Lumbar Neuraxial Ultrasound Corrrelation With MRI

Overview

Ultrasound is used increasingly to aid placement of neuraxial blocks. Good visibility of ligamentum duramater complex and posterior longitudinal ligament (LF, PLL) has a positive predictive value of 85% for successful neuraxial block. Dural puncture could still be feasible despite the absence of a good view, as reflected by a negative predictive value of 30%. Poor neuraxial ultrasound view could be due to limitations of ultrasound or various other anatomical reasons which might include absent LF, calcification of ligaments, spinous process contact and presence or absence of epidural or subdural fat. This will be the first study of its kind to compare neuraxial ultrasound with MRI. The aim of this study is to look for anatomical correlation between neuraxial ultrasound and MRI at various lumbar interspinous levels and to look at factors that contribute to abnormal neuraxial imaging. Assuming the predicted abnormal ultrasound images to be around 30% and predicted abnormal MRI images to be 5%, 79 images should give a power of 0.9 and alpha 0.01 to analyse the association between the two. Sample Size Calculations were performed using statistical software R version 2.15.2

After ethical committee approval, 20 consenting patients scheduled for MRI lumbar spine were included in the study. Patients with previous spinal surgeries or with gross spinal deformities were excluded. In all patients ultrasound scanning of the lumbar interspinous spaces was performed with curvilinear 2-5 MHz Sonosite probe. At each lumbar interspinous level (L1-2 to L5-S1) the best possible transverse and paramedian oblique sagittal view was obtained. The obtained ultrasound images were the graded based on the visibility of LF and PLL. The lumbar spine MRI images were analyzed by radiologists who were blinded to the ultrasound images. On MRI various parameters such as depth of epidural space, absence of LF, thickness of LF, characteristics of muscular and intervening fatty tissue, epidural/subdural fat thickness etc was noted. The correlation between neuraxial ultrasound images and various parameters observed in MRI was identified.

Conditions