Does Anterior Impaction Affect Radiographic Outcomes of Pilon Fractures?

Abstract

Background

The outcomes of pilon fractures are multifactorial. Anterior articular impaction requires sagittal plane correction (anterior distal tibia angle (ADTA)) with articular reduction. However, there is a risk of avascular necrosis of the articular fragments and postoperative tibiotalar arthritis. The purpose of this study was to determine if the presence of anterior impaction affects radiographic alignment after definitive fixation.

Methods

Retrospective cohort study of patients who underwent operative management for pilon fractures at two academic, level 1 trauma centers between September 2005-September 2016. Fractures were categorized as having anterior impaction or no anterior impaction after review of preoperative radiographic and computer tomography imaging. Patient demographics and postoperative time to union was recorded. Quality of reduction was measured using (ADTA) (degrees), lateral distal tibia angle (LDTA) (degrees), and lateral talar station (LTS) (millimeters) from postoperative radiographs. Statistical analysis compared fracture patterns with anterior impaction to those without.

Results

208 patients met inclusion criteria. 132 fractures (63.4%) were determined to have anterior impaction. Cohorts were similar in demographics and medical comorbidities (p>0.05). Mean ADTA, LTDA, and LTS for the anterior impaction group 83.5°, 89.7°, and 2.4mm versus 84.6°, 89.9°, and 2.0mm in the group without anterior impaction. Cohorts significantly differed in ADTA(p=0.01), but not LDTA(p=0.12) or LTS(p=0.44). No significant differences were found between cohorts with infection (>0.05), nonunion(p=0.76), unplanned reoperation(p=0.56), or amputation(p=0.34).

Conclusion

This study demonstrated no significant differences in the coronal or sagittal plane alignment when comparing definitively fixed pilon fractures with and without anterior impaction. Additional studies are needed to evaluate the longterm clinical impact of failing to restore ADTA.

Level of Evidence: III

Introduction

Pilon fractures are commonly associated with articular comminution, soft tissue injury and neurovascular injury.^1-5 The known injury mechanism consisting of axial compression of the talus into the tibial plafond often results in significant articular impaction.⁶ Mechanism of injury, severity of the associated soft tissue injury and fracture characteristics determine operative strategy as well as prognosis.⁶

As with all articular fractures, anatomical reconstruction of the articular surface and overall restoration of the mechanical axes of the distal tibia should be the goal. Postoperative malalignment contributes to accelerated post-traumatic osteoarthritis of the tibiotalar joint.^6-8 Sagittal plane deformities alter biomechanics and joint forces to a greater degree than coronal plane deformities.^5,9,10 However, there is not a current consensus on gold-standard radiographic parameters used to assess postoperative reduction in pilon fractures. This is largely due to the significant variability in baseline anatomy of patients, variability in radiographic positioning of injured limb, training level of evaluators, as well as poor inter-and intra-rater reliability.^1,2,7,11-18 Furthermore, secondary radiographic parameters that are more difficult to measure, including the presence of anterior impaction, create a unique challenge for surgeons regarding attaining anatomical reduction.

While previous studies have investigated the impact of soft tissue injury and fracture patterns on outcomes,^6-8 associated anterior impaction has not been emphasized. Incidence of avascular necrosis has been reported up to 25% after injury and post-traumatic arthritis up to 31% depending on initial fracture severity and length of follow-up.³⁸ Anterior articular impaction makes both sagittal plane correction and obtaining and maintaining articular reduction difficult. Initial fracture reduction of the comminuted anterior articular surface and postoperative fragment resorption due to avascular necrosis can create problems with joint containment that increases the chance of postoperative tibiotalar arthritis. This is exacerbated when there is limited ankle motion that continues to stress the anterior tibiotalar joint (Figure 1). The purpose of this study was to determine if the presence of anterior impaction affects radiographic alignment outcomes of pilon fracture after undergoing open reduction internal fixation (ORIF). We hypothesize that the presence of anterior impaction will result in worse radiographic alignment when compared to fracture patterns without anterior impaction.

Figure 1.

(A) A lateral radiograph showing left pilon fracture with anterior impaction, (B) An immediate postoperative lateral radiograph showing anterolateral primary plate and accessory anterior rim plate; (C) A two-year postoperative lateral radiograph showing anterior talar subluxation, anterior plafond avascular necrosis and post-traumatic tibiotalar arthrosis.

Methods

After Institutional Review Board (IRB) approval, an electronic medical record (EMR) retrospective review was performed on patients who underwent pilon fracture ORIF at two American College of Surgeons-verified level 1 academic trauma centers between September 1, 2005, and September 30, 2016. The surgeons were all fellowship-trained orthopedic traumatologists.

Inclusion criteria included skeletally mature patients, AO/OTA 43B or C fractures¹⁹ managed acutely or in a staged fashion with ≥6 months of clinical follow-up. Patients with missing/unidentifiable data, lack of measurable radiographs/computed tomography (CT) scan, or lack of follow up were excluded.

Clinical data points including patient demographics and comorbidities, AO/OTA fracture classification,¹⁹ open fracture type,²⁰ intraoperative complications, type and number of additional procedures performed after index ORIF, time to fracture union and complications were recorded. Complications included superficial surgical infection (SSI, requiring antibiotics/wound care), deep infection (requiring return to operating room), nonunion, unplanned reoperation, and amputation.

Radiographic Measurements

Fractures were categorized as having anterior impaction or no anterior impaction through methodical, detailed review of pre-operative radiographic and CT imaging. Anterior impaction was defined as presence of cranial axial compression of the anterior one-third of distal tibia as assessed on pre-operative imaging (Figure 1). Quality of reduction was quantified, via the postoperative AP and lateral ankle radiographs, utilizing the following measures: anterior distal tibial angle (ADTA (degrees), lateral distal tibial angle (LDTA (degrees)), and lateral talar station (LTS (millimeters)). Postoperative CT scans were not part of the institutional protocols and therefore not routinely obtained. Although CT scans are more accurate for judging reduction, plain radiographs are the current gold standard for postoperative reduction evaluation to minimize patient radiation exposure and cost. However, on an individual basis, CT scans were obtained to further evaluate concerns for postoperative reduction or changes in the postoperative period.

Anterior Distal Tibial Angle (ADTA)

The ADTA was measured as described by Magerkurth and colleagues.²¹ This angle was used to assess sagittal plane alignment. First, a line is drawn on the anterior and posterior tibial joint surface, with care taken to exclude osteophytes, on the lateral radiograph. A second line is then drawn within the anatomic, longitudinal axis of the tibia. The anterior angle between these lines defines the ADTA. (Figure 2) Normal angles range from 78° to 82°.²¹ Other references support a much wider variation from 76°-92°.¹⁶

Figure 2.

(A) An injured right ankle lateral radiograph showing a pilon fracture with anterior impaction and comminution; (B) an axial CT scan showing anterior pilon comminution; (C) an immediate postoperative lateral radiograph of the right ankle showing anterolateral primary plate and accessory anterior plate with ADTA of 83.3º; (D) an immediate postoperative lateral radiograph of the right ankle showing LTS of -5.1mm; (E) an immediate postoperative anteroposterior (AP) radiograph of the right ankle showing anterolateral primary plate with accessory anterior and medial plates, bone graft substitute, and an LDTA of 88º.

Lateral Distal Tibial Angle (LDTA)

The LDTA is measured by drawing a line across the weight bearing articular surface of the tibial plafond on the mortise radiograph. A second line is created along the anatomical, long axis of the tibia. The lateral angle is considered the LDTA (Figure 2). Normal angles vary widely and range from 86 to 92°.^16-18,21

Lateral Talar Station (LTS)

The LTS is measured by creating a sector of a circle that is fitted to the talar joint surface to first define the talar joint radius.^20,22 The distance from the center of this circle and the anatomic, longitudinal axis of the tibia is measured on the lateral radiograph (Figure 2). Normal ranges from –0.8076 mm to 3.1496 mm.²³

Statistical Analysis

Statistical analysis included two-tailed t-test of unequal variance to compare fracture patterns, postoperative complications, and radiographic measurements of anterior impaction to patterns without anterior impaction, with a p-value of <0.05 determined as significant. A Chi-Square test was used to evaluate significant difference regarding OTA-AO classification between the groups.

Results

A total of 208 pilon fractures met inclusion criteria. Of the total, 132 fractures (63.4%) were determined to involve anterior impaction and 76(36.6%) without anterior impaction prior to definitive fixation. The two cohorts were similar regarding age, BMI, tobacco use, and associated medical comorbidities (p<0.05; Table 1). The mean clinical and radiographic follow-up was 1.9 ± 1.7 years (min: 6 months, max: 8.4 years) for the anterior impaction cohort and 2.0± 1.7 years (min: 6 months, max: 8.3 years) for the no anterior impaction cohort.

Table 1.

Patient Cohort Characteristic Comparison Between Patients With and Without Anterior Impaction

	Anterior Impaction	No Anterior Impaction	P-value
Total Patients	192	130
Age (SD)	43.3 (± 14.2)	46.9 (± 16.3)	0.06
Male	128	71
Female	64	59	0.42
BMI (range)	31.3 (± 12.9)	32.6 (18.3-50.2)	0.18

SD = standard deviation; BMI = Body Mass Index

Radiographic Results

The mean postoperative ADTA, LTDA, and LTS the anterior impaction cohort was 83.5° (69.7-99.6°), 89.7° (79-101.8°), and 2.4mm (-9.6–12.0mm), respectively. The mean postoperative ADTA, LTDA, and LTS for fractures without anterior impaction was 84.6° (75.3-99.8°), 88.9° (74.8-97.8°), and 2.0mm (-10.5–12.1mm), respectively. There was no significant difference in ADTA (p=0.06), LDTA (p=0.09) or LTS (p=0.22; Table 2).

Table 2.

Radiographic Alignment Measure Comparison Between Patients With and Without Anterior Impaction

Alignment Parameters
Measure Alignment	Anterior Impaction	No Anterior Impaction	P-value
ADTA	83.44	84.88	0.01*
LDTA	89.70	89.36	0.12
LTS	2.41	1.75	0.44

* indicates significant; ADTA = Anterior Distal Tibia Angle (degrees); LDTA = Lateral Distal Tibia Angle (degrees); LTS = Lateral Talar Station (mm)

Clinical Outcomes

Pilon fractures with anterior impaction did not demonstrate statistically significant differences for the rates of SSI (p=0.92), deep infection (p=0.81), nonunion (p=0.76), unplanned reoperation (p=0.98), or amputation (p=0.34) (Table 3).

Table 3.

Postoperative Complication Comparison Between Patients With and Without Anterior Impaction

Complications	Anterior Impaction (number of complications)	No Anterior Impaction (number of complications)	P-value
Superficial Infection	45	26	0.47*
Deep Infection	45	23	0.22*
Unplanned Reoperation	83	52	0.56
Nonunion	30	22	0.76
Amputation	12	5	0.34*

* indicates significance (p<0.5)

Discussion

Our data demonstrates that pilon fractures with anterior impaction do not have significant differences in postoperative ADTA, LDTA and LTS or clinical rates of complications including SSI, deep infection, non-union, unplanned reoperation, or amputation. While the radiographic measures are not significantly changed with the presence of anterior impaction in pilon fractures, the clinical significance is still unknown.

Anterior impaction is often recognized as comminution and cranial displacement of the anterior tibial articular surface that occurs in the event of significant dorsiflexion of the ankle during high-energy axial compression.¹⁵ Frequently associated with OTA/AO type 43C fractures, anterior impaction can be difficult to reduce and, due to the associated comminution, difficult to maintain the reduction.¹⁵ This often requires accessory plate fixation and graft or graft substitute augmentation (Figure 2). However, failure to achieve and maintain this reduction and anterior containment leads to persistent instability, anterior escape of the talus, and an increased risk of post-traumatic osteoarthritis due to asymmetric joint wear.^{1-6,8-11,13,24}

The current literature fails to report the relationship of anterior impaction on both radiographic and clinical outcomes. Striving for anatomical reduction and maintaining it is the goal.^{1-6,8-11,13-15,24-28} Anterior articular impaction increases the difficulty of correcting the sagittal plane and the articular surface due to comminution and the void left behind by reducing the impacted bone.⁴ Even if the reduction is anatomic, the comminuted anterior fragments may resorb over time due to avascular necrosis and can create problems with anterior joint containment that increases the risk of postoperative tibiotalar arthritis.³⁸ Furthermore, limited ankle motion is common after complex pilon fractures and exacerbates the abnormal stress at the anterior tibiotalar joint surface (Figure 1). However, our average 2-year follow up study, although relatively short-term, suggests that presence of anterior impaction does not have a significant influence on postoperative clinical complications. Two years is enough time for avascular necrosis and collapse to happen.

The goals of surgical management of pilon fractures include reconstruction of the articular surface, restoration of the mechanical axes across the ankle joint, and stable fixation to allow early joint motion. Rüedi and Allgöwer²² proposed these four classic technical principles that are still followed with the inclusion of advances in plating techniques, and updated imaging technology such as CT scans for preoperative planning. The majority of pilon fractures present a great risk of instability in the sagittal plane, whilst malleolar fractures often result in coronal plane instability.^29-31 Chronic sagittal malalignment with anterior subluxation of the talus causes progressive post-traumatic ankle arthritis due to abnormal contact stresses.^29-31 Post-traumatic ankle arthritis follows a consistent form and the tibiotalar joint compensates by forming anterior tibial rim osteophytes to contain the talus. This ultimately leads to anterior impingement, commonly associated with pain with ankle dorsiflexion and decreased ankle dorsiflexion which propagates the existing problem. Therefore, maintaining proper sagittal alignment has been considered an essential factor for favorable long-term outcomes.^29-31

More recently, coronal plane alignment has been cited as an important factor during consideration of the arthritic ankle for long-term success of secondary procedures including tibiotalar arthrodesis or ankle arthroplasty.^29-32 In clinical practice, both sagittal and coronal planes should be carefully considered and maintained during definitive fixation. Our study results support that regardless of anterior impaction presence, objective radiographic outcomes are similar between the two groups at an average of 2-year follow-up.

Injury severity of the fracture pattern as well as the quality of reduction at time of definitive fixation have been known to correlate with function.^9,10,21,26 Multivariate analysis demonstrated lower physical component summary (PCS) scores in OTA/AO C-type fractures and patients with lower education level. Additionally, inadequate reduction (retained joint incongruity ^≥ 2 mm), and asymmetric ankle range of motion were significant risk factors for decreased scores in PCS.²⁷ However, other studies argue that the most predictive factors for clinical outcome are patient demographic factors rather than radiographic injury or accuracy of reduction.²⁵ Despite variation in injury severity and patient-related factors, patient outcomes after pilon fractures have been reported to improve for up to five years after injury.²⁵ While our current data does not demonstrate difference in postoperative outcomes with the presence of anterior impaction, it is also difficult to control for presence of this fracture characteristic alone as high-energy pilon fractures are often unique in their presentation. Additional PROs in these patients would also be useful for determining effect on quality of life and functional status postoperatively.

Our study has inherent limitations including the retrospective nature and use of plain radiographs for postoperative alignment measurement. Over the last decade, postoperative imaging modalities have changed. Several studies have demonstrated the limitations in fluoroscopy and plain radiographs in the assessment of the articular surface and fracture reduction.^28,33-35 Further studies have demonstrated the improved accuracy of CT compared to conventional radiographs.^11,36-38 However, CT is not routinely obtained postoperatively due to cost, radiation exposure, as well as metal implant artifact obscuring accurate evaluation. Many of our cohorts did not have postoperative CTs, therefore these were not utilized. Additionally, our focus was to utilize radiographs to measure reduction alignment, as radiographs are the main imaging modality obtained during clinical follow-up. We, however, did not use comparison ankle radiographs to understand what measurements would be considered normal for each patient. Finally, patient reported outcomes (PROs) would help further clarify the relationship between anterior impaction and mid- to long-term functional outcomes.

Conclusion

The presence of anterior impaction did not demonstrate significantly different radiographic parameters post-operatively compared to pilon fractures without anterior impaction, which emphasizes that “normal” radiographic parameters can be achieved in these complex fractures. Further research is required to determine the most useful postoperative radiographic parameters and their correlation with clinical outcomes.