In a prospective randomised study, the Hansson pin technique for internal fixation of fractures of the femoral neck was compared with the Uppsala screw technique. The series consisted of 115 consecutive patients, 97 women and 18 men, with a mean age of 80 years. There were 56 patients treated with Hansson pins and 59 with Uppsala screws. After one year 22 patients had died, eight in the Hansson pin group and 14 in the Uppsala screw group (p = 0.28). Twenty-four complications had occurred, 18 in the Hansson pin group and six in the Uppsala screw group (p = 0.008). After exclusion of those with complications, the patients in the Hansson pin group had significantly more pain, less mobility, and a smaller proportion were living in their own homes.

We measured the stability of fixation in femoral neck fractures treated with von Bahr screws, investigated the influence of impaction and correlated peroperative stability with the clinical results. Stability was measured at operation using a metal probe fitted with strain gauges. Its tip was anchored in the subchondral bone of the femoral head and its lateral end was fixed in the lateral femoral cortex. The shearing force produced by longitudinal compression applied to the foot of the operated leg was recorded. The results in 41 consecutive patients all followed for 30 months, showed that fractures with early loosening or nonunion had all had significantly poorer stability than the fractures that had healed. Impaction improved stability in only 23 out of the 41 fractures; in the others stability had deteriorated or was unchanged.

We describe a method of internal fixation for femoral neck fractures which has been newly developed to reduce the frequency of early complications. Two cannulated screws are inserted in the axis of the femoral neck to reach into the subchondral bone of the femoral head. The screws are inserted over guide pins and the tip of the screw is self-tapping and designed to provide good anchorage in the femoral head. We used this method in 44 consecutive patients in a prospective study with no exclusions, followed for a minimum of 24 months. All fractures healed within 12 months, and there were no cases of early loosening or nonunion. In four cases, late segmental collapse had developed during the mean follow-up period of 30 months.