Morphological Characteristics of the Posterior Wall Associated with Complex Acetabular Fractures: A Radiological Study Using 3D Software and Fracture Mapping Technique

Background The aim of the study was to compare the morphological distinctions of the posterior wall (PW) in different complex acetabular fractures using 3D software and fracture mapping technique and ultimately to provide for improved clinical treatment. Methods One hundred and fourteen patients with complex acetabular fracture associated with PW were recruited. All patients were divided into two groups according to the injury mechanism of the PW: Group A (both-column and PW) and Group B (including posterior column and PW; T shape and PW; and transverse and PW). Fracture mapping was generated on the intra- and extrasurface of a standard template. The radiological parameters including spatial displacement, articular surface area, articular range, marginal impaction, and multifragments of the two groups were compared. Results The spatial displacement, intra-/extra-articular surface area, and start and end point in Group A were 10.9 mm (IQR, 8.4-15.2), 8.2 ± 2.6 cm2, 17.9 ± 5.3 cm2, 0.8° (IQR, -6.0-16.2), and 107.5° (IQR, 97.2-116.9), respectively. The results in Group B were 30.4 mm (IQR, 16.8-48.7), 4.1 ± 2.0 cm2, 10.6 ± 4.4 cm2, 29.5° (IQR, 19.2-38.0), and 117.5° (IQR, 98.2-127.2), respectively. Marginal impaction was defined by Letournel et al. All the differences between two groups were significant (P < 0.05). The fracture map in Group A showed an “L”-shaped pattern and a “cusp” on the ilium, and the PW was located at 1/5 to 1/4 of the posterosuperior part of the acetabulum. The fracture maps in Group B were scattered and lacked consistency, and the PWs were confined to 1/10 to 1/8 of the posterior acetabulum. Conclusions Quantitative measurements and fracture mapping represented the differences in morphological characteristics of PWs associated with complex acetabular fractures.


Introduction
Fractures of the posterior wall (PW) are the most common types of the acetabulum [1,2]. It has been reported that the PW fractures may be isolated or associated with other acetabular injuries, including posterior column, T shape, transverse, and both-column fractures [3,4]. Similar to acetabular isolated PW fractures, the first three PW fractures mentioned above are all caused by a direct strike posteriorly of the femoral head, which often involves marginal impac-tion, multifragments, and lesions of the femoral head [5]. In contrast, the injury mechanism of PW fracture in bothcolumn fractures is created by a "pull-type" mechanism and often a large-sized fragment [3,6]. Although the differences between the two types of PW fractures have been reported in recent literature [3,7,8], to our knowledge, there are no quantitative morphological comparisons of two kinds of PW associated with complex acetabular fractures.
As computed tomography (CT) processing software has emerged, an imaging technique called "fracture mapping"   [9] was utilized to reveal the distribution of fractures and guide surgical planning. In the past decade, the combination of 2D/3D CT and fracture mapping has also been proven to be effective in understanding the injury mechanism and fracture classification and morphology [10,11]. In addition, this technique was widely used in the acetabulum. Yang et al. [12,13] established a frequency map of acetabular both-column fractures and quadrilateral fractures to help surgeons gain insights into the surgery. Cho et al. [14] evaluated the characteristics of isolated PW fractures of the acetabulum with a cohort of 51 patients to show fracture patterns and recurrent fracture zones. The aim of the study was to compare the morphological distinctions of PW in different complex acetabular fractures using 3D software and fracture mapping technique and ultimately to provide for improved clinical treatment.

Subjects.
A retrospective study of patients with acetabular fractures was performed at our level I trauma center between September 2015 and April 2020. The inclusion criteria were as follows: (1) age of 18 years or older, (2) complex acetabular fractures with PW involvement, and (3) pelvic CT images with high quality and those with slice thicknesses of <1.5 mm. Exclusion criteria included (1) bilateral acetabular injuries, (2) sacral fracture and/or sacroiliac dislocation, and (3) pathological fractures. A total of 114 patients were identified in the picture archiving and communication system (PACS) database and classified by two trained orthopedic trauma physicians. The detailed inclusion and exclusion process is shown in Figure 1. All patients were divided into two groups according to the injury mechanism of PW: Group A (both-column and posterior wall, BC +PW) and Group B (including posterior column and posterior wall, PC+PW; T shape and posterior wall, T+PW; and transverse and posterior wall, TV+PW).  [12,15], a 3D model reconstruction was produced in which different fracture segments were distinguished in different colors. The 3D models were exported to 3-matic 12.0 software (Materialise Inc., Belgium). Using the sacrum and contralateral hip as a template, a mirrored hemipelvis was generated. All segments were manually reduced and automatically calibrated by "registration" to finish the reduction of the fracture.

Fracture Mapping.
A 3D fracture mapping technique previously described by Xie et al. [11] was used to best match a standard acetabular template. Smooth curves were drawn precisely on the intra-and extrasurface of the template. All the fracture lines were superimposed on the same template to show the distribution of fracture lines in PW ( Figure 2).

Parameter Measurement.
To compare the different types of PWs quantitatively, we measured the following radiological parameters using Mimics 20.0 and 3-matic 12.0 software.
Spatial displacement: this measure was defined as the furthest 3D distance of a point on the acetabular rim of the PW before and after reduction.
Articular surface area: the intra-and extra-articular surface area of the PW was measured.
Articular range: as noted in the previous literature [14], a clock was drawn in the lateral view. The articular range of the PW was defined as the range between the start and end points on the rim and expressed in angle (°) (Figure 3).
In addition, marginal impaction and multifragments were recorded in each group.

Data
Analysis. The analysis of the PW fracture maps was descriptive. All the data were processed using SPSS 23.0 (SPSS Inc., Chicago, IL, USA). Independent t-tests and Mann-Whitney U tests were used for the comparison of continuous variables. Differences in categorical variables igure 3: The clock was drawn in lateral view. The inferior point of the acetabulum was defined as +180°, and the superior point was 0°.

Results
The patient demographic characteristics and classifications are shown in Table 1.
3.1. Radiological Parameter Measurements. All parameters are presented as the median (IQR), the mean and standard deviation, or the proportion in Table 2  2), respectively. The differences were statistically significant between patients in the two groups (P < 0:05).
Regarding marginal impaction, several patients in Group B had marginal impaction (18, 26%), whereas no patients in Group A had marginal impaction (P < 0:05). The differences in multifragments of patients in Group A (3, 7%) and Group B (36, 52%) were statistically significant (P < 0:05).    (Figure 4(a)). The high density of the two corridors formed an "L"shaped pattern ( Figure 5). All the fracture lines extended upward beyond the level of the greater sciatic notch, continued upward toward the iliac crest in 34 patients (76%), and oriented around the anterior superior spine anteriorly in 11 patients (24%). The highest point of the PW formed a "cusp" on the outer table of the ilium ( Figure 5). Moreover, the intra-articular fracture line revealed that the majority of PWs were located at the posterosuperior part and occupied approximately 1/5 to 1/4 of the acetabular fossa (Figure 4(a)). The extra-articular fracture maps in Group B demonstrated that the majority of fracture lines were concentrated in the middle-lateral 3/4 of the PW; these fracture lines were scattered and lacked consistency (Figures 4(b)-4(d)). In addition, the main fracture lines were concentrated in 1/10 to 1/8 of the posterior acetabulum (Figures 4(b)-4(d)).

Discussion
Previous studies have widely reported radiological findings of the PW of the acetabulum. Some investigators described various methods for measuring the size and features of the PW to predict hip stability [3,[16][17][18][19][20]. However, these methods were confined to 2D and did not provide a 3D evaluation. A novel fracture mapping technique was applied by Cho et al. [14] to evaluate the features of isolated PW fragments, while PWs associated with complex acetabular fracture were not included. To our knowledge, this is the first work to compare the morphological characteristics of PWs associated with complex acetabular fractures using a combination of fracture mapping and 3D software.
The differences in injury mechanisms resulted in the differences in the PWs observed between patients in the two groups. Both-column fractures frequently accompanied central dislocation of the femoral head. The force was transmitted medially to the main fragments, and the PW fracture was created by a "pull-type" mechanism anteromedially (Figures 6(a) and 6(b)), while the joint capsule was intact [3,6,21]. In contrast, the PW fractures in Group B were similar to acetabular isolated PW fractures, which were caused by a direct impact of the femoral head, posteriorly [4] (Figure 6(c)).
Except for the displacement direction of the PW, the displacement distances of Group A were significantly smaller than those of Group B (P < 0:05). Secondary congruency of the acetabulum and femoral head is a characteristic of acetabular both-column fractures in Group A, which means that the joint contact stress is evenly distributed over the entire articular surface [12]. And the displacement of PW in Group A disappeared following the reduction of the anterior column, quadrilateral plate, and femoral head [22]; this occurrence is known as a "congruency reduction." However, after reduction of the main fragments (posterior column, T shape, and transverse) and femoral head in Group B, the vis-ible displacement of the PW remained and required additional posterior fixation. It should be mentioned that displacements in this study were 3D distances, which were different from the measurements of the fracture gaps noted in radiographs or CT scans in other studies [23]. Therefore, the results reported in this study are quite different from those of other studies [24].
The articular surface area and fracture map demonstrated that the PW in BC fractures were often a single large-size fragment. The main PWs in Group A occupied 1/5 to 1/4 of the acetabular fossa, in which the intraarticular surface area accounted for nearly 22% of the acetabulum. In addition, the results showed that the multifragments in Group A accounted for 3 of 45 cases (7%). All 3 cases were two-part intact fragments, and the fracture line did not affect the articular surface ( Figure 7). In contrast, the smaller size in Group B took up approximately 1/10 to 1/8 of the acetabulum, which the intra-articular surface area was nearly 11% of the acetabular surface. Also, scattered fracture lines showed that 52% (36/69) patients had multifragments, even accompanied by intra-articular comminution of the PWs [4,5], which was a related predictor for poor clinical outcomes [25].
Furthermore, the location characteristics of the two types of PWs were also different. The articular range and fracture map reflected the PWs in Group A at a higher location, even involving the roof wall [26]. Eighty-seven percent (39/45) of the start points were between -25°and 25°, and 91% (41/45) of the end points were between 70°and 120°, whereas, in Group B, 91% (63/69) of the start points were between 0°a nd 50°, and 81% (56/69) of the end points were between 90°and 140°. This meant that the majority of the PWs in Group B were mainly confined below the roof [27], as described by Letournel and Judet. However, due to the uncertainty regarding the term "roof" [14] and obvious grouping differences between articles [14,27], we did not divide the PWs of Group B into subgroups (posterior, posterior-superior, and posterior-inferior). Additionally, the fracture map of Group A elucidated a "cusp" of the PW on the outer table of the ilium. The "cusp" of the PW, which can be regarded as an auxiliary marker for diagnosing BC+PW fractures, was often obvious in the obturator-oblique view and was named the "antispur" sign [22] (Figure 8). In Group A, 76% fracture lines of PW extended upward toward the iliac crest and 24% oriented around the anterior superior spine anteriorly. This result demonstrated the morphological distinction of PWs in different AO/OTA subgroups (C1.3 and C2.3, respectively) and the incidence of the two subtypes, similar to the results revealed by Yin et al. [28].
Marginal impaction was defined by Cho et al. and e Souza et al. as a rotated, impacted fracture with depression of the osteochondral fragments into the underlying cancellous bone [14,29], which was regarded as an independent risk factor for the development of arthritis and early clinical failure [5,29,30]. Our studies showed that 26% of "posterior-type" PWs had marginal impaction, which was similar to other results (21.5%-41.4%) [27,31,32]. However, because there was no direct impact on the surroundings of PWs in BC fractures, no marginal impactions were found in Group A.
All qualitative and quantitative morphological comparisons theoretically revealed the differences in posthip stability [33] and clinical prognosis [30] of the two types of PW. Combined with previous studies and our results, recommendations for the treatment of PW associated complex acetabular fractures were proposed. BC+PW fractures mainly involve the anterior stress of acetabulum and can be managed by a single ilioinguinal or iliac fossa combined with the Stoppa approach [8,26,34]. After reduction and fixation of the two columns, the posthip stability was evaluated using intraoperative dynamic stress examination [35]. Nonfixation  BioMed Research International or antegrade lag-screw fixation was determined based on evaluation results, and an additional Kocher-Langenbeck (KL) approach was unnecessary. The KL approach was obviously performed for the PWs in PC+PW, T+PW, and TV +PW. Furthermore, the intra-articular fragments and marginal impaction must be identified carefully and stabilized appropriately because these are the known factors affecting clinical outcomes [2]. This retrospective study had several limitations. First, there was a relatively small sample size gathered from one database. The paradox is that additional cases can improve the accuracy of the measured results; however, the increased fracture lines may lead to a complexity of mapping that is difficult to expound. Second, factors such as age, sex, and the force of impact endured upon injury were not taken into account. Third, because of the anatomical differences, some pelvis did not match the standard template perfectly.
In conclusion, 3D radiological measurements and fracture mapping represented the characteristics of PWs associated with complex acetabular fractures. There were significant differences among the three "posterior-type" PWs (posterior column, T shape, and transverse) and the PW associated with both-column fractures; also, the treatment for PWs should be different.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.

Ethical Approval
The study was approved by the Institutional Review Board of the Third Hospital of Hebei Medical University.

Conflicts of Interest
All authors declare that they have no conflict of interest.

Authors' Contributions
Siyu Tian and Shaobo Liang contributed equally to this work.