Introduction: The development of the ilioinguinal approach by the pioneering work of Letournel in 1965 has transformed the treatment of acetabular fractures. To date, this approach has been well established and few modifications have been described of the original approach. However, this approach is difficult, takes long time for exposure and closure of abdominal layers. The aim of this article is to report a modification of the approach which the authors have found particularly useful.

Material and Method: Cadaveric dissection showed that it was easier to detach the inguinal ligament from the anterior superior iliac spine and reflect the anterior abdominal wall as one layer, than by the classical approach through layers of the anterior abdominal wall. Closure was also simpler, in the cadaver, with the entire anterior abdominal wall reattaching satisfactorily by a single transosseous suture. The rest of the approach, including division of iliopectineal fascia and developing the three windows remains the same as in the original approach.

Results: To date, three acetabular fractures have been reduced and fixed using this modification. Besides substantially speeding up the exposure and closure, this approach allows superior distal visualization of the anterior column and wall, and the impression of rather less bleeding. No complications developed with the three cases treated through this modified approach, and specifically, there have been no hernias, nor has the lateral femoral cutaneous nerve of the thigh been damaged.

Discussion: This small study demonstrates a modification to a classic approach that seems to be both safe and fast We hope that further experience will also show reduction in problems associated with wound healing. Further work in progress may also show that femoral venous flow is less impeded by this approach as retraction is not against the unyielding inguinal ligament.

Various frames of reference are routinely used for hip and knee arthroplasty. We hypothesised that the linea aspera is a constant anatomical feature which can be used as a frame of reference.

Twenty cadaveric femora were CT scanned with high resolution 1mm slices. Robin 3D software was used to manipulate the CT data. Three points were identified on the posterior aspect of the lesser trochanter, medial and lateral femoral condyles to position the femora in similar positions based on the posterior femoral plane (PFP). Centres of the femoral head and neck were derived by surface markers placed on the head and around the neck respectively. Joining the 2 centres gave head neck axis (HNA). The most prominent point on the linea aspera was identified at a level midway along the length of the femur. At that level the centre of the canal was derived by placing surface markers. Joining the most prominent point on the linea aspera to the centre of the canal identified our plane, linea aspera – centre plane (LCP). Angle measurements were made between PFP to HNA, PFP to LCP and LCP to HNA.

PFP to HNA is the traditional method for measuring anteversion angle which in our series had a mean of 13°, SD of 5 (range 5–24). PFP to LCP gave very similar results with mean 101°, SD 6 (range 92–112). However it was noted that there is weak correlation between PFP to HNA angle and PFP to LCP angle for each femur. LCP to HNA measurements were more variable with mean 89°, SD 8 (range 76–108).

From these data we conclude that the proximal half of the femur has more variable torsion compared to the distal half. This study shows that the linea aspera should not be used as a frame of reference for hip nor knee arthroplasties. However, further studies are needed to evaluate the linea aspera in-vivo where it is expected to be more prominent and easier to identify.

The use of intramedullary column screws in the treatment of acetabular fractures is becoming more widely utilized. The development of percutaneous methods to insert these screws under image intensifier guidance is one of the main reasons for their increased use. Few groups are navigating insertion of these screws. The available screws are cannulated 6.5–8 mm screws. Most surgeons prefer using 3.2 mm guide wires to reduce deflection. With a shank diameter of 4.5 mm, 3.2 mm cannulation significantly weakens the screws. We postulated that both columns, specially the posterior column can accommodate larger screw diameters which will increase the stability of fixation allowing earlier full weight bearing. The currently used screws were designed for fixation of femoral neck fractures. As percutaneous fixation of acetabular fractures is a growing area of interest, this warrants designing suitable screws with larger diameters.

Eight CT scans of the adult pelvis –performed for non fracture related indications-, were studied (7 females, 1 male). We found that the anatomical cross-section of the columns is irregular but approximately triangular. The method we used to determine the largest diameter of a screw to fit each column was fitting cylinders in the columns. Robin’s 3D software was used to segment acetabula and convert the CT data into polygon mesh (stereolithography STL format) bone surfaces at an appropriate Hounsfield value. The resulting STL files were imported in Robin’s Cloud software, where polygon mesh cylinders of 10 mm diameter were fitted in each column. These cylinders were then manipulated to achieve best fit and their diameters were gradually increased to the biggest diameter which still fitted in the column.

The mean diameters of the fitted cylinders were 10.8 mm (range: 10–13mm) and 15.2 mm (range 14–16.5mm) for the anterior and posterior columns respectively.

To our knowledge, this is the first investigation to study the cross sectional dimensions of the anterior and posterior columns of the acetabulum. Our small sample shows that both columns can safely accommodate larger screws than those currently used. We plan to investigate this further using cadavers.

Various frames of reference are routinely used for hip and knee arthroplasty. We hypothesised that the linea aspera is a constant anatomical feature which can be used as a frame of reference.

Twenty cadaveric femora were CT scanned with high resolution 1mm slices. Robin 3D software was used to manipulate the CT data. Three points were identified on the posterior aspect of the lesser trochanter, medial and lateral femoral condyles to position the femora in similar positions based on the posterior femoral plane (PFP). Centres of the femoral head and neck were derived by surface markers placed on the head and around the neck respectively. Joining the 2 centres gave head neck axis (HNA). The most prominent point on the linea aspera was identified at a level midway along the length of the femur. At that level the centre of the canal was derived by placing surface markers. Joining the most prominent point on the linea aspera to the centre of the canal identified our plane, linea aspera – centre plane (LCP). Angle measurements were made between PFP to HNA, PFP to LCP and LCP to HNA.

PFP to HNA is the traditional method for measuring anteversion angle which in our series had a mean of 13°, SD of 5 (range 5–24). PFP to LCP gave very similar results with mean 101°, SD 6 (range 92–112). However it was noted that there is weak correlation between PFP to HNA angle and PFP to LCP angle for each femur. LCP to HNA measurements were more variable with mean 89°, SD 8 (range 76–108).

From these data we conclude that the proximal half of the femur has more variable torsion compared to the distal half. This study shows that the linea aspera should not be used as a frame of reference for hip nor knee arthroplasties. However, further studies are needed to evaluate the linea aspera in-vivo where it is expected to be more prominent and easier to identify.