Mechanical performance of traditional distraction-based dual growing rod constructs

Abstract

Background: Growing rod constructs are an important contribution in the treatment of children with early onset scoliosis even though these devices experience high rates of rod fracture. The mechanical performance of traditional, distraction-based dual growing rod constructs is not well understood, and mechanical models for predicting device performance are limited.

Purpose: Two mechanical models were developed and used to determine the mechanical performance of various growing rod configurations by increasing construct complexity.

Study design/setting: Mechanical bench testing and finite element (FE) analysis.

Methods: Static and dynamic compression bending tests were based on an ASTM F1717 method modified to accommodate dual growing rod constructs. Six construct configurations were tested, mechanical properties were recorded, and statistical analyses were performed to determine significant differences between groups: (1) no connectors (rods only), (2) side-by-side connectors, (3) side-by-side connectors plus 4 crosslinks, (4) (40-mm long tandem connectors, (5) 80-mm long tandem connectors, and (6) 80-mm long tandem connectors plus 4 crosslinks. FE analysis was used to predict the stress distribution within the constructs.

Results: The static results indicated greater stiffness, yield load, and peak load as the axial connector length increased (side-by-side to 40 mm tandem to 80 mm tandem). The dynamic results showed similar cycles to failure for side-by-side and tandem connector (40 and 80 mm) construct configurations without crosslinks. Crosslinks shifted the location of rod fracture observed experimentally and significantly reduced the fatigue life of the construct. The flexibility of the construct decreased significantly as the axial connector length increased. FE predictions were highly consistent with the experimentally measured values and provided information on stress distribution within the rod for comparison to experimental fracture locations.

Conclusions: This is the first study to evaluate mechanical performance of various configurations of pediatric growing rod constructs using preclinical models. The current study is consistent with a previous retrieval study in that rigid constructs lacking flexibility (ie, higher stiffness and lower displacement), such as those with 80-mm tandem connectors and multiple crosslinks, demonstrated decreased mechanical performance as shown through both experimental and computational models. Additionally, the experimental and computational findings suggest that surgeons should strategically consider the number of interconnecting components and subsequent stress concentrations along the posterior side of the rod. For example, changing the placement of crosslinks to low stress regions of the construct or not using crosslinks in the construct are options.

Keywords: Computational or finite element analysis model; Early onset scoliosis; Growing rod; Mechanical or bench testing; Pediatric; Spinal instrumentation.