Summary

Tomographic analysis of mineral density and cortical thickness

Abstract

Introduction

When it comes to complex reverse shoulder arthroplasty cases, reconstructing the glenoid bone loss to achieve an optimal baseplate position is crucial. From this perspective, metal-augmented glenoid baseplates and Bony increased-offset (BIO) RSA using humeral head or other donner site autograft or allograft have been developed to compensate for the glenoid bone loss. The use of bone tissue bank frozen allografts implies the possibility of using the amount of graft as needed without adding morbidity to the patient. However they are not exempt from complications, such as allograft fractures, non-union or graft resorption that could lead to glenoid component failure. It could be assumed that the better the bone mineral quality and the greater the cortical thickness of the allograft used in a BIO RSA, the better its biomechanical characteristics. Therefore, the objective of this study was to evaluate which areas of the proximal femur frozen allograft present the best biomechanical conditions to eventually be used as a structural graft in a BIO RSA.

Methods

This study prospectively analyzed a consecutive series of cadaveric femurs from donors admitted to our institution’s bone bank. Cadaveric bone grafts were frozen at -70°C for 4 weeks. Subsequently, they underwent irradiation with Cobalt-60, delivering a dosage of 25 kGy to eradicate any remaining pathogens. They were stored in separate plastic bags to perform a multislice CT scan. The femoral head and neck were divided into 3 thirds from proximal to distal. Each third was divided into 4 quadrants. Then, the mineral density of each quadrant, the thickness of its respective cortex, and the total diameter of the analyzed surface were recorded.

Results

After determining the sample size, forty specimens of donors aged more than 18 years old were analysed. The humeral head was the area with the largest surface area (15.9 +- 4.81mm²), exceeding by almost 20% the second (distal third of the head). The section with the tickest cortical was the distal neck (4.06 +- 1.57 mm). Within this section, the quadrant with the greatest cortical thickness was the posteroinferior with 6.18 +-2.15mm, followed by the anteroinferior with 4.86 +- 1.83mm. The proximal third of the head had the highest density, with a value of 373 ± 64.8 Hounsfield Units (HU). the anterosuperior and posterosuperior sectors of the femoral head recorded the highest values, followed by the posteroinferior section of the proximal third of the neck (460 ± 104 HU, 426 ± 69.1 HU, and 422 ± 149.9 HU, respectively). In contrast, the middle and distal thirds of the neck were the sections with lower density, with values ​​of 216 ± 107.3 HU and 220 ± 97.6 HU, respectively.

Conclusion

Although the distal third of the femoral neck was the area with the greatest cortical thickness, its bone mineral density was the lowest. On the contrary, the proximal third of the humeral head showed the highest bone density and was also one of the areas with the greatest cortical thickness. Of this, the upper quadrants had the best results in all the variables analyzed. According to these results, we recommend the use of the proximal third of the femoral head for allografts in a bioRSA. Additionally, in cases when a wedge to cover an eccentric defect is necessary, we recommend placing the two superior quadrants in the defect area since they have the best ratio of mineral density and cortical thickness.