No value is given for a normal red cell count, because the definition of a ‘traumatic tap’ varies from 500 to 1000 red blood cells (RBC)/mm ³ . Byington used a definition of <1000/mm ³ , and found that the mean red cell count was 95 RBC/mm ³ , the median value was 5 with a range of 0–236 ( ).

More recent studies tend to give lower white cell counts (WCCs) than older studies, and as a result the upper limit of normal for the WCC in neonatal cerebrospinal fluid (CSF) has gradually been revised down in successive editions of this textbook. give values of up to 112 white blood cells (WBC)/mm ³ in preterm and 90 in term babies, but the study included babies with over 1000 RBC/mm ³ and was done before routine ultrasound was available. The results are so much higher than other more recent studies that they now appear historical. Recent studies have used polymerase chain reaction to exclude viral infection, which was not available in the past.

In 278 babies less than a month old who did not have meningitis the mean WCC was 6.1 with a median of only 5 WBC/mm ³ ( ). Babies less than a month old have higher cell counts than older babies. Similar low values were reported by . The 95th percentile value for CSF WCCs in the newborn babies in this study was 19/mm ³ . The results were not different in the preterm group, but only 22 preterm babies less than a month old were included. found that using 21 WBC/mm ³ as the upper limit of the threshold led to a sensitivity of 79% and a specificity of 81% for the diagnosis of meningitis; in this study there were 95 babies with meningitis, of whom 12 had a WCC ≤21 cells/mm ³ . found that the 90th percentile value for CSF WCC was 26 cells/mm ³ in the first week of life, and 9 cells/mm ³ thereafter. Occasionally babies with meningitis have falsely reassuring CSF WCCs ( ; ). Sometimes CSF obtained very early after meningeal invasion may not reflect the inflammatory response, and brain abscess or other parenchymal foci may not trigger a CSF pleocytosis. Viral encephalitis can also be present even when there is no CSF pleocytosis.

Protein values also fall after the first week of life ( ). These authors found that the mean CSF protein in the first week was 1.06 g/l, with the 90th percentile 1.53 g/l. The value fell to around 0.6 g/l thereafter, with the 90th percentile less than 1 g/l. The protein value does depend on the red cell count, and in older children it has been said that every additional 1000 RBC/mm ³ adds 1 mg/dl (0.1 g/l) to the protein concentration.

CSF glucose is usually 70–80% of plasma glucose, and at least 50%. Take the plasma glucose sample before doing the lumbar puncture to avoid the effect of stress. The concentration of glucose does not seem to be affected in the short term by the presence of red cells, but low CSF glucose levels can develop and persist for weeks after intraventricular haemorrhage ( ).