Staged Growing Rods With Preimplantation of Spinal Anchors for Complex Early Onset Scoliosis

BACKGROUND:Treatment of early onset scoliosis (EOS) with growing rods (GR) can be challenging in patients with significant deformity, hyperkyphosis, or poor bone quality, due to risks of neurological deficit and hardware pull-out. The objective of this study is to report a series of EOS patients managed with a 2-stage GR technique used to minimize these complications. METHODS:Two-stage GR technique was performed in 8 patients at mean age of 5.4 (range, 3.4 to 7.9) years. At stage 1, proximal and distal anchors were implanted with local fusion. At stage 2, the distraction rods were inserted. There were at least 3 months between stages 1 and 2, and halo-gravity traction was used before stage 2 unless contraindicated. Demographic, clinical, and surgical data were retrospectively reviewed with mean 4.9 (range, 2.0 to 9.4) years of follow-up. Radiographic measurements including Cobb and kyphosis angles were evaluated before stage 1, after halo-gravity traction, after stage 2, and at last follow-up. Indications for staging, anchor healing time, and complications were collected and analyzed. RESULTS:Indications for 2-stage surgery were poor bone quality in 5 patients and neurological changes during initial attempt at GR placement in 3 patients. The mean time between stage 1 and 2 was 23 (15 to 45) weeks. Patients have undergone mean 7 (3 to 16) lengthenings. Three patients have been converted to magnetically controlled GR. The major coronal Cobb angle improved from mean 81 degrees (range, 61 to 97) preoperatively to 40 degrees (24 to 50) after stage 2 and remained at 40 degrees (27 to 53) at last follow-up. Kyphosis remained controlled from 45 degrees (10 to 76) preoperatively to 38 degrees (9 to 61) after stage 2 to 41 degrees (17 to 65) at last follow-up. Complications included superficial wound problems (4 patients), broken rods (2 patients), proximal migration (2 patients), and implant prominence (1 patient). At minimum 2-year follow-up, no patients had lingering neurological complications or instrumentation-bone failure of the GR construct. CONCLUSIONS:Two-stage GR technique can effectively be utilized in EOS patients with poor bone quality or in the event of intraoperative neurological changes. Satisfactory deformity correction can be obtained and routine serial lengthenings can take place with minimal complications. LEVEL OF EVIDENCE:Level IV— case series.