Background Screw fixation of scaphoid fractures has gained popularity. A long central screw has been shown to be biomechanically advantageous. Questions/purposes We compared the ability of different screw designs to obtain this goal and determined the influence of sex and approach on screw length. Methods We performed all measurements on threedimensional reconstructions of 20 CT scans of normal scaphoids (10 men and 10 women) with the use of software. The three-dimensional computer models were analyzed, the central axis was defined, and the screws were placed along this axis. We compared 15 different available screw designs and volar and dorsal screw placement. Results The length of the scaphoid along its central axis was longer in men (mean, 27.14 mm; standard error of the mean, 0.97 mm) than in women (mean, 23.86 mm; standard error of the mean, 0.37 mm). The screw length that can be used was longer in the volar approach (mean, 23.72 mm; standard error of the mean, 0.19 mm) than in the dorsal approach (mean, 23.31 mm; standard error of the mean, 0.19 mm) regardless of the screw design. Screws with a trailing thread diameter greater than 3.9 mm and leading thread diameter greater than 3.0 mm were shorter. Conclusions Scaphoids in women are smaller than in men. Theoretically, fixation of scaphoid fractures through a volar approach will allow the surgeon to use longer screws. The screw design has a significant influence on the screw length that can be used in scaphoid fracture fixation. We recommend using a differential pitch screw with a thread diameter of 3.9 mm or less.