Early versus delayed bilateral subthalamic deep brain stimulation for parkinson's disease: a decision analysis.

The long-term benefits of subthalamic nucleus deep brain stimulation (STN DBS) applied earlier in the disease course, before significant disability accumulates, remain to be determined. We developed a Markov state transition decision analytic model to compare effectiveness in quality-adjusted life years (QALYs) of STN DBS applied to patients with PD at an "early" ("off time" 10-20%) versus "delayed" stage ("off time" >40%). A lifelong time horizon and societal perspective were assumed. Probabilities and rates were obtained from literature review; utilities were derived using the time trade-off technique and a computer-assisted utility assessment software tool applied to a cohort of 22 STN-DBS and 21 non-STN-DBS PD patients. Uncertainty was assessed through one- and two-way sensitivity analyses and probabilistic sensitivity analysis using second-order Monte Carlo simulations. Early STN DBS was preferred with a quality-adjusted life expectancy of 22.3 QALYs, a gain of 2.5 QALYs over those with delayed surgery (19.8 QALYs). Early STN DBS was preferred in 69% of 5,000 Monte Carlo simulations. Early surgery was robustly favored through most sensitivity analyses. Delayed STN DBS afforded greater QALYs when using utility estimates exclusively from non-STN-DBS patients and, for the entire group, if the rate of motor progression were to exceed 25% per year. Although decision modeling requires assumptions and simplifications, our exploratory analysis suggests that STN DBS performed in early PD may convey greater quality-adjusted life expectancy when compared to a delayed procedure. These findings support further evaluation of early STN DBS in a controlled clinical trial.