Hip fracture is a common serious injury in the elderly. Between 1982 and 1998 the number of hip fractures reported annually in Scotland in patients over 55 years rose from 4,000 to 5,700. The optimum method of treatment for the various fracture types remains in contention.

We compare outcome measures between displaced, intracapsular fractures in patients over 70 years fixed with cannulated screws and sliding hip screw with side plate.

Between 1998 and 2005 a total of 30,482 patients were reviewed by the Scottish Hip Fracture Audit (SHFA). Of these 15,823(53.3% of the total) had sustained intracapsular fractures. 13,587 of these occurred in patients aged 70 or over. Of these 2,428 had undisplaced and 11,159 displaced fractures. Chi test statistical analysis compare outcome measures in this group of displaced intracapsular fractures with respect to aspects of early failure.

534(3.9%) of patients were treated conservatively. 509 (4.7%) fractures were fixed using cannulated screws and 499 (4.6%) using a sliding hip screw. Readmission within 120 days for any cause occurred in 62 patients(14.1%) treated with cannulated screw fixation and in 63 patients(15.7%) for those treated with a sliding hip screw(P=0.509). Of these 36 patients(8.2%) in the former and 23 patients(5.7%) in the latter group were readmitted for complications related to hip fracture(P=0.033). Mortality within this period included 69 patients(13.5%) in the CS and 98(19.6%) in the SHS group. In terms of re-operation within 120 days of the original admission, 53 patients(10.6%) receiving cannulated screws compared to 24 patients(4.8%) treated with a sliding hip screw requiring further surgery(P=0.0006). The fracture was seen to displace in 12(22.6%) patients originally treated with cannulated screws compared to 6 patients(25%) treated with sliding hip screw(P=0.156). More significantly the fixation device was seen to have migrated in 24(45.3%) of the cannulated screw as compared to 7(29.2%) patients in the sliding hip screw group(P=0.002). Periprosthetic fractures were recorded in 4(7.5%) of the former and 3(12.5%) in the latter group(P=0.708). Wound infection was higher in the SHS group(2 patients) as compared to the screw fixation group(1 patient)(P=0.565).

Statistical analysis demonstrates a dramatic difference exists between these 2 fixation types in terms of re-operation within 120 days of the original admission for which published literature has previously only recorded biomechanical, in vitro comparisons.

Data regarding specific implant factors such as number of screws, position, configuration, starting point, thread length and use of washers in cannulated screws, and position, tapping, supplementary screw and compression screw in sliding hip screws was not recorded and may be considered to bias our results.