How to choose infra-acetabular screw? Option of All-In or In-Out-in Screw Based on Computed Tomography Measurement of Infra-Acetabular Corridor

Background: The infra-acetabular corridor is quite narrow, which makes a challenge for the orthopedists to insert the screw. This study aimed to explore the relationship between the infra-acetabular corridor diameter (IACD) and the minimum thickness of medial acetabular wall(MTMAW), and to clarify the way of screw placement. Methods: The Computed tomography (CT) data of 100 normal adult pelvises (50 males and 50 females respectively) were collected and pelvis three-dimensional(3D) reconstruction was performed by using Mimics software and the 3D model was imported into Geomagic Studio software. The perspective of acetabulum was carried out orienting from iliopubic eminence to ischial tuberosity and the IACD was measured by placing virtual screws which was vertical to the corridor transverse section of "teardrop". The relationship between IACD and MTMAW was analyzed. When IACD was ≥ 5 mm, 3.5mm all-in screws were placed. When IACD was < 5 mm, 3.5mm in-out-in screws were placed. Results: The IACD of males and females were (6.15 ± 1.24) mm and (5.42 ± 1.01) mm and the MTMAW in males and females were (4.40 ± 1.23) mm and (3.60 ± 0.81)mm respectively. The IACD and MTMAW in males were signicantly wider than those of females (P < 0.05), and IACD was positively correlated with MTMAW (r = 0.859), the regression equation was IACD = 2.111 + 0.917 MTMAW. In the all-in screw group, 38 cases (76%) were males and 33 cases (66%) were females respectively. The entry point was located at posteromedial of the apex of iliopubic eminence, and the posterior distance and medial distance were (8.03±2.01)mm and (8.49±2.68)mm respectively in males. As for females, those were (8.68±2.35)mm and (8.87±2.79)mm respectively. In the in-out-in screw group, 12 cases (24%) were males and 17 cases (34%) were females, respectively. The posterior distance and medial distance between the entry point and the apex of iliopubic eminence were (10.49±2.58)mm and (6.17±1.84)mm respectively in males. As for females, those were (10.10±2.63)mm and (6.63±1.49)mm respectively. The angle between the infra-acetabular screw and the sagittal plane was medial inclination (0.42 ± 6.49) °in males, lateral inclination (8.09 ± 6.33) °in females, and the angle between the infra-acetabular screw and the coronal plane was posterior inclination (54.06 ± 7.37) °. Conclusions: The placement mode of the infra-acetabular screw(IAS) can be determined preoperatively by measuring the MTMAW in the CT axial layers. Compared with all-in screw, the in-out-in screw entry point was around 2mm outwards and backwards, and closer to true pelvic rim.

Introduction Acetabular fractures are intra-articular fractures. As for those fractures, open reduction and internal xation has been the gold standard to achieve anatomic or near-anatomic reduction of the articular surface, avoid complications of recumbency and return to pre-injury function as quickly as possible [1,2] . Periacetabular xation frame has been introduced by Culemann [3] for the anterior column fractures. Infra-acetabular screw closes the incomplete periacetabular xation frame which consists of both osseous columns, the ilioinguinal plate, and supra-acetabular screw xation. The entry point for the screw is 1 cm caudal of the eminentia iliopectinea in the mid-width of the pubic ramus and the target point is in ischial tuberosity. Infra-acetabular screw placed strictly parallel to the quadrilateral plate surface trans xes both columns, which signi cantly increases the xation strength of a standard plate xation for anterior column fractures. The indications of Infra-acetabular screw include anterior column, anterior column and posterior hemitransverse, T-type and both column fractures. The 3.5mm screws and reconstruction plates are commonly used for acetabular fractures, and the screw perpendicular to the fracture line is more effective and less traumatic [4,5] . Gras [6] reported that an infra-acetabular corridor(IAC) with a diameter of at least 5 mm was found in 93% of pelvises. The placement of a 3.5-mm cortical screw through the corridor was achievable. However, another study showed that over 20% of infra-acetabular corridors were not feasible for infra-acetabular screw placement even with the perfect reduction of fragments when treating acetabular fractures [7] . The corridor is adjoining with obturator never, obturator vessels and hip joint and the corridor size exists variance in races and sex, which could cause iatrogenic injury when operation is not proper [8] .
Therefore, the present study using the Mimics software and Geomagic Studio software simulated the placement of infraacetabular screw, measured the infra-acetabular corridor diameter and thickness of medial acetabular wall, clari ed the entry point, orientation, safety range and length of infra-acetabular screw, analyzed the relationship between infra-acetabular corridor diameter and minimum thickness of medial acetabular wall based on CT data to provide reference values for the placement of infra-acetabular screw.
Materials And Methods

Incusion and exclusion criteria
Inclusion criteria: 1. Pelvis and acetabulum were included in the Computed tomography examination.
3. Participator had no compulsive position while being scanned. tube voltage was 130kV, tube current was 100mA, layer thickness was 0.625mm and scanning matrix was 512×512.

Semi-pelvic model building
The DICOM format data were imported into Mimics 19.0 software (Materialise, Belgium) and 3D reconstruction of the pelvis was performed. The reconstructed 3D model of the pelvis was imported into the Geomagic Studio 2015 software (Geomagic, USA) in stereolithography (STL) format. According to the method proposed by Feng et al [9] , the triangular facets in the acetabular medullary cavity were removed to make the medullary cavity hollow. The processed model was imported into Mimics 19.0 software in STL format to match the original model.

Measurement of IACD and grouping
The prepared pelvis model was rotated to inlet view, then the perspective of acetabulum was carried out orienting from iliopubic eminence to ischial tuberosity in the Mimics software. The teardrop area encircled by the medial acetabular wall, the quadrilateral plate, and the posterior of obturator was demonstrated. We adjusted the pelvic model to make the medial of the ischial tuberosity and the roof of the acetabular fossa just overlap. Whereafter, the virtual screw was inserted perpendicular to the cross section of teardrop area. The maximum IACD [9] was achieved by gradually increasing the screw diameter when the screw just did not penetrate the cortex (Fig.1). We de ned the all-in screw as the full length of the screw was in the infra-acetabular corridor and the screw did not penetrate the cortex. Correspondingly, the in-out-in screw was de ned as 1/2 screw diameter exposed out of the cortex of quadrilateral plate. When IACD was ≥ 5.0 mm, a 3.5 mm diameter all-in screw (all-in screw group) was virtually inserted, and when IACD was <5.0 mm, a 3.5 mm diameter in-out-in screw was inserted (in-out-in screw group).

Measurement of MTMAW
We used the re-cut function of the Mimics software to cut the acetabulum from the superior to the inferior of the acetabulum with an interval of 1mm, which was perpendicular to the longitudinal axis of the human body. The thickness of the medial acetabular wall was measured in every layer to obtain the MTMAW (Fig.2), then the distance between the superior of the acetabulum and the layer of MTMAW was determined.
6. Parameter measurement of all-in screw When IACD was ≥ 5.0 mm, an all-in screw was virtually inserted. We reduced the screw diameter to 0.5mm, then the screw The angles between the virtual screw and the coronal and sagittal planes of the pelvis were measured in the standard anatomical position, and the optimum screw directions were recorded. When all-in screw model was chosen, the point O was xed, and point P was mobilizable, however the screw could not penetrate the cortex in any direction. We measured the angle between the virtual screw and the coronal and sagittal planes in the critical state to achieve the maximum safety range. In the same way, the maximum safety range of in-out-in screw were gained. We assumed that the angle was positive when the screw was oriented laterally in the sagittal plane, correspondingly the angle was negative when the screw was oriented medially (Fig.5).

Statistical analysis
All analyses were performed using SPSS software 23.0(SPSS Inc., Chicago, Illinois). The Shapiro-Wilk test was used to determine whether the data obeyed a normal distribution. The differences of measurement data were compared with independent sample t test. The rank sum test was used on the difference in the composition ratio of ranked data. Simple linear regression analysis was applied to analyze the relevance between the IACD and the MTMAW. The threshold for statistical signi cance was set at p smaller than 0.05.

Results
1.Sex-speci c differences in IACD and the relationship between the IACD and the MTMAW  The entry point was located at posteromedial of the apex of iliopubic eminence, and the posterior distance and medial distance were (10.49±2.58)mm and (6.17±1.84)mm respectively in males. As for females, those were (10.10±2.63)mm and (6.63±1.49)mm respectively. These differences did not reach statistical signi cance(p >0.05). Differences of distance between entry point and true pelvic rim and distance between entry point and the proximal perforation in the quadrilateral plate did not reach statistical signi cance either.

5.Comparison of optimum screw angle
There was no statistically signi cant difference in the angle between screw and coronal plane between genders in both groups, however, the difference in angle between screw and sagittal plane had statistical signi cance. There was no statistical signi cance on the difference in optimum screw angle between groups in same gender (Table.4). The screw safety angle range was wider in male in both groups, however, there was no statistical signi cance on the difference in safety range between groups in same gender (Table.5.).

Discussion
Open reduction and internal xation of acetabular fractures provide the most consistent functional results when an anatomic restoration and a congruent articular surface can be achieved [10] . Infra-acetabular screw rstly mentioned by Culemann [3] strengthens the xation dramatically by constituting periacetabular xation frame with both osseous columns, the ilioinguinal plate, and supra-acetabular screw xation. In several biomechanical analysis studies [11,12] , some authors found that additional placement of an infra-acetabular screw signi cantly increased the fracture xation strength and reduced the displacement of the fracture. However, compared with the acetabular anterior and posterior corridors, the infraacetabular corridor is signi cantly narrower, which may be even more signi cant for Asians. Additionally, infra-acetabular corridor is surrounded by obturator nerves and vessels [13] , which increases the risk of iatrogenic injury when screw is inserted. Therefore, it is essential to conduct a detailed study on the placement method and anatomical parameters of the infra-acetabular screw.

Placement method for infra-acetabular screw
In the original article by Gras et al [6] , they found the mean IACD of males and females were 7.7 mm and 6.9 mm respectively, the corridor with a diameter of at least 5 mm existed in 93% of cases (90% in females versus 94% in males) and the mean corridor length of males and females were 106.4mm and 96.2mm. However, in our study, the mean IACD of males and females were (5.15±1.25) mm and (4.42±1.01) mm respectively, the corridor with a diameter of at least 5 mm existed in 71% of cases (66% in females versus 76% in males), and the mean corridor length of males and females in all-in screw group were (99.43±4.04)mm and (88.83±3.71)mm and those data in in-out-in screw group were (98.19±4.37)mm and (87.31±4.48)mm. Obviously, the diameter and length of infra-acetabular corridor studied by Gras et al are greater than the results in our study, which may be caused by the racial difference. The infra-acetabular corridor is narrow and long and the safety range is limited. Therefore, tiny angular deviation may cause iatrogenic damage to the peripheral neurovascular bundles and hip joint [14] . Cai XH et al [15] have shown that when 1/3~1/2 of the screw diameter is located in the bone of the quadrilateral plate, it can effectively resist the separation of fracture fragment provide rigid xation and avoid the screw from penetrating the joint. In our study, we found that the minimum ICAD of males and females were respectively 4.09mm and 3.72mm, and it was di cult to insert 3.5mm screws in this situation. Whereupon we proposed that the screw placement method can be selected according to IACD. When IACD was ≥5mm the 3.5mm all-in screw was selected; When IACD was 5mm, we inserted the 3.5mm in-out-in screw and made 1/2 of the screw diameter exposed out of the quadrilateral plate cortex.
Arlt et al [16] found that the infra-acetabular corridor showed a double-cone shape with the isthmus located in the region of the acetabular fovea as the limiting anatomical structure. Additionally, the body weight, body height, and the diameter of Köhler's teardrop in the anteroposterior X-ray view showed signi cant positive correlations with the corridor volume. In our study, IACD had a positive correlation with MTMAW. Hence, based on the regression equation, we can calculate the IACD by the MTMAW which can be measured by CT scan, thereby the screw placement method can be selected preoperatively. When the MTMAW is greater than 3.15mm, all-in screw can be inserted.

infra-acetabular screw entry point and orientation
Culemann et al [3] described that the entry point for the infra-acetabular screw is 1 cm caudal of the iliopectineal eminence in the mid-width of the pubic ramus. Gras et al [4] measured the distance between anterior superior iliac spine and pubic symphysis and the distance between pubic symphysis and screw entry point, they found the mean ratio of these two distances is 1.36. Whereas these two methods are not utility in clinical practice. Baumann [17] found that the entry point is located at the posteromedial of apex of iliopubic eminence. The relationship between the entry point and the iliopubic eminence has no relevance with gender, age, or body type. In our study, the apex of iliopubic eminence and true pelvic rim were regarded as the reference point to measure the entry point, which could be con rmed by palpation [18] and was convenient to locate the entry point during the operation. We found that the all-in screw entry point was closer to the apex of iliopubic eminence than the results in the study by Baumann, which may be caused by the racial differences. The all-in screw exit point was roughly located at the medial of the middle ischial tuberosity. The infra-acetabular corridor was surrounding by obturator nerves and vessels, the in-out-in screw entry point was located at lateral of the obturator groove in order to avoid iatrogenic injury. Compared with all-in screws, the in-out-in screw entry point was round 2mm outwards and backwards, and closer to true pelvic rim (Fig.8). When in-out-in screw is applied, the periosteum dissector can be used to push away the obturator nerves and vessels to protect them. 90% of length of in-out-in screw in the sclerotin can ensure the stability the screw xation.
According to the data in our study, there is no statistical signi cance on the difference of optimum screw angle between groups in same gender. The angle between screw and the sagittal plane was more medial tilted in males than that in females in both groups, which may be caused by the morphological differences of pelvis in genders. The pelvis is "funnelshaped" in males, however that is "barrel-shaped" in females. The infra-acetabular corridor was nearly parallel to the sagittal plane and the medial inclination angle was round (0.42±6.49) ° in males. However, the angle was lateral inclined in females, which was round (8.09±6.33) °. Additionally, the screw was (54.06±7.37) ° posterior inclination with the coronal plane in both genders.

Method of intraoperative uoroscopy to determine the screw position
The inlet, outlet, iliac oblique and obturator oblique views are commonly used to verify whether the screw perforates the joint or cortex during the process of screw placement [19,20] . However, the anatomy peri infra-acetabular corridor is quite complicated, and the above-mentioned uoroscopic methods cannot fully assess whether the screw penetrates the joint, the quadrilateral plate and the posterior of obturator. If the screw posterior inclination is too small, it is easy for the screw to penetrate the posterior of obturator and damage the obturator nerve and vascular bundles. The infra-acetabular corridor is narrow and long and the safety range is limited. If the angle between the screw and the sagittal plane is too large during the screw placement process, there is a risk of penetrating the acetabular joint. In the original research by Culemann et al [3] carm rotated to the injured side and tilted 30°caudally in the obturator oblique and outlet views with patient supine for the control of correct screw path (Fig.9A). In our study, we trend to rotate the c-arm to the injured side and tilted 50°-55° cranially.
The infra-acetabular screw in males is nearly parallel to the sagittal plane, however it is slightly lateral tilted in females.
According to the orientation of screw, we adjust the position of c-arm to eliminate the angle between the screw and the coronal and sagittal planes. In this way, the screw becomes a dot in the perspective view. We can verify whether the screw penetrate the joint or quadrilateral plate through this view (Fig.9B).
The infra-acetabular corridor is quite narrow, which makes a challenge for the orthopedists to insert the screw. Hence, Gras et al [21] recommended inserting the infra-acetabular screw under the guidance of 3D navigation system. Lehmann et al [18] designed a new electromagnetic navigation system and completed infra-acetabular screw placement for 22 of 24 patients with the help of this new system. However, these technologies are demanding for the hospital facilities, which makes a di culty to widely apply around the world, especially in the developing countries. Therefore, if a portable infra-acetabular screw sighting device could be designed based on the results in this study, the safety and effectiveness of the operation will be improved.

Consent for publication
All patients involved in this study gave their consent for the anonymized data to be used for scienti c purposes and published in a scienti c journal.

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