Ultrasound guidance for difficult lumbar puncture in children: pearls and pitfalls
OBJECTIVE The objective of this study was to determine whether subarachnoid space width at the site of lumbar puncture in infants changed between 3 positions: flat lateral decubitus, 45-degree tilt, and sitting. METHODS Healthy infants younger than 4 months presenting electively to a pediatric radiology department were enrolled. Exclusion criteria included signs of dehydration, illness, or previous spine/brain surgery. Subarachnoid space width at L3-L4 was measured by ultrasound with the subject placed in 3 randomly ordered positions: flat lateral decubitus, 45-degree tilt lateral decubitus, and sitting. The 3 positions were collectively compared using both repeated-measures analysis of variance and linear mixed models (LMMs) adjusted for potential confounders. Pairs of positions were compared using LMM adjusted for potential confounders. RESULTS Fifty subjects were enrolled (15 male and 35 female patients). Patient weight was significantly correlated with subarachnoid space width (P = 0.02). There was no statistically significant difference in subarachnoid space width between the 3 positions (repeated-measures analysis of variance P = 0.32, LMM P = 0.40). Comparisons of pairs of positions were not significantly different: flat and 45 degrees P = 0.24, 45 degrees and sitting P = 0.98, and flat and sitting P = 0.23. CONCLUSIONS The subarachnoid space width did not significantly change between the 3 positions. An increase in lumbar puncture success rate with sitting or tilt position could be due to other factors such as increased cerebrospinal fluid pressure, increased interspinous space widening, or improved identification of landmarks.