Effect of screw torque level on cortical bone pullout strength

The objectives of this study were 2-fold: (1) to perform detailed analysis of cortical screw tightening stiffness during automated insertion, and (2) to determine the effect of 3 torque levels on the holding strength of the bone surrounding the screw threads as assessed by screw pullout.

Ten pairs of ovine tibiae were used with 3 test sites spaced 20 mm apart centered along the shaft. One side of each pair was used for measuring ultimate failure torque (Tmax). These Tmax and bone-density values were used to predict Tmax at contralateral tibia sites. Screws were inserted and tightened to 50%, 70%, and 90% of predicted Tmax at the contralateral sites to encompass the average clinical level of torque (86% Tmax). Pullout tests were performed and maximum force values were normalized by cortical thickness.

Torque to failure tests indicated tightening to 86% Tmax occurs after yield and leads to an average 51% loss in stiffness. Normalized pullout strength for screws tightened to 50% Tmax, 70% Tmax, and 90% Tmax were 2525 +/- 244, 2707 +/- 280, and 2344 +/- 346 N, respectively, with a significant difference between 70% Tmax and 90% Tmax groups (P < 0.05).

Within the limitations of our study involving the testing of 1 type of screw purchase in ovine tibiae, results demonstrate that clinical levels of lag screw tightening (86% Tmax) are past the yield point of bone. Tightening to these high torque levels can cause damage leading to compromised holding strength. Further research is still required to establish the appropriate level of torque required for achieving optimal fracture fixation and healing.