Purpose of the study: Certain patients with a total knee arthroplasty (TKA) require large range of flexion postoperatively to enable squatting and sitting cross-legged. Several factors have an effect on this flexion, including prosthesis design. The purpose of this study was to examine the influence of prosthesis design on intra- and postoperative flexion of three modes of TKA with a pure rotational mobile plateau: two NexGen posterostabilised (PS) prostheses (LPS-Standard and LPS-Flex) and one ultracongruent prosthesis (SAL). It was hypothesised that PS prostheses would have a better flexion than the ultracongruent prosthesis and that the flexion would be greater with the LPS-Flex than the LPS-standard.

Material and methods: This was a prospective randomised study of consecutive patients from January 2006 to January 2007 to compare maximal flexion for each model. All patients requiring a first-intention TKA were included in this study. Seventy-tow TKA were studied: LPS-standard (n=24), LPS-Flex (n=22), SAL (n=26). Flexion was measured pre and postoperatively goniometrically. Intraoperative measures were made with the navigation system (Navitrack-Orthosoft). Minimum follow-up was one year.

Results: There was a significant difference in flexion, intraoperatively and at last follow-up, in favour of the PS models over the ultracongruent SAL. Conversely, there was no significant difference between the LPS-standard and the LPS-Flex. The analysis of factors predictive of flexion were the SAL model with a negative influence (loss of 8° intraoperatively [p< 10-4] and 15° at one year [p< 10-4] compared with the LPS models). Preoperative flexion appeared to be a positive predictive factor (p=0.00023).

Discussion: The design of the TKA has an influence on knee flexion: from implantation, flexion of the PS models was superior to the ultracongruent model, a difference which persisted late after the operation. For the PS models, the LPS-Flex model presented as a large flexion model did not, in our study, demonstrate its superiority over the LPS-standard model, irrespective of the time of the comparison. It should be noted that for implantation the LPS-Flex model required a posterior cut 2mm more than for the LPS-standard. Good preoperative flexion is an essential factor for obtaining good postoperative flexion, irrespective of the model implanted.

Purpose of the study: Conventional treatment for recurrent lumbar disc herniation is repeated discectomy. Other methods such as fusion, ligamentoplasty or implantation of a discal prosthesis are sometimes proposed but all increase morbidity. The purpose of this work was to ascertain the efficacy of isolated repeated radicular release for the treatment of recurrent discal herniation.

Material and methods: Thirty-four patients underwent surgery for recurrent discal herniation. Repeated radicular release was used in all patients included in this analysis who completed a self-administered questionnaire at last follow-up to assess the final functional outcome.

Results: The cohort included 13 women and 21 men, mean age at surgery 45 years. Mean time from first discectomy to revision surgery for recurrence was 55 months. At the time of the review, four patients had died, all four from cancer. None of these patients had undergone a revision procedure on the lumbar spine. One patient was lost to follow-up so 85% of the cohort was analyzed with 60 months average follow-up. A dural tear occurred during the proscedure in six patients (17%. Five patients (14.7%) required revision surgery, one for deep infection, four for recurrent or persistent lumboradiculalgia (recurrent discal herniation, isthmic fracture, lateral stenosis associated with inflammatory discopathy). The rate of revision for painful failure was 11.4%. The final outcome could be assessed for 25 patients and was satisfactory for 22/25 (88%). The self-administered questionnaire revealed 65% average improvement with more than half of the patients reported better than 80% improvement. Ten patients (40%) complained of lumbar pain and a third had residual, generally intermittent, radiculalgia. Eighteen of 25 patients resumed their work at a comparable level after six months on average; 84% of the patient would accept the same operation again.

Discussion: In terms of morbidity and rate of revision, the results are comparable to reports in the literature. Repeated release does not increase the risk of a new recurrence.

Conclusion: This work enabled us to demonstrate that in the large majority of patients repeated discectomy provides satisfactory functional outcome with little morbidity.