572222

research-article2015

FAIXXX10.1177/1071100715572222Foot & Ankle InternationalWarner et al

Article

Fracture-Dislocations Demonstrate Poorer Postoperative Functional Outcomes Among Pronation External Rotation IV Ankle Fractures

Foot & Ankle International® 2015, Vol. 36(6) 641­–647 © The Author(s) 2015 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1071100715572222 fai.sagepub.com

Stephen J. Warner, MD, PhD1, Patrick C. Schottel, MD2, Richard M. Hinds, MD1, David L. Helfet, MD3, and Dean G. Lorich, MD3

Abstract Background: Pronation external rotation (PER) ankle fractures are relatively uncommon but serious ankle injuries. Although recent studies have demonstrated good outcomes of PER IV fractures after operative treatment, the effect of dislocation on functional outcomes has not yet been evaluated. The objective of this study was to compare short-term functional outcomes in PER IV ankle fractures with and without dislocation. Methods: Our database of ankle fractures surgically treated using an anatomic fixation approach by the senior author from 2003 to 2013 was reviewed. All PER IV ankle fracture patients older than 18 years with a minimum of 12 months of follow-up, including Foot and Ankle Outcome Score (FAOS), were included for analysis. Patient demographics, injury characteristics, FAOS, ankle range of motion (ROM), and rate of postoperative complications were compared in PER IV fractures with and without dislocation. Of the 47 PER IV fractures included for analysis, 20 (43%) were fracture-dislocations and 27 (57%) had no dislocation. Mean age of the study cohort was 49 years (range, 24-91 years). Results: The fracture-dislocation cohort demonstrated significantly poorer FAOS (symptoms, 46 vs 70, P = .002; pain, 56 vs 82, P < .001; activities of daily living, 61 vs 84, P = .002; sports, 37 vs 59, P = .036; quality of life, 25 vs 59, P < .001) than the nondislocation cohort. Articular malreduction (33% vs 14%, P = .147) was also more common in the PER IV dislocation group. Rates of syndesmotic malreduction (44% vs 48%, P = .951) were similar between PER IV fractures with and without dislocation. Conclusion: PER IV fracture-dislocations had higher rates of articular malreduction and demonstrated statistically poorer functional outcomes than PER IV fractures with no dislocation. Dislocation in this select subset of ankle fracture patients likely represents a higher energy injury resulting in substantial articular damage and should spur appropriate preoperative patient counseling by the orthopaedic surgeon concerning functional outcome expectations. Level of Evidence: Level III, clinical outcome comparison. Keywords: ankle, fracture, dislocation, pronation external rotation, PER, outcome, prognosis, FAOS Rotational ankle fractures are commonly categorized using the Lauge-Hansen classification system based on his study of ankle position and injurious force using cadaveric models.11,12 Supination external rotation (SER), supination adduction, pronation external rotation (PER), and pronation abduction comprise the 4 Lauge-Hansen designations. Of the 4 types, PER ankle fractures are relatively uncommon but serious ankle injuries and have been reported to account for 14% to 22% of ankle fractures.1,3,10,13 The Lauge-Hansen classification further subcategorizes PER fractures in a graded scale of injury from I to IV, with PER IV fractures associated with higher energy mechanisms of injury.20 However, there is a paucity of literature regarding outcomes of PER IV ankle fractures. Prior investigations

include a case series of 98 PER III and IV fractures and a cohort study comparing the clinical outcomes of SER IV and PER IV fractures.4,20 Furthermore, only a case series of 15 PER IV fracture-dislocations exists to portend prognosis and guide operative management of PER IV 1

Hospital for Special Surgery, New York, NY, USA University of Texas–Houston, Houston, TX, USA 3 Hospital for Special Surgery and New York Presbyterian Hospital, New York, NY, USA 2

Corresponding Author: Stephen J. Warner, MD, PhD, Hospital for Special Surgery, 535 East 70th St, New York, NY 10021, USA. Email: [email protected]

Downloaded from fai.sagepub.com at MOUNT ALLISON UNIV on July 13, 2015

642

Foot & Ankle International 36(6)

fracture-dislocations,21 although these fractures may represent the highest energy rotational ankle injuries.13,20 To our knowledge, functional outcome comparison of fracture-dislocations and fractures without dislocation in a cohort of PER IV ankle fractures has not been previously reported in the literature. The objective of this study was to compare short-term functional outcomes (greater than 12 months) in PER IV ankle fractures with and without dislocation. We hypothesized that patients with PER IV fracturedislocations would demonstrate worse functional outcomes than PER IV fracture patients without ankle dislocation.

Methods Institutional review board approval was obtained prior to a retrospective review of our prospectively created database of ankle fractures operatively treated from 2003 to 2013. All PER IV ankle fractures patients 18 years and older with a minimum of 12 months of follow-up, including Foot and Ankle Outcome Score (FAOS), were included for analysis. Ankle fractures were classified upon assessment of injury radiographs and preoperative magnetic resonance imaging (MRI) and confirmed intraoperatively. PER IV fractures were defined as having a deltoid ligament rupture or medial malleolus fracture, anterior inferior tibiofibular ligament (AITFL) rupture, anterosuperior to posteroinferior fibula fracture, and either a posterior inferior tibiofibular ligament (PITFL) rupture or posterior malleolus fracture.8,16 The MRI scans were routinely obtained and evaluated by a fellowship-trained musculoskeletal radiologist and the senior author to better determine the osseoligamentous ankle injury pattern. Using a 1.5-Tesla magnet (Signa Horizon LX; General Electric Medical Systems, Milwaukee, WI) with a dedicated extremity coil, MRIs were evaluated for rupture of the anterior talofibular ligament (ATFL), AITFL, interosseous membrane, PITFL, and deltoid ligaments. All fractures were treated by a single surgeon in an injury-specific manner. A posterolateral ankle approach was typically performed for antiglide plate fixation of the fibula and for open reduction and internal fixation of posterior malleolus fractures using mini-fragment reconstruction plates when present. A 3.5-mm cortical screw with a soft tissue washer was used to reapproximate all delaminated PITFL injuries. Medial malleolar fractures were open reduced and plated using 2.0-mm fragment reconstruction plates. Deltoid ligament ruptures underwent suture anchor repair when present. Syndesmotic instability was assessed fluoroscopically and via direct visualization following deltoid ligament repair, PITFL repair, and/or malleolar fixation by placing a narrow Hohmann retractor in the interosseous space with application of an external rotation force.23 The syndesmosis was found to be unstable if tibiofibular diastasis was noted.17,24 One or 2 transsyndesmotic screws with 4

cortices of fixation were used to address syndesmotic instability. However, syndesmotic instability and subsequent utilization of syndesmotic screws have greatly decreased in recent years with the advent of deltoid ligament and PITFL repair at our institution. Postoperatively, all patients were placed in a lower leg splint for immobilization, and bilateral ankle computed tomography (CT) scans were routinely obtained prior to hospital discharge to evaluate for articular and syndesmotic malreduction. Patients were transitioned to a controlled ankle motion (CAM) boot at the 2-week postoperative visit, and ankle range-of-motion (ROM) exercises were initiated at this time. Patients were nonweightbearing for a total of 6 weeks, after which full weightbearing and strengthening exercises were initiated. Syndesmotic screws were typically removed no sooner than 4 months after the index procedure.15 Patient age, sex, and medical history, including incidence of hypertension (HTN), history of smoking, diabetes (DM), peripheral vascular disease (PVD), and obesity (body mass index >30 kg/m2), were reviewed. Mechanism of injury (MOI), geriatric fracture (patient age 65 years or older), and osseoligmentous injury pattern, including trimalleolar fracture (medial, posterior, and lateral malleolar fracture), bimalleolar fracture (PITFL rupture with medial and lateral malleolar fracture), posterior malleolar fracture (deltoid ligament rupture with posterior and lateral malleolar fracture), and ligamentous equivalent (deltoid ligament and PITFL rupture with lateral malleolar fracture), were also evaluated. Tibiotalar dislocation was defined as complete loss of articular congruence (Figures 1 and 2). Of the 47 PER IV fractures included in the study, 43% (20/47) were fracture-dislocations and 57% (27/47) did not have dislocation. Mean age of the study cohort was 49 years (range, 24-91 years), with 66% (31/47) being male. Mean study follow-up was 30 months (range, 12-128 months). Thirty-nine of the 47 fractures had preoperative MRI, while 41 fractures underwent CT scan postoperatively. Trimalleolar fractures represented 33% of the study cohort, with bimalleolar (PITFL rupture with medial and lateral malleolar fractures), posterior malleolar (deltoid ligament rupture with posterior and lateral malleolar fractures), and PER IV ligamentous equivalent (deltoid ligament and PITFL rupture with lateral malleolar fracture) patterns comprising the remaining 21%, 28%, and 18%, respectively. No significant difference in demographic characteristics, injury profile, incidence of syndesmotic screw fixation, or medical history data was noted between the 2 groups. Tables 1, 2, and 3 summarize these data. Primary clinical outcome was assessed via the latest postoperative FAOS. FAOS is a validated functional outcome measure evaluating the subcategories of symptoms, pain, activities of daily living (ADL), sports, and quality of life (QOL) with clinical significance defined as a difference of 10 points.2,18 Secondary clinical outcomes included

Downloaded from fai.sagepub.com at MOUNT ALLISON UNIV on July 13, 2015

643

Warner et al

Figure 1.  Anteroposterior (A, C), and lateral (B, D) injury radiographs of a pronation external rotation IV fracture-dislocation.

if a complete bony bridge was observed on the latest postoperative anteroposterior (AP), mortise, and lateral ankle radiographs.7 Statistical analyses were performed to determine differences in postoperative outcomes between PER IV fractures with and without dislocation. Associations between categorical variables were determined with χ2 or Fisher exact tests. Relationships between continuous variables were evaluated with independent samples t tests or MannWhitney U tests if nonnormally distributed. For all statistical tests, P < .05 indicated the level of significance. Descriptive statistics are shown as frequencies and percentages for categorical variables and means ± standard deviations (SD) for continuous variables. Statistical analyses were performed with SAS version 9.3 (SAS Institute, Cary, NC). Figure 2.  Anteroposterior (A), mortise (B), and lateral (C) 2-month postoperative radiographs of the pronation external rotation IV fracture-dislocation.

ROM and rates of articular and syndesmotic malreduction, loss of reduction, complete synostosis, postoperative infection, and wound complications. Articular malreduction was noted if a joint surface gap of greater than 2 mm, articular step-off greater than 2 mm, or intra-articular loose bodies were present on postoperative CT scan.2 Syndesmotic malreduction was defined as a greater than 2-mm difference between the anterior and posterior incisura distance in the operative ankle on postoperative CT scan.6 Loss of reduction was determined by a greater than 2-mm change in the medial clear space upon comparison of immediate postoperative and final radiographs.20 Synostosis was designated

Results The fracture-dislocation cohort demonstrated significantly poorer FAOS results (symptoms, 46 vs 70, P = .002; pain, 56 vs 82, P < .001; activities of daily living, 61 vs 84, P = .002; sports, 37 vs 59, P = .036; quality of life, 25 vs 59, P < .001) than the nondislocation cohort. Range of motion was reduced in the fracture-dislocation group but not significantly different from the nondislocation group. Articular malreduction (33% vs 14%, P = .147) was also more common in the PER IV dislocation group than the PER IV nondislocation group, although this difference was not statistically significant. Syndesmotic malreduction (44% vs 48%, P = .951) was slightly more common in the nondislocation cohort but did not show a statistically significant difference. Rates of loss of reduction, complete synostosis, postoperative infection, and wound complications were

Downloaded from fai.sagepub.com at MOUNT ALLISON UNIV on July 13, 2015

644

Foot & Ankle International 36(6)

Table 1.  Patient Demographic Characteristics Summary. Demographic Characteristics

Fracture-Dislocation (n = 20)

No Dislocation (n = 27)

Total (n = 47)

P Value

48 (18.7) 11 (55.0) 31 (30.0)

50 (18.1) 20 (74.1) 31 (30.9)

49 (18.2) 31 (66.0) 30 (30.2)

.669 .172 .621

Age, mean (SD), y Male, No. (%) Mean follow-up, mean (SD), mo

Table 2.  Injury and Treatment Profile Summary. Injury and Treatment Profile

n

Mechanism of injury   High-energy NOS  Fall   Pedestrian struck  MVA Geriatric fracture ATFL rupture AITFL rupture PITFL rupture IO rupture Deltoid ligament rupture Fracture pattern  Trimalleolar  Bimalleolar   Posterior malleolar   PER IV eq Syndesmotic screw fixation

47

Fracture-Dislocation, No. (%)

No Dislocation, No. (%)

1 (5.0) 15 (75.0) 4 (20.0) 0 (0.0) 5 (25.0) 11 (61.1) 18 (100.0) 5 (27.8) 16 (88.9) 7 (38.9)

3 (11.1) 17 (63.0) 6 (22.2) 1 (3.7) 5 (18.5) 11 (52.4) 21 (100.0) 11 (52.4) 18 (85.7) 8 (38.1)

8 (44.4) 3 (16.7) 5 (27.8) 2 (11.1) 16 (80.0)

5 (23.8) 5 (23.8) 6 (28.6) 5 (23.8) 21 (77.8)

47 39 39 39 39 39 39

47

P Value   .885

.723 .584 >.999 .119 .768 .960   .572

.854

Abbreviations: AITFL, anterior inferior tibiofibular ligament; ATFL, anterior talofibular ligament; eq, equivalent; IO, interosseous membrane; MVA, motor vehicle accident; NOS, not otherwise specified; PER IV, pronation external rotation; PITFL, posterior inferior tibiofibular ligament.

similar between the 2 cohorts. Tables 4 and 5 summarize the primary outcome and secondary outcomes, respectively.

Discussion In the current investigation, we compared short-term functional outcomes in PER IV ankle fractures with and without dislocation. Our hypothesis that patients with PER IV fracture-dislocations would have worse functional outcomes than PER IV fracture patients without ankle dislocation was supported by our results. All 5 FAOS subcategories were statistically and clinically poorer in the PER IV fracturedislocation group. Outcomes of our entire PER IV fracture cohort were comparable with the literature. FAOS subcategory scores of our entire study cohort were symptoms = 60, pain = 71, ADL = 74, sport = 50, and QOL = 45. Donken et al4 reported on 60 cases of PER III-IV injuries and found that 90% had good or excellent clinical outcomes using the Olerud

scoring system. Of their study cohort, 51 fractures were PER IV, 4 fractures were PER III, and 5 fractures were not classified as either and deemed to have an unclear injury pattern. The authors also did not identify fracture-dislocations in that study. In addition, 15% of their study cohort was treated without surgery. To the contrary, the results of our PER IV fracture-dislocation cohort differed from previously reported findings. The FAOS subcategory scores of our fracture-dislocation cohort were symptoms = 46, pain = 56, ADL = 61, sport = 37, and QOL = 25. Stiehl and Schwartz21 presented the outcomes of PER IV fracture-dislocations in their case series from 1990. Using a scoring system devised by Phillips et al17 with a maximum score of 150, Stiehl and Schwartz found an average score of 136.6 among their 15 PER IV fracture-dislocations. The etiology of the difference in reported outcomes between their study and our current investigation is likely multifactorial. Our study used FAOS, a validated functional outcome metric, as our primary

Downloaded from fai.sagepub.com at MOUNT ALLISON UNIV on July 13, 2015

645

Warner et al Table 3.  Patient Medical History Summary. Medical History

Fracture-Dislocation (n = 20), No. (%)

No Dislocation (n = 27), No. (%)

Total (N = 47), No. (%)

P Value

0 (0.0) 6 (30.0) 5 (25.0) 1 (5.0) 7 (35.0)

0 (0.0) 6 (22.2) 3 (11.1) 0 (0.0) 10 (37.0)

0 (0.0) 12 (25.5) 8 (17.0) 1 (2.1) 17 (36.2)

>.999 .545 .258 .426 .886

DM HTN History of tobacco use PVD Obesity

Abbreviations: DM, diabetes mellitus; HTN, hypertension; PVD, peripheral vascular disease.

Table 4.  Foot and Ankle Outcomes. Primary Outcome

Fracture-Dislocation (n = 20), Mean (SD)

No Dislocation (n = 27), Mean (SD)

Total (N = 47), Mean (SD)

P Value

FAOS symptoms FAOS pain FAOS ADL FAOS sport FAOS QOL

46 (21.7) 56 (20.7) 61 (23.3) 37 (32.2) 25 (21.3)

70 (23.3) 82 (20.1) 84 (21.6) 59 (34.4) 59 (29.4)

60 (25.6) 71 (23.7) 74 (24.9) 50 (34.9) 45 (31.0)

.002

Fracture-Dislocations Demonstrate Poorer Postoperative Functional Outcomes Among Pronation External Rotation IV Ankle Fractures.

Pronation external rotation (PER) ankle fractures are relatively uncommon but serious ankle injuries. Although recent studies have demonstrated good o...
345KB Sizes 0 Downloads 5 Views