Inclusion of asymptomatic degenerative discs in a two-level anterior cervical discectomy and fusion: a decision analysis.

OBJECTIVE

To perform a decision analysis model to compare single-level fusion versus two-level fusion in patients with an asymptomatic disc adjacent to a symptomatic disc.

METHODS

Probabilities and utilities of alternative outcomes in the decision tree were assigned based on systematic review of the literature and expert opinion. Rollback analysis determined the optimal treatment. Sensitivity analyses and Monte Carlo simulations were performed to identify effects of varying model parameters.

RESULTS

Rollback analysis provided expected values of 0.92 versus 0.84 in favor of observation as the optimal decision. Sensitivity analysis identified the probability of developing adjacent segment disease (ASD) and the likelihood of surgery given a diagnosis of ASD as the most critical parameters influencing the decision. Observation was the preferred strategy at all values of probability of ASD < 100%. At a probability of ASD of 100%, fusion was the preferred strategy only when the probability of surgery for ASD was ≥ 66% or the utility assigned to successful nonoperative management was ≤ 0.84.

CONCLUSIONS

Observation was the preferred strategy for management of asymptomatic adjacent degenerative discs (AADDs) given the probabilities and utilities used in the decision analysis model. The study was limited by unavailability of precise estimates of the probability of development of ASD and the probability of surgery after diagnosis of ASD, the most critical factors influencing the decision. However, the conclusions were robust given wide ranges used for these parameters in the sensitivity analysis.