The effect of screw length and position on fixation of four-stranded hamstring grafts for anterior cruciate ligament reconstruction.

The aim of this study was to examine interference screw fixation of four strand hamstring grafts for anterior cruciate ligament reconstruction in vitro. Bovine tibiae and cannon tendons were used. Screws were introduced from 'outside-in'. The tendons were loaded cyclically to 150 N to simulate walking and 450 N to simulate jogging. Slippage of the tendon from the anchorage was recorded after 100, 300 and 1000 cycles. Cortico-cancellous and cancellous only fixations were compared using a 7-mm screw in an 8-mm tunnel with a 7-mm graft. The effect of screw length was studied by comparing 8 x 25 and 8 x 45 mm screws. Graft slippage with cancellous only fixation was significantly greater than with cortico-cancellous screw fixation (P<or=0.007). The grafts tended to slip more with a 25-mm screw than with a 45-mm screw but, while this difference was not significant (P>0.105), the longer screws gave more consistent behaviour. We concluded that the screw should be placed so that its head engages the cortex, while increasing screw length within cancellous bone did not significantly improve graft fixation. Cyclical load testing reflects the repetitive forces imposed in the early post-operative period following hamstring anterior cruciate ligament reconstruction and is important for the evaluation of graft fixations.