• Richard B North, David Kidd, Jane Shipley, and Rod S Taylor.
• Department of Neurosurgery, The Johns Hopkins University, Baltimore, Maryland 21287, USA. RNorth@jhmi.edu
• Neurosurgery. 2007 Aug 1;61(2):361-8; discussion 368-9.

ObjectiveWe analyzed the cost-effectiveness and cost-utility of treating failed back-surgery syndrome using spinal cord stimulation (SCS) versus reoperation.Materials And MethodsA disinterested third party collected charge data for the first 42 patients in a randomized controlled crossover trial. We computed the difference in cost with regard to success (cost-effectiveness) and mean quality-adjusted life years (cost-utility). We analyzed the patient-charge data with respect to intention to treat (costs and outcomes as a randomized group), treated as intended (costs as randomized; crossover failure assigned to a randomized group), and final treatment costs and outcomes.ResultsBy mean 3.1-year follow-up, 13 of 21 patients (62%) crossed from reoperation versus 5 of 19 patients (26%) who crossed from SCS (P < 0.025) [corrected]. The mean cost per success was US $117,901 for crossovers to SCS. No crossovers to reoperation achieved success despite a mean per-patient expenditure of US $260,584. The mean per-patient costs were US $31,530 for SCS versus US $38,160 for reoperation (intention to treat), US $48,357 for SCS versus US $105,928 for reoperation (treated as intended), and US $34,371 for SCS versus US $36,341 for reoperation (final treatment). SCS was dominant (more effective and less expensive) in the incremental cost-effectiveness ratios and incremental cost-utility ratios. A bootstrapped simulation for incremental costs and quality-adjusted life years confirmed SCS's dominance, with approximately 72% of the cost results occurring below US policymakers' "maximum willingness to pay" threshold.ConclusionSCS was less expensive and more effective than reoperation in selected failed back-surgery syndrome patients, and should be the initial therapy of choice. SCS is most cost-effective when patients forego repeat operation. Should SCS fail, reoperation is unlikely to succeed.

23 keywords...

hide…