Aims Samples of cerebrospinal fluid (CSF) are required for the management of many disease states. One of the commonest errors in performing lumbar punctures (LPs) is inaccurate insertion of the spinal needle (too shallow or too deep), with blood contamination often complicating subsequent laboratory interpretation.
We previously demonstrated a good correlation between ultrasound measures of spinal canal depth (SCD) and body weight in neonates, producing a nomogram (1) which has been tested in clinical practice (2). Here, we present preliminary data from ultrasound and auxological assessment in an older general paediatric population.
Other studies have been limited by small cohort size, measurements being taken in the supine position during CT, or data gathered post-LP procedure, based on recall of depth of needle insertion.
Methods Single-centre study of patients aged 0–18 years presenting at a tertiary paediatric centre. Patients were recruited from cardiology outpatients, PICU and paediatric inpatient wards. Individuals with pre-existing spinal abnormalities were excluded.
Ultrasound measures of the spinal canal were performed in the left lateral position to obtain anterior and posterior SCDs; from these measures the mid-SCD (MSCD) was calculated. Patient age and auxological parameters including patient weight, height and body surface area (BSA) were recorded.
Results 125 children were recruited and had complete data recorded (figure 1). The median age of the study population was 7.0 years (range 0.3-16.0). All auxological data are presented as means (standard deviation; range). Weight was 30.6 kgs (16.1; 7.8-88.9) and height 128 cms (29.1; 67.9-179.6), resulting in a mean BSA of 1.03 m2 (0.39; 0.35-1.88). We identified a linear correlation between MSCD (mm) and age (R2 0.68), weight (R2 0.79), height (R2 0.74) and BSA (R2 0.79). The approximated formula for predicting MSCD from body weight (W Kg) in this cohort is MSCD (mm) = 0.4W + 20 (figure 1).
Conclusion Our preliminary data demonstrates good correlation between both patient weight and BSA and MSCD in an unselected older general paediatric population. The most practical MSCD correlation is likely to be that based on weight, as height is infrequently routinely measured in clinical practice. This needs testing in clinical practice.