Dynamics of improvement following total lumbar disc replacement: is the outcome predictable?

STUDY DESIGN: Prospective clinical study of total lumbar disc replacement (TDR) with ProDisc II (Synthes, Paoli, PA).

OBJECTIVES: To examine whether baseline variables VAS (Visual Analogue Scale) and ODI (Oswestry Disability Index) correspond with late and final postoperative outcome parameters and to identify early predictors of late outcome following total lumbar disc replacement (TDR).

SUMMARY OF BACKGROUND DATA: Previously published TDR studies reported on the pooled data averages collected from various cohort sizes. The individual patient's prognosis as well as prognostic factors of postoperative improvement remain unestablished.

METHODS: Data were accumulated prospectively and included VAS and ODI scores. The subjective outcome evaluation was based on a 3-scale grading system ("highly satisfied," "satisfied," "not satisfied"). An analysis was performed to ascertain whether the late and final outcome following TDR can be predicted based on preoperative and early postoperative data from the 3 and 6 month follow-up (FU).

RESULTS: The overall results from 161 patients with an average FU of 4 years (mean: 45.5 months, range: 24.1-94.4 months) revealed a significant and maintained improvement of VAS and ODI scores (P < 0.0001). The most pronounced changes occurred within the early postoperative period (P < 0.0001) with no significant changes thereafter (P > 0.05).Baseline ODI levels were significantly correlated with VAS/ODI scores and patient satisfaction rates at the final FU (P < 0.0001).After surgery, early and late ODI levels were highly significantly correlated with each other (r = 0.84, P < 0.0001). Similar associations were observed between early and late VAS scores and patient satisfaction rates (P < 0.006).The individual patient's subjective outcome evaluation revealed stable postoperative results. An improvement or a deterioration by 2 classes on a 3-scale grading system was only observed in 3.1% (n = 5/161) of all cases overall. Patients with an early "highly satisfactory" result (n = 83) maintained either a satisfactory (15.7%, n = 13/83) or a highly satisfactory outcome (79.5%, n = 66/83) in 95.2% of all cases (n = 79/83).Conversely, the probability that patients with an "unsatisfactory" outcome would still achieve a "highly satisfactory" result after the early postoperative period was 5.0%.

CONCLUSION: Baseline ODI and early postoperative outcome parameters (< or =6 months) revealed significant and strong associations with the final results following TDR. While the vast majority of patients with an early highly satisfactory outcome maintained satisfactory results at later FU stages, any significant improvement considered as "highly satisfied" is unlikely in a group of patients which reported early unsatisfactory results. In summary, any clinically relevant changes are unlikely to occur after the early postoperative period.The current findings offer a foundation for weighing both the patients and the spine surgeons expectations against possible realistic achievements. Although the data show that the midterm outcome at a FU of 4 years (mean: 45.5 months, range: 24.1-94.4 months) is predictable following TDR, the long-term results of lumbar disc replacements still need to be established.