Growing rods for spinal deformity: characterizing consensus and variation in current use

Abstract

Summary of background data: Growing rods are a commonly used form of growth guidance for patients with early onset scoliosis, but no studies exist to characterize their use among a large group of surgeons.

Methods: A survey regarding growing rod use preferences and a case-based survey regarding early onset scoliosis were completed by an international group of surgeons. Two hundred and sixty-five growing rod patients treated over 4.7+/-2.1 years in the Growing Spine Study Group database were analyzed to characterize actual practice and compare it with the survey results. All patients had at least 2 years of treatment.

Results: In the case-based survey, there was correlation (P=0.04, r=0.58) between increasing curve size and choice of growing rods over nonoperative treatment, rib-based distraction (vertically expandable prosthetic titanium rib), growth guidance (Shilla), and primary fusion. In practice, growing rods were used for most types of early onset spine deformity. Most surgeons stated that their indication for growing rod treatment was a curve over 60 degrees (10/13) in a patient younger than 8 to 10 years (14/17). In practice, mean curve at rod insertion was 73+/-20 degrees and age was 6.0+/-2.5 years. Other factors favoring growing rods included curve rigidity (8/17), brace intolerance (6/17) and syndromic diagnoses (2/17). In the database, idiopathic scoliosis represented <50% of diagnoses. The most common preferred surgical lengthening interval was 6 months. However, in practice, lengthening actually occurred at a mean of 8.6+/-5.1 months. In the database, the number of growing rod insertions per year (P=0.02, r=0.96) and percentage of surgeons using dual rods over single rods (P=0.065, r=0.93) increased over time. Insertion age (P=0.075, r=-0.87) and lengthening interval (P=0.006, r=-0.69) decreased as time progressed. The most common stated indication on the survey for final fusion was skeletal maturity (13/17), and 7/13 surgeons used Risser 3 or more. Indications to stop lengthening included complications such as infection or implant failure (14/17), curves progressing past 90 degrees (8/17), and failure to distract (6/13). The most common method of final fusion was replacement of implants with more intermediate anchors.

Conclusions: Significant practice variation exists in growing rod treatment, but there is some consensus on indications for surgery including curve size, diagnosis and age, and lengthening intervals and final fusion methods. Mean curve size and lengthening interval are greater in practice than in surgeons' stated aims. In principle and in practice, most growing rods are used for curves over 60 degrees in patients under 10, in all diagnoses. This information may form a starting point as practice variation is studied.