Aims. One method of femoral head preservation following avascular necrosis (AVN) is core decompression and insertion of a tantalum rod. However, there may be a high failure rate associated with this procedure. The purpose of this study was to document the clinical and radiological outcomes following total hip arthroplasty (THA) subsequent to failed tantalum rod insertion. Patients and Methods. A total of 37 failed tantalum rods requiring total hip arthroplasty were identified from a prospective database. There were 21 hips in 21 patients (12 men and nine women, mean age 37 years, 18 to 53) meeting minimum two year clinical and radiographic follow-up whose THAs were carried out between November 2002 and April 2013 (mean time between tantalum rod implantation and conversion to a THA was 26 months, 6 to 72). These were matched by age and gender to individuals (12 men, nine women, mean age 40 years, 18 to 58) receiving THA for AVN without prior tantalum rod insertion. Results. There were no functional outcome differences between the two groups. Tantalum residue was identified on all post-operative radiographs in the tantalum group. Linear wear rates were comparable between groups with no evidence of catastrophic wear in either group. Conclusion. In the short term, tantalum rod implantation does not demonstrate an adverse effect on subsequent total joint replacement surgery. There is however, a high rate of retained tantalum debris on post-operative radiographs and thus there is an unknown risk of accelerated articular wear necessitating longer term study. Cite this article: Bone Joint J 2016;98-B:1175–9

The introduction of a trabecular tantalum rod has been proposed for the management of early-stage osteonecrosis of the femoral head but serves as a single-point of support of the necrotic lesion. We describe a technique using two or three 4.2 mm (or later 4.7 mm) tantalum pegs for the prevention of collapse of the necrotic lesion. We prospectively studied 21 patients (26 hips) with non-traumatic osteonecrosis of the femoral head treated in this manner. Of these, 21 patients (24 hips) were available for radiological and clinical evaluation at a mean follow-up of 46 months (18 to 67). Radiological assessment showed that only eight hips deteriorated according to the Association Research Circulation Osseous classification, and four hips according to the Classification of the Japanese Investigation Committee of Health and Welfare. Functional improvement was obtained with an improvement in the mean Harris hip score from 65.2 (33.67 to 95) to 88.1 (51.72 to 100), the mean Merle D’Aubigné-Postel score from 13 (6 to 18) to 16 (11 to 18), a mean visual analogue score for pain from 5.2 (0 to 9.5) to 2.6 (0 to 7), and the mean Short-Form 36 score from 80.4 (56.8 to 107.1) to 92.4 (67.5 to 115.7). Of these 24 hips followed for a minimum of 18 months, three were considered as failures at the final follow-up, having required total hip replacement. One of the hips without full follow-up was also considered to be a failure. In more than two-thirds of the surviving hips a satisfactory clinical outcome was achieved with promising radiological findings. The estimated mean implant survival was 60 months (95% confidence interval 53.7 to 66.3).