Cannulated hip screws are frequently used in the management of hip fractures. There have been concerns over the failure rate of the technique and the outcomes of those that subsequently require conversion to total hip replacement (THR). This study utilised a database of over 600 cannulated hip screw (CHS) fixations performed over a 14-year period and followed up for a minimum of one year (1-14). We identified 57 cases where a conversion to THR took place (40 females, 17 males, mean age: 71.2 years). Patient demographics, original mechanism of injury, fracture classification, reason for fixation failure, time until arthroplasty, implant type and post-arthroplasty complications were recorded. Clinical outcomes were measured using the Oxford Hip Score. The failure rate of cannulated screw treatment was 9.4% and the mean time from initial fixation to arthroplasty was 15.4 (16.5) months. Thirty six fractures were initially undisplaced and 21 were displaced. As one might expect the displaced cases tended to be younger but this didn't reach statistical significance [66.5(14.3) vs 72.7(13.1), p=0.1]. The commonest causes of failure were non-union (25 cases, 44%) and avascular necrosis (17 cases, 30%). Complications after THR consisted of one leg length discrepancy and one peri-prosthetic fracture. The mean Oxford score pre-arthroplasty was 12.2 (8.4), improving to 38.4 (11.1) at one-year. Although the pre op Oxford scores tended to be lower in patients with undisplaced fractures and higher ASA scores, the improvement was the same whatever the pre-op situation. The one-year Oxford score and the improvement in score are comparable to those seen in the literature for THR in general. In conclusion, CHS has a high success rate and where salvage arthroplasty is required it can provide good clinical outcomes with low complication rates