Cost Analysis of Magnet-driven Growing Rods for Early-onset Scoliosis at 5 Years.

STUDY DESIGN: Prospective case series of nine children with early-onset scoliosis (EOS) treated by a single surgeon with a novel implant, the magnet-driven growing rod (MdGR) in a publicly funded health care service accounting for "payer costs" (PC) incurred.

OBJECTIVE: The aim of this study was to compare the cost-effectiveness of MdGR versus conventional growing rods (CGRs) with respect to the PC incurred for treating EOS at 5 years.

SUMMARY OF BACKGROUND DATA: Cost estimate and mathematical modeling study projections of MdGR have shown despite high insertional costs, it breaks even with CGR by 3 to 4 years. However, no clinical study to date exists either supporting or refuting this hypothesis.

METHODS: Nine patients with EOS secondary to idiopathic (two), congenital (one), syndromic (three), and neuromuscular (three) etiologies treated by submuscular insertion of MdGR against stringent inclusion criteria formed the study cohort. We collected costs incurred with all aspects of care over the lifetime of device (or at least 5 years) from payers' perspective to compute and report average PC incurred per patient. We performed this cost analysis by comparing the MdGR PC against literature reported PC for CGR at 5 years.

RESULTS: There were five single rod (SR) and two dual rod (DR) de novo MdGR insertions, while two patients had conversion of CGR to MdGR. MdGR alone accounted for at least 50% of overall budget. The MdGR was at least 40% more cost-effective in comparison to the CGR (£34,741 vs. £52,293) and there were seven MdGR graduates.

CONCLUSION: The first study reporting direct PC incurred in EOS treated by MdGR that is devoid of any mathematical modeling and deterministic sensitivity analysis is presented. The true societal/human cost savings taking into consideration indirect costs are likely to be significantly higher. MdGR is a promising novel implant that may eventually become the "standard of care" for certain EOS etiologies.

LEVEL OF EVIDENCE: 4.