Introduction: The AO/OTA 31 A-3 fractures are very unstable and biomechanically differ from the 31 A-1 and A-2 fractures. Recent papers state that the extra-medullary implants used to treat these fractures have a failure rate as high as 56%. Few papers report the results with intramedullary implants, and, sometimes to heal the fracture, the nails had to be dynamised by to removing the distal screw or the fracture is “self-dyna-mised” by breaking the distal screw.

Material and Methods: A prospective clinical study was designed to treat 57 consecutive patients with a 31 A-3 fracture. Mean age (84.2 years). The GT Short Nail (17-cm long, 16-mm upper diameter, and a distal locking oval hole allowing 12 mm of proximal sliding), was inserted through the apex of the greater trochanter to stabilize these fractures with a single 9-mm rotationally unlocked hip-screw, and a distal dynamically locked screw. 43 patients followed for six months were included in the study. Radiological studies: screw-tip migration (Doppelt’s method), hip-screw sliding, tip-apex distance (Baumgaertner), and proximal sliding of the distal locking screw (intra-op, one week, one month, three and six months). Full weight bearing with the needed help was encouraged as soon as possible.

Results: All fractures but one, healed uneventfully. No cutouts, no thigh pain, 1 implant failure in the only delayed healing case, and no deep infections. Three patients had further surgery to heal a bleeding skin incision. The failed implant was removed and replaced by a 90° Synthes hip plate. The average tip-screw migration was 2.4 mm, the tip-apex-distance was < than 25 mm in 94% of the cases, the average hip-screw sliding was 6.4 mm, and the average proximal sliding of the distal locking screw was 4.9 mm. In 24.5 % of the cases (14) the proximal sliding of the distal dynamically locked screw was over 10 mm.

Conclusions: A single 9-mm diameter rotationally unlocked hip screw works very well through the healing process of these very unstable fractures. A dynamically distal locked screw controls the femoral shaft rotation. Our study shows that the distal locking hole of the trochanteric nails should allow at least 10 mm of proximal sliding to provide the unknown needed proximal sliding of the distal fragment to minimize delayed or non-healing of these fractures.