Study design: A consecutive series of patients with idiopathic scoliosis treated with single-rod instrumentation was followed prospectively. Outcomes were compared with results obtained from a retrospective review of a consecutive series of patients treated with double-rod instrumentation.
Objective: To compare single-rod instrumentation with segmental fixation with double-rod instrumentation for the treatment of adolescent idiopathic scoliosis.
Summary of background data: Mechanical testing of single-rod instrumentation with segmental fixation at every level showed it to be as resistant to torsion as a double-rod construct. A clinical trial was initiated to document the clinical outcome in single-rod patients.
Methods: A total of 43 of 51 consecutive patients underwent spinal fusion with a single rod. Outcome was evaluated at a minimum of 2 years after surgery. The control group comprised 103 patients who had standard double-rod instrumentation at the same institution.
Results: The single- and double-rod groups were similar with respect to age, sex, curve type, length of follow-up, curve magnitude, and best bend. For King III-V curves undergoing posterior spinal fusion, there was significantly less blood loss in the single-rod group (703 mL vs 1011 mL), less cell saver collection (189 mL vs 367 mL), and less operating time (220 minutes vs 260 minutes). Blood loss and operating time were not different for patients with King I and King II curves. There were eight patients (19%) requiring reoperation because of hardware-related problems in the single-rod group compared with four (4%) in the double-rod group. There were nine patients (21%) with broken rods in the single-rod group, six of whom were symptomatic and five of whom required reoperation. Two patients required multiple operations because of pseudarthrosis in the single-rod group. There were no broken rods in the double-rod group. The single-rod group had 2 early postoperative infections and no late infections compared with 10 late infections in the double-rod group. There was a statistically significant relationship between hardware problems and fusion below L1 in the single-rod group.
Conclusion: Because of rod failure, single-rod instrumentation should be considered only in curves that can be instrumented to L1 and higher.