Condylocephalic nails versus extramedullary implants for extracapsular hip fractures.

BACKGROUND: Condylocephalic nails are intramedullary nails which are inserted up through the femoral canal from above the knee, for example Ender and Harris nails. OBJECTIVES: To compare condylocephalic nails with alternative implants (extramedullary implants such as fixed nail plates and sliding hip screws, or other intramedullary nails) for the treatment of extracapsular hip fracture in adults. SEARCH STRATEGY: We searched the Cochrane Musculoskeletal Injuries Group trials register, Medline (1983 to August 1999) and reference lists of relevant articles. Date of the most recent search: August 1999. SELECTION CRITERIA: Randomised or quasi-randomised trials comparing condylocephalic nails with other implants. DATA COLLECTION AND ANALYSIS: All reviewers independently assessed trial quality and extracted data. Data were pooled where relevant and possible. Ender nails and Harris nail data were presented separately. Results from fixed nail plates and sliding hip screws were sub-grouped to explore differences in these two implant types. MAIN RESULTS: Eleven trials were included. Ten compared Ender nails with either a fixed nail plate or a sliding hip screw. One compared the Harris condylocephalic nail with a sliding hip screw. The only advantages of condylocephalic nails were a reduced deep wound sepsis rate (0.9% versus 4.2%; odds ratio 0.26, 95% confidence interval 0.12 to 0.56), length of surgery and operative blood loss. However there was an increased risk of re-operation (20.9% versus 5. 5%; odds ratio 3.78, 95% confidence interval 2.67 to 5.36) and later fracture of the femur when compared with extramedullary implants. There was also an increased risk of cut-out of the implant from the femoral head for Ender nails compared with the sliding hip screw, but not for fixed nail plates. Backing out of the nail was a frequent complication (30%) of Ender nails and often resulted in revision surgery. Ender nails also had an increased risk of shortening of the leg and external rotation deformity and potentially a poorer return to previous walking ability. An increase in residual pain resulting from an excess of knee pain was also evident in patients undergoing condylocephalic nailing. There was no apparent difference in mortality between the condylocephalic nail and extramedullary implant groups. REVIEWER'S CONCLUSIONS: Any advantages in intra-operative outcomes of condylocephalic nails are outweighed by the increase in fracture healing complications, re-operation rate, residual pain and limb deformity when compared with an extramedullary implant, particularly a sliding hip screw. The use of condylocephalic nails (in particular Ender nails), for trochanteric fracture is no longer appropriate.