Should cannulated screws be removed after a femoral neck fracture has healed, and how? A finite element analysis of the femur before and after cannulated screw removal.

Abstract

The removal of three inverted triangular parallel cannulated screws after the femoral neck fracture healing is sometimes accompanied by osteonecrosis of the femoral head (ONFH) or its refracture. The purpose of this study was to determine the biomechanical changes of the femur before and after the screw removal using a finite element analysis. The CT data of the femurs were obtained from a 69-year-old healthy female to establish the femur models. The established finite element models consisted of N, C, and R groups: N group, the normal femur; C group (to simulate the femoral neck fracture healing after the internal fixation), the normal femur with three inverted triangular parallel cannulated screws (C1) or with two upper parallel screws (C2). For the R1 or R2 groups, the screws in the model of the C1 or C2 individuals were gradually removed in seven or three types of different screw combinations, respectively. The stresses and displacements of the femur were determined. (1) Compared with the N group, a uniform stress distribution was stopped by the addition of three screws in the C1 group; in contrast, all screw removals resulted in the stress concentration on the screw holes and the disappearance of stress shielding. Moreover, the average stress of the femoral head in C1 group increased by 37.85%, while that of the femoral neck decreased by 23.03%. (2) Compared with the C1 group, while only the lowest femoral calcar screw was removed, there was a similar stress distribution in the proximal femur, and the average stress of the femoral head increased only by 0.35% although that of the femoral neck increased by 63.62%; however, removal by any other means resulted in a significant stress concentration in the proximal femur and a significant increase in the average stresses of the femoral head and neck (5.96-40.95% and 12.82-75.46%, respectively). (3) Compared with the N or C1 group, there was a significant stress concentration on the screws and its surrounding proximal femur in the C2 group. (4) Compared with the C2 group, the simultaneous removal of two upper screws not only did not cause a significant stress concentration on the proximal femur but also significantly reduced the average stresses of the femoral head and neck by 87.49% and 65.51% respectively. On the contrary, the gradual removal of two screws caused a significant stress concentration on the screw and its surrounding proximal femur although the average stresses of the femoral head and neck decreased by 88.79-89.06% and 67.00-67.22%, respectively. (5) Compared with the N group, the average displacements of the femoral head and neck in the C1 group increased only by 3.12% and 3.37%, respectively. Additionally, compared with the C1 group, while three, two, or one screw was simultaneously removed, the average displacements of the femoral head and neck only fluctuated - 5.51-1.65% and 1.78-9.03%, respectively. Residual internal fixation after femoral neck fracture healing may lead to stress concentration on screws and stress shielding around screws. The first removal of the lowest femoral calcar screw and then the second removal of two upper residual screws have a minimum effect on the stress concentration on the proximal femur and the average stress on the femoral head. The incorrect screw removal and resulting femoral load may well be closely related to occurrence in ONFH or its refracture.