The failure of survivorship.

This editorial considers the shortcomings of assessing outcome after joint replacement only by the survival of the implant. Survivorship of a prosthesis is used as an objective outcome measure to determine whether surgery has been a success or a failure. It is clearly important to know the length of time that an implant has remained in place so that any early failure of a new implant may be identified. However, to define a successful outcome solely on implant survivorship is inadequate and lacks sensitivity. Because revision is uncommon, large sample sizes and long-term follow-up are needed to detect significant differences in survival rates between implants. 1 If an operation is only defined as having failed if it is revised, then patients with a poor outcome who have not been offered revision surgery or elect not to undergo further surgery will not be captured. This results in an overly optimistic representation of the success of joint replacement. From the patient's perspective, success of their joint replacement is unlikely to be judged solely on how long the implant remains in place. Whereas survival analysis would classify an implant with a 15-year survivorship as a success, the patient is unlikely to agree if they have experienced persistent joint pain and disability during that time. Therefore, it is imperative that outcome assessment in orthopaedics moves beyond defining success by survivorship alone and routinely incorporates patient-reported outcome measures (PROMs). Research using PROMs has revealed a subgroup of patients (approximately 10% to 30% of total knee replacements (TKRs) 2 and 10% to 15% of total hip replacements (THRs)) 3 who report persistent pain and functional limitation after surgery. Survival analysis is unable to identify those patients for whom surgery has failed to relieve their pain and disability. However , by incorporating measures of pain, disability and dissatisfaction into the definition of failure alongside revision, a more accurate assessment of outcome can be made. For example, in one series of TKRs, the survivor-ship at seven years using revision surgery as the endpoint was 98%, but this decreased to 75% when moderate to severe pain was added as a second endpoint. 1 Similarly, Bullens et al 4 described a cohort of patients who underwent TKR, in which success was measured in three different ways. The rate of success defined purely by survivorship was 96.7%. When defined as survivorship plus satisfaction > 80 of 100 on a visual analogue …