Comparison of Intended Lengthening of Magnetically Controlled Growing Rods: Ultrasound Versus X-Ray

Background: In the treatment of early onset scoliosis (EOS), there has been a trend to use magnetically controlled growing rods (MCGR) in order to reduce the number of surgeries. To confirm the amount of lengthening, spine radiographs were required. Recently, ultrasound (US) has been added to monitor lengthening of MCGR to avoid radiation exposure. Our aim was to determine whether US is as accurate as plain radiography (x-ray) in determining the amount of length achieved at individual MCGR lengthening episodes. Methods: Retrospective study; inclusion criteria: EOS cases with dual MCGR with minimum 12 months follow-up. Intended lengthening IL (mm), lengthening on US (mm) and x-ray (mm) were documented from medical records for both right and left rods. Primary (no surgery before MCGR) and conversion (other types of instrumentation were replaced with MCGR) cases were reviewed separately. P-values determined with analysis of variance. Results: Sixteen cases with 100 lengthening episodes met the inclusion criteria. Eleven were primary MCGR cases with 67 episodes. Mean follow-up was 19 +/- 5 months. Significant differences were found between IL (3.4 +/- 1 mm), US (2.7 +/- 1.9 mm), and x-ray (4.1 +/- 2.2 mm) (P<0.001). The difference between IL and x-ray was minimal, but statistically significant (P=0.046). US showed statistically lower values than both IL (P=0.001) and x-ray (P<0.001). The mean ratio of x-ray/IL, US/IL, and US/x-ray were 1.1, 0.75, and 0.84, respectively. Five conversion cases had 33 episodes. Mean follow-up was 21 +/- 2 months. Significant differences were found between IL (3.4 +/- 0.8 mm), US (1.3 +/- 0.8 mm), and x-ray (1.7 +/- 0.9 mm) (P<0.001) but there was no significant difference between US and x-ray (P=0.283). IL was significantly higher than both US (P< 0.001) and x-ray (P<0.001). The mean ratio of x-ray/IL, US/IL, and US/x-ray were 0.64, 0.41, and 1.1, respectively. Conclusions: US can provide confirmatory information of noninvasive lengthening of MCGR. However, US tended to underestimate the achieved length as measured by x-ray in primary cases. Conversion cases demonstrate better concordance between US and x-ray but in these cases less overall length was achieved at each lengthening episode.