Investigation of the radiological relationship between iliac crests, conus medullaris and vertebral level in children.

Background: The inability of anaesthetists to accurately identify vertebral spaces has been documented. The aim of our magnetic resonance imaging (MRI) study was to find the relationship of vertebral body level with Tuffier's line and the conus medullaris in children.

Methods: Forty-nine children aged 10 years or less who had MRI scanning in the region of the lumbar spine were identified. If either the conus medullaris or iliac crests were visible, the corresponding vertebral body level was recorded.

Results: The level of the conus could be identified in 43 patients. The median conus level was at L1 with an interquartile range (IQR) of 0.5 vertebral levels. The level of the crests could be identified in 35 patients. The median crest height was L5 with an IQR of 0.5 vertebral levels. The difference between conus and crest levels could only be calculated in 29 patients. In the remainder of the children one of the two pieces of the data was not available. The median difference was 4 vertebral levels with an interquartile range (IQR) of 1. Two of 49 children, both of whom had tethered cords, had differences of <2 vertebral levels.

Conclusions: Our data confirmed that the iliac crests and conus medullaris were consistently related to the predicted vertebral levels in children. In the absence of a tethered spinal cord, subarachnoid puncture at L4-5 or below should present little risk of conus damage. Tuffier's line corresponded with a radiological level of L4-L5 in our study population.