Virtual mapping of 260 three-dimensional hemipelvises to analyse gender-specific differences in minimally invasive retrograde lag screw placement in the posterior acetabular column using the anterior pelvic and midsagittal plane as reference

Anatomic reduction and retention are pivotal in surgical therapy of acetabular fractures to prevent posttraumatic osteoarthritis. Open reduction and stabilization via ilioinguinal or Kocher-Langenbeck-approach and in rare cases extended approaches is the internationally accepted standard [ 1].

Minimally-invasive osteosynthesis approaches are limited due to the demanding reduction, which cannot always be completely achieved, and because of the complex three-dimensional geometry of the pelvis with limited bone corridors for screw placement. Fluoroscopy-assisted percutaneous screw osteosynthesis of the acetabulum may require several trials for correct screw placement which leads to high radiation doses [ 2].

Based on previous findings [ 3, 4] suggesting that there are gender-specific differences regarding the best implant position, we first defined an optimal screw position for the retrograde placed posterior column screw and then analysed the surrounding preformed bone corridor. In order to gain statistically relevant data without the limitation of a reduced sample size of previously performed studies [ 3, 5– 8], we decided to perform a virtual study using the technique described in 2010 [ 2]. We examined 260 hemipelvises virtually using data from CT scans of 130 uninjured European pelvises which were collected during clinical routine exams at our trauma centre. Our first hypothesis was that there are gender-specific differences considering the optimal implant position.

In a second step, we wanted to analyse, if there are landmark pelvic planes which could be used as reference, thus facilitating the visualization and screw placement for the surgeon. We have chosen the anterior pelvic plane as well as the midsagittal pelvic plane as reference planes since they are commonly used in hip arthroplasty [ 9]. We analysed the angle between the screw trajectory and these planes. Our second hypothesis was, that there are gender-specific differences considering the angles between these planes and the optimal screw position.

This study has been approved by the ethical committee of the University of Tuebingen (No 580/2012R). A patient collective of 260 uninjured hemipelvises was examined in regard to significant gender-specific differences relevant in the placement of the posterior column screw. Therefore, a CT-scan archive spanning from 06/2003 till 06/2010 was searched for exams of the pelvis to a standardized protocol (16 line helical CT Siemens Somatom Sensation 16, 120 kV, 160 mAs, slice thickness of 0.75 mm, collimation 0.75; Siemens AG, Munich, Germany). Patients with fractures of the acetabulum or pelvic ring, hip dysplasia or advanced osteoarthritis of the hip and age < 16 years or > 86 years were excluded as well as patients who underwent arthroplasty of the hip or osteosynthesis of the pelvic ring or acetabulum. Moreover, the study included only European patients.

Percutaneous stabilization as a minimally invasive treatment option is possible with non-displaced but instable fractures of the acetabulum with a risk of secondary displacement to allow early mobilization of the patient [ 10, 11]. Moreover, displaced acetabular fractures which can be reduced via lag screw are suitable for percutaneous stabilization [ 3]. Patients with an elevated perioperative risk from an open approach such as old patients or patients with several injuries in which a reduced approach related morbidity outweighs perfect anatomic fracture reduction, should also be considered for percutaneous stabilization. In addition to this, patients who will need total hip arthroplasty in the short run due to the configuration of the fracture might also benefit from earlier mobilization and protection of soft tissue when undergoing percutaneous stabilization instead of open reduction.

Percutaneous screw placement in the posterior column is difficult due to the complex three-dimensional pelvic geometry. Despite improved imaging with new fluoroscopic views of the acetabulum, 3D-CT scans and the intraoperative use of computer-assisted navigation, there is a danger of perforation the bone corridor or the acetabulum [ 2, 12]. Therefore, it is important that the surgeon visualizes the bone stock with the screw corridor in preoperative planning.

The present work defined reproducible screw positions in order to analyse the narrow bone corridor of the posterior column. The study was performed on CT-scans of 130 uninjured Europeans as described by [ 2]. The advantage of this method used in the present work is the fact, that all screws were virtually planned using advanced 3D planning software which is used in daily clinical routine, especially in neuro—and craniofacial surgery (iPlan® CMF 3.0, BrainLAB AG, Feldkirchen, Germany). One of the main focuses of the study was to investigate gender-specific differences in minimally invasive posterior column screw placement, especially in regard to reference planes which are clinically used in pelvic surgery. So far, there is only one study with this focus [ 3].

The present study showed that mean implant length in men was 141.8 mm and 127.2 mm in women with significant difference between men and women ( p < 0.001). Shahulhameed et al. in 2010 found a mean posterior column screw length of 138.5 mm for the whole collective of 11 pairs of adult cadaveric pelves; this is in accordance to our findings [ 8]. Mu et al. however found a mean screw length of 104.8 mm for the whole collective of 30 adult dried hemipelvis specimens [ 7]. The difference in screw length can be attributed to the fact, that Mu et al. placed antegrade screw whereas we used retrogradely placed screws in the posterior column. The percutaneous minimally invasive approach only allows for retrograde screw placement. We deliberately chose this approach since it offers the advantage of highly reduced soft tissue trauma compared to open approaches [ 1].

The entry point of the posterior column screw was located at mean 12.7 mm dorsal from the transition of ramus pubis inferior and ischial tuberosity in anterior-posterior direction and at mean 12.1 mm lateral of medial margin of ischial tuberosity in male patients. In female patients, the entry point was located significantly more medial than in male patients (mean: 10.8 mm). Similar gender-specific differences of the entry point could be found by Puchwein et al. [ 13] and by Dienstknecht et al. [ 3]. In another study, Dienstknecht also found a gender-specific entry point in posterior column screw placement when measuring the distance from the screw entry point to the landmarks ischial tuberosity, centre of symphysis, iliopectinal eminence and anterior inferior iliac spine [ 14].

The analysis of the anatomically preformed bone corridor of the posterior column was virtually performed using axial sections referenced to the screw axis according to the method described by Ebraheim et al. [ 6]. In order to account for the individual anatomy, the axial sections were adjusted to anatomical landmarks such as the walls of the acetabulum. Other workgroups did not take individual anatomical characteristics into account but used sections in defined distances along the screw axis [ 3, 5, 6]. Thus, in our study, two constrictions of crucial clinical importance in the bone stock of the anterior column could be identified. These constrictions were located at mean 49.6 mm (inferior margin of acetabulum) and 77.0 mm (centre of acetabulum) from the entry point of the implant in men and respectively 43.7 mm (men vs. women: p < 0.001) as well as 71.2 mm in women. Puchwein et al. analysed 50 polytrauma patients. Here, the mean distance between the narrowest zone and entry point was 54.8 mm (men) respectively 49.4 mm (women) [ 13]. The different findings in this study compared to the present study can be attributed to different measurement algorithms: To achieve interindividual comparability, in the present study, defined sectors along the screw axis according to the method described by Ebraheim et al. [ 6] were used, whereas Dienstknecht et al. measured the narrowest acetabular zone along the screw axis [ 3].

In our collective, the maximally possible implant diameter was at mean 12.9 mm in men and 11.2 mm in women. This means that the established use of 7.3 mm screws in men and women would have been save. Similar results were found by Banerjee in 2011 [ 15] and by Shahulhameed et al. in 2010 [ 8]. It is also in accordance to Attias et al. [ 5], Puchwein et al. [ 13] as well as Dienstknecht et al. [ 3].

As mentioned above we wanted to evolve from single osseous landmarks to established reference planes which can easily be identified intraoperatively; this is why we have chosen the anterior pelvic plane and the midsagittal plane as reference plane and measured the angles between these planes in different CT scans. The anterior pelvic plane [ 9] and the midsagittal plane are well known from hip arthroplasty [ 16– 21] and are commonly used and easily visualized in the OR. We measured the angle between the posterior column screw as well as the midsagittal plane and the anterior pelvic plane for each hemipelvis comparing the angles found in male and female patients.

Dienstknecht et al. also measured angles formed between the posterior column screw and several references lines and two different planes on a.p. radiographs. These reference lines were the tangent to the pubic tubercule, the caudal border of superior pubic ramus, the linea ilioischiadica and the tangent to acetabular rim. A defined vertical and horizontal plane were used as reference planes to measure the angles [ 3]; this approach has some limitations: The measurement was performed on standard a.p. radiographs, thus only in a two dimensional system compared to the three-dimensional CT scans used in the present study. Moreover, the planes chosen were defined as vertical and horizontal and are not further described. Even more important with respect to the used a.p. radiographs are the findings of Pullen et al. who could show in 2014 that the individual positional variability in pelvic standard radiographs can lead to large-magnitude changes in radiographic acetabular measures [ 22]. The present study eliminates these problems, since CT scans were used and the reference planes were defined by the pelvic bone landmarks and individually adopted for each pelvis examined. Moreover, the used planes are well established in daily clinical routine and are easily visualized both during procedure planning and in the OR.

However, there are limitations to the present study. The findings are limited to retrograde lag screw placement in the posterior acetabular column and the study has been performed in healthy individuals with no fracture present. Moreover, only European patients were examined, so the findings may be limited for use in Europeans. The effectiveness of the described technique in regard to safe screw placement has to be proved in clinical use.

The anatomical differences in the pelvic geometry of men and women lead to gender-specific differences in preformed bone corridor, implant length and diameter as well as angles formed between screw trajectory of posterior acetabular column screw and the reference planes (anterior pelvic and midsagittal plane). The angles and the reference planes are reproducible and can be easily visualized in the OR, so they can be helpful in minimally invasive retrograde screw placement in the posterior column. The effectiveness of the described technique in regard to safe screw placement has to be proved in clinical use.