New Classification Based on CT and Its Value Evaluation for Fractures of the Lateral Process of the Talus.

This study aimed to develop a comprehensive classification system for fractures of the lateral process of the talus (LPTF) based on CT, and to evaluate its prognostic value, reliability and reproducibility. We retrospectively reviewed 42 patients involving LPTF with an average follow-up of 35.9 months for clinical and radiographic evaluations. In order to develop a comprehensive classification, a panel of experienced orthopedic surgeons discussed the cases. All fractures were classified according to Hawkins, McCrory-Bladin and new proposed classifications by 6 observers. The analysis of interobserver and intraobserver agreements was measured using kappa statistics. The new classification included 2 types based on presence of concomitant injuries or not, with type I consisting of 3 subtypes and type II of 5 subtypes. Average AOFAS score was 91.5 in the type Ia of new classification, 86 in type Ib, 90.5 in type Ic, 89 in type IIa, 76.7 in type IIb, 76.6 in type IIc, 91.3 in type IId, and 83.5 in type IIe. Interobserver and intraobserver reliability of the new classification system were almost perfect (κ = 0.776 and 0.837, respectively), showing a higher interobserver and intraobserver reliability compared to the Hawkins classification (κ 0.572 and 0.649, respectively) as well as McCrory-Bladin classification (κ = 0.582 and 0.685, respectively). The new classification system is a comprehensive one that takes into account concomitant injuries and shows good prognostic value with clinical outcomes. It is more reliable and reproducible and could be a useful tool for decision-making on treatment options for LPTF.


Introduction
Fracture of the lateral process of the talus (LTPF) occurs infrequently, accounting for 20% of talar fractures and 0.02% to 0.17% of all fractures in human body [1][2][3][4]. Due to the progressive expansion in recreational sport activities, mainly snowboard, their incidence has markedly increased [5,6]. For this reason, the LTPF has been termed a 'snowboarder's fracture' [7]. Despite its increasing incidence, the LTPF is di cult to diagnose and commonly overlooked on initial plain radiographs, as it may be subtle and di cult to visualize, with a misdiagnosis rate of 15% and up to 21% by radiographs alone [8]. Therefore, CT scan can be used to verify the LTPF and should be considered when more accurately diagnosing the LTPF [9,10].
In 1965, Hawkins initially classi ed LTPF into three types, simple fractures (Type I), comminuted fractures (Type II) and chip fractures (Type III) [11]. Based on Hawkins' classi cation, McCrory and Bladin reorganized and proposed a similar fracture classi cation system, but just in a different order [12]. After then, Tinner and Sommer subdivided type III of McCrory-Bladin classi cation into three subtypes, but didn't consider the concomitant injuries of LTPF [13]. All of the above classi cation systems are based on radiographs and possibly that they might underestimate the complexity of LTPF. It was estimated that, LTPF were associated with 19.6% of talar neck fractures and 24% of all talar body fractures [14,15]. LTPF may also be combined with fractures of talar head and posterior process [3]. However, by far as we know, these concomitant injuries have not been considered in any classi cation system regarding LTPF. Prompt and accurate diagnosis and classi cation of LTPF are crucial for optimized perioperative management [16]; moreover, detailed and extensive knowledge of concomitant injuries are conductive in surgical scheme making and contributed to obtaining of a better surgical outcomes. Sadly, up to date we are missing a comprehensive classi cation for the LTPF, which should include evaluation of LTPF and their concomitant injuries based on CT scan [17].
The objective of this study was to propose a comprehensive classi cation system for the LTPF based on CT and to evaluate its prognostic value, reliability and repeatability, which will help surgeons to develop treatment plans for patients with LTPFs.

Patients
The study was approved by the Ethics Committee of the Third Hospital of Hebei Medical University (K2015-001-12), according to the Helsinki Declaration. Written informed consent was obtained from all participants. We retrospectively reviewed 370 patients with talus fractures treated at the Third Hospital of Hebei Medical University between 2014 and 2020. Inclusion criteria to the study were: (1) older than 18 years; (2) fractures involving the lateral process of the talus; (3) preoperative and postoperative ankle plain radiographs available and (4) a preoperative CT scan available for review. Cases were excluded if they involved age less than 18 years, incomplete medical records or radiologic images, and lost to followup. A total of 42 consecutive patients (43 cases of LTPF, including one patient with bilateral fractures) were nally enrolled in our study. Demographical and clinical data such as sex, age, affected side, and accompanying injuries were collected from the electronic medical record, and the radiographs were evaluated by the Picture Archiving and Communication System. On plain radiograph the presence of complications like, non-union arthrosis, or malunion were assessed and clinical evaluation using AOFAS (The American Orthopaedic Foot and Ankle Society) score were performed at last follow-up. Based on this, we examined the localization and conformation of each fracture. Furthermore, the description of the fracture mentioned in the operative report was analyzed. In this context, fracture size, location, number, and comorbidities were documented. These cases were discussed jointly by three orthopedic surgeons, who are experienced in orthopedic trauma surgery. Within the scope of this case discussion, we arrived at a new classi cation and treatment recommendations.

Investigators and Survey
After fully understanding the classi cation system, four senior orthopaedic surgeons classi ed independently the chosen fractures (n 43) according to the Hawkins classi cation, the McCrory-Bladin classi cation and the new classi cation proposed. Among the four surgeons, two had more than 5 years of experience in orthopedic trauma surgery and two were senior residents. A second round of classifying was performed two weeks later, and the case order was scrambled using a random number generator.

Statistical analysis
The Spearman correlation coe cients between the AOFAS scores and the new classi cation, Hawkins classi cation and McCrory-Bladin classi cation were used to assess the prognostic value of each types of fracture we evaluated. Multivariate regression analysis was used to detect factors (age, sex, type of surgery, etc.) other than classi cations to predict AOFAS. Finally, the Generalized Linear Model (GLM) was developed with AOFAS as the dependent variable and classi cations as the covariate to compare the prediction strength of each classi cation. The partial ETA square was used to measure the prediction strength.
Interobserver reliability was evaluated to determine the reliability of the opinions of different observers for each case. By contrast, intraobserver reliability was evaluated to determine the reliability of individual observers by comparing the rst and second-round surveys for each case.
To determine the reliability of the classi cation systems we evaluated the interobserver agreement for each classi cation system using the Fleiss Kappa coe cient [18]. In order to evaluate the reproducibility of the classi cation system we calculated the intraobserver agreement using the Cohen Kappa coe cient [19]. Data analysis was conducted with IBM SPSS Statistics 21.0. The coe cients are interpreted using the Landis and Koch grading system [20], which de nes the reliability or reproducibility of κ values 0.2 as slight, between 0.2 and 0.4 as fair, between 0.4 and 0.6 as moderate, between 0.6 and 0.8 as substantial, and values > 0.8 as perfect.

Proposed Classi cation System
The new classi cation includes two types according to whether the LTPF is an isolated fracture or not.
Type I is an isolated fracture. This type is further divided into three subtypes ( Fig.1): Ia: chip fracture without inclusion of the talo bular joint.
Ib: simple fracture with involvement of the talo bular joint.
Ic: multiple fragment fracture with joint involvement.
Type II is a fracture of the lateral process of the talus combined with other parts of the ipsilateral talus, whether or not the articular surface is involved. This type is divided into ve subtypes ( Fig.2): IIa: lateral process fracture combined with talar head fracture.
IIb: lateral process fracture combined with talar neck fracture.
IIc: lateral process fracture with extension into the remainder of the talar body.
IId: lateral process fracture combined with talar posterior process fracture.
IIe: lateral process fracture combined with any two or more other fractures of the ipsilateral talus.

Baseline characteristics
A total of 42 patients were included in this study, of whom 37 were male and 5 were female, with a mean age of 34.9 years (range 18-65). Baseline data for all patients included are shown in Table 1. Among all 43 cases of LTPF, there were 9 fractures that were visible on CT but not on radiographs and the overall missed diagnosis rate was 20.9% (9/43  The average κ value among the three reviewers was 0.737 when classifying LTPFs using the Hawkins classi cation system, 0.748 when using McCrory-Bladin classi cation and 0.846 when using the new classi cation system ( Table 3). As can be seen, the new classi cation showed the highest interobserver reliability.

Intraobserver agreement
The κ value was 0.689 when classifying LTPFs using the Hawkins classi cation system, 0.714 when using the McCrory-Bladin classi cation and 0.823 when using the new classi cation proposed (Table 4).
Similarly, the new classi cation system has a higher intraobserver agreement than the other two classi cations.

Discussion
In this study, we proposed a new and comprehensive classi cation for LTPF based on CT scan. It was divided into two types according to whether the LTPF was an isolated fracture or not: type I was further divided into three subtypes and type II was divided into ve subtypes. This proposed new classi cation system is simple, comprehensive and easy to remember, and has higher prediction strength, inter-observer reliability, and intra-observer reliability, which we believe will provide useful information on treatment scheme making.
An ideal fracture classi cation system should be simple, comprehensive, reliable and reproducible [21].
Although various foot classi cation systems, such as the AO/OTA classi cation or Sneppen classi cation [22], do include LTPF, Hawkins as well as McCrory and Bladin are the only two classi cations speci cally focusing on LTPF [11,12,17]. Unfortunately, both classi cations are based on plain radiographs. The diagnosis may be di cult and underestimate the extent of such fractures, and the radiographs may be normal [23]. The main objective of the radiology report should be to convey the full range of fracture patterns rather than attempting to t complex LTPF into a limited classi cation scheme. Additional CT scan must be employed in every patient in whom clinical doubts exist, providing additional information on concomitant injuries or associated fractures [24], which is possibly bene cial for operative program making. Our study showed that although the currently used LTPF classi cation system (Hawkins and McCrory-Bladin) is simple enough, its main limitations are demonstrated by its lack of comprehensiveness, failure to take into account concomitant injuries, and exhibiting moderate to substantial reliability and reproducibility. Thus, it appears that an establishment of a classi cation based on CT scanning, which relates in particular to the concomitant injuries or associated fractures, is necessary.
Therefore, we proposed a new CT-based classi cation system for comprehensive typing of LTPF and evaluated the correlation between different types of fractures and clinical prognosis using the AOFAS score to check the prognostic value of the proposed classi cation. Not only did the new classi cation comprehensively cover the types of LTPFs, but our study also demonstrated the new classi cation system had a higher prognostic value. Furthermore, consistent with our hypothesis, the analysis results showed higher reliability and reproducibility of the new classi cation than currently used classi cation systems. While the Hawkins and McCrory-Bladin classi cation showed considerable substantial reliability and reproducibility, but the new classi cation showed a perfect reliability and reproducibility. This difference may be due to the use of a CT scan as a preoperative imaging modality, which permits a better identi cation of all fragments, their displacements and the areas of comminution.
LTPF is an injury that is often missed because most cases are considered trivial ankle sprains due to nonvisualization of fracture on radiographs, with a resultant overall misdiagnosis rate of 15% and up to 21% by radiographs alone [25,26]. In our study, there were 9 cases of LTPF that were visible on CT but not on radiographs and the overall missed diagnosis rate was 20.9% 9/43 , which was in line with previous literature. Missed or untreated LTPFs can potentially lead to permanent pain, healing in malposition, impingement syndrome, pseudoarthrosis development and also, due to joint instability, induced a potential development of severe subtalar arthritis [27]. The presence of normal radiographs creates di culty assessing the progress of the fracture. Jonathan recommended that both skeletal scintigraphy and CT scanning should be used in the resistant ankle sprain [9], and we believe that CT examinations are at least required for such injuries.
If an early and accurate diagnosis of the fracture pattern is made and adequate treatment is undertaken, the prognosis of LTPF would be favorable [28,29]. The lateral process of the talus possesses large articulations for both the distal bula and the posterior facet of the calcaneus. Fixation of LTPF is essential to restore the native architecture of these joint surfaces [30]. An attempt on open reduction and internal xation can potentially improve outcome and delay development of arthritis. For the small avulsion fracture or severe comminuted fracture which cannot be treated by open reduction and internal xation, conservative treatment or resection should be reserved [31]. Patients with type Ia and Ic of the proposed classi cation usually fall into this category. In our study, three patients with type Ia received non-operative treatment and one patient with type Ic underwent surgical excision of the fragment. The rest of the patients (including all type II patients) were treated with ORIF. The LTPFs of type II frequently extend into the midportion of talar body and neck of talus, and are often comminuted. Excision of a large number of fragments may result in an uneven joint surface or joint instability and is not recommended. Vallier suggested the fragments can be temporarily stabilized with Kirschner wires and then xed with mini-fragment screw and/or plate [32]. In patients with type II, it is usually most effective to reposition and stabilize the neck, body, and other fragments before incorporating the lateral process into the internal xation. The new classi cation provides a more comprehensive and practical method for classifying the LTPF, which can improve preoperative planning and subsequently the surgical treatment outcome.
Some possible limitations to this study should be acknowledged. First, the sample size of our study was improved compared to previous studies, however, due to the low incidence of LTPF, the available sample size remained small and the overall sample size was less than 50 patients. Second, this study has the intrinsic weakness of retrospective studies. A prospective study including clinical data of relatively large sample is necessitated to evaluate its clinical relevance. Third, in a clinical setting, it is critical to recognize LTPFs complicating soft tissue injuries, as they may require different treatment strategies. Due to the inadequate magnetic resonance imaging (MRI) in these patients, our study was too underpowered to determine the relationship between LTPFs and soft tissue injuries. This leaves room for further research.

Conclusions
CT scanning is an important adjunct in characterizing LTPFs to determine the most effective treatment. The new classi cation system is prognostic, reliable and reproducible and could be a useful instrument to well assist surgeons in selecting appropriate treatment strategies. Further studies on the evaluation of the classi cation are warranted.

Consent for publication
Not applicable.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Con ict of interest
Yuchuan Wang, Xiangtian Deng, Yanbin Zhu, Zhongzheng Wang, Lei Fu, Wei Chen, and Yingze Zhang declare that they have no con ict of interest.

Ethics approval and consent to participate
This study was approved by the ethics committee of the 3rd Hospital of Hebei Medical University (K2015-001-12). All patients of our study gave their written consent for participation and publication of their anonymized data.

Funding
This study was supported by the Innovation Project for Postgraduates of Hebei Province Education Department (grant number: CXZZBS2021083).

Contributions
Yingze Zhang conceived the idea for the study; Yuchuan Wang and Xiangtian Deng designed the study. Zhongzheng Wang, Yanbin Zhu, Xiangtian Deng and Wei Chen collected the relevant data and classi ed the fractures. Yuchuan Wang performed the statistical analyses and prepared the tables. Lei Fu did the drawings. All the authors interpreted the data and contributed to preparation of the manuscript. Table 1 Clinical details of 43 Fractures of the lateral process of the talus.   Lateral process fracture in combination with other fractures of the ipsilateral talus: IIa: lateral process fracture combined with talar head fracture; IIb: lateral process fracture combined with talar neck fracture; IIc: lateral process fracture with extension into the remainder of the talar body; IId: lateral process fracture combined with talar posterior process fracture; IIe: lateral process fracture combined with any two or more other fractures of the ipsilateral talus