndesmotic instability at union.

When we review the decision making for each patient, itwould seem that part of the judgment used in the discussionwith the patient was the amount of talar subluxation seenon the stress examination and the degree of posterior bulardisplacement, with more unstable injuries being treatedoperatively. Although the patients all ultimately made thesedecisions, this process was undoubtedly guided by the impres-sion of the injury by the surgeon. Based on the incidence ofsyndesmotic instability found in this group, it is likely reason-able to offer surgery to patients who have greater degrees ofinstability on their initial stress radiograph, or who are notanatomically aligned in the postreduction cast. This methodof treatment yielded 100% union with no subluxation at nalfollow-up for any patient while avoiding the surgical risk inabout half of the study group.

There are several limitations with our study, includingits retrospective nature and the lack of clinical follow-upwith functional outcome scores. Although we lack functionaloutcome scores at follow-up, we were able to obtain lms atunion for all patients. This is likely because of a combinationof factors, including our treatment protocol of maintainingsolid berglass cast until clinical and bony union and thelack of alternative treatment options in this largely indigentpopulation. None of the patients demonstrated instability withthis stress maneuver. It is our belief that a stable and reducedankle mortise at the time of union has the greatest predictivevalue for long-term outcomes.

Our results show that patients with an intact anklemortise, minimal MCS displacement on stress radiographs,and minimal bular displacement on initial lateral radio-graphs can be successfully treated closed in a cast (Fig. 1). Anexact number with regard to the amount of MCS displace-ment is difcult to ascertain from this data, although wideningor 5 mm or less in our series showed predictable healingwhen offered cast treatment. This is consistent with the datapresented by Yang et al. in which patients with less than5 mm of displacement on stress examination and treated ina cast had an average OlerudMolander ankle score of 89 atnal follow-up.11 As we do not have a protocol that includesMRI studies on patients with these types of injuries, it isimpossible to determine the extent of ligamentous injury inthese patients. However, based on previous studies that foundno correlation between the extent of MCS widening and theseverity of deep deltoid injury, our results would suggest thatthese injuries heal regardless of severity as long as a the mor-tise is reduced and maintained by surgical or cast treatment.

REFERENCES1. McConnell T, Creevy W, Tornetta P III. Stress examination of supi-

nation external rotation-type bular fractures. J Bone Joint Surg Am.2004;86-A:21712178.

2. Stark E, Tornetta P III, Creevy WR. Syndesmotic instability in Weber Bankle fractures: a clinical evaluation. J Orthop Trauma. 2007;21:643646.

3. Mizel MS. Technique tip: a revised method of the Cotton test for intra-operative evaluation of syndesmotic injuries. Foot Ankle Int. 2003;24:8687.

4. Michelson JD, Varner KE, Checcone M. Diagnosing deltoid injury in anklefractures: the gravity stress view. Clin Orthop Relat Res. 2001;(387):178182.

5. Park SS, Kubiak EN, Egol KA, et al. Stress radiographs after anklefracture: the effect of ankle position and deltoid ligament status onmedial clear space measurements. J Orthop Trauma. 2006;20:1118.

6. Koval KJ, Egol KA, Cheung Y, et al. Does a positive ankle stress testindicate the need for operative treatment after lateral malleolus fracture?A preliminary report. J Orthop Trauma. 2007;21:449455.

7. Cheung Y, Perrich KD, Gui J, et al. MRI of isolated distal bular frac-tures with widened medial clear space on stressed radiographs: whichligaments are interrupted? AJR Am J Roentgenol. 2009;192:W7W12.

8. DeAngelis NA, Eskander MS, French BG. Does medial tenderness pre-dict deep deltoid ligament incompetence in supination-external rotationtype ankle fractures? J Orthop Trauma. 2007;21:244247.

9. Nielson JH, Sallis JG, Potter HG, et al. Correlation of interosseousmembrane tears to the level of the bular fracture. J Orthop Trauma.2004;18:6874.

10. Nielson JH, Gardner MJ, Peterson MG, et al. Radiographic measure-ments do not predict syndesmotic injury in ankle fractures: an MRI study.Clin Orthop Relat Res. 2005;(436):216221.

11. Yang E, Wu Y, Dorcil J. Surgical versus nonsurgical treatment of theSE4-equivalent ankle fracture: a retrospective functional outcome study.Orthopedics. 2011;34:271.

FIGURE 1. Patient treated with closed reduction and casting.Presentation film in (A) with stress radiograph shown in (B).Final weight bearing films at union (C).

J Orthop Trauma Volume 26, Number 11, November 2012 Treatment of Stress Positive SE Pattern Injuries

2012 Lippincott Williams & Wilkins www.jorthotrauma.com | 661