A load-to-failure test was used to study the biomechanical properties of scaral screw fixation in human cadaveric specimens. The goals of this study were 1) to determine the effects of the two commonly chosen sacral screw orientations of fixation characteristics; 2) to determine the effects of selected screw-instrumentation linkages on the biomechanics of sacral screw fixation; 3) to correlate the biomechanical properties with a noninvasive assessment of sacral bone density; and 4) to corralate the torque during screw insertion with these biomechanical properties. The bone density of each specimen was measured with quantitative computed tomography. A screw was inserted from the dorsal surface either anteromedially or anterolaterally into the body of S 1, and the torque needed to insert each screw was measured. The screw head was attached to a constrained or semiconstrained loading linkage. Force was applied to the screw in an inferior direction until the maximum load was achieved. The maximum load, screw translation, rotation at maximum load, and initial compliance of the bone-screw interface were determined. It was found that the anteromedial screw orientation, combined with a rigidly constrained loading linkage, resulted in the greatest maximum load to failure, the least screw rotation, and the least initial compliance of the four groups studied. The maximum load and the initial stiffness of bone-screw fixation increased signficantly with bone density. Torque measurements correlated significantly with maximum load to failure, initial interface stiffness, and bone density. It was therefore concluded that bone density and torque measurements can be useful in assessing sacral screw fixation.