Identifying Risk Factors for Osteonecrosis After Talar Fracture.

OBJECTIVE

To identify patient, injury, and treatment factors associated with the development of avascular necrosis (AVN) after talar fractures, with particular interest in modifiable factors.

DESIGN

Retrospective chart review.

SETTING

21 US trauma centers and 1 UK trauma center.

PATIENT SELECTION CRITERIA

Patients with talar neck and/or body fractures from 2008 through 2018 were retrospectively reviewed. Only patients who were at least 18 years of age with fractures of the talar neck or body and minimum 12 months follow-up or earlier diagnosis of AVN were included. Further exclusion criteria included non-operatively treated fractures, pathologic fractures, pantalar dislocations, and fractures treated with primary arthrodesis or primary amputation.

OUTCOME MEASUREMENTS AND COMPARISONS

The primary outcome measure was development of AVN. Infection, nonunion, and arthritis were secondary outcomes.

RESULTS

In total, 798 patients (409 men; 389 women; age 18-81 years, average 38.6 years) with 798 (532 right; 264 left) fractures were included and were classified as Hawkins I (51), IIA (71), IIB (113), III (158), IV (40), neck plus body (177), and body (188). In total, 336 of 798 developed AVN (42%), more commonly after any neck fracture (47.0%) versus isolated body fracture (26.1%, P < 0.001). More severe Hawkins classification, combined neck and body fractures, body mass index, tobacco smoking, right-sided fractures, open fracture, dual anteromedial and anterolateral surgical approaches, and associated medial malleolus fracture were associated with AVN ( P < 0.05). After multivariate regression, fracture type, tobacco smoking, open fractures, dual approaches, age, and body mass index remained significant ( P < 0.05). Excluding late cases (>7 days), time to joint reduction for Hawkins type IIB-IV neck injuries was no different for those who developed AVN or not. AVN rates for reduction of dislocations within 6 hours of injury versus >6 hours were 48.8% and 57.5%, respectively. Complications included 60 (7.5%) infections and 70 (8.8%) nonunions.

CONCLUSIONS

Forty-two percent of all talar fracture patients developed AVN, with talar neck fractures, more displaced fractures, and open injuries having higher rates. Injury-related factors are most prognostic of AVN risk. Surgical technique to emphasize anatomic reduction, without iatrogenic damage to remaining blood supply appears to be prudent.

LEVEL OF EVIDENCE

Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.