What's in a Tap: CSF Analysis

A Little Background on Cerebral Spinal Fluid Physiology

Cerebral spinal fluid (CSF) is created in the choroid plexus, which is found in the lateral, third, and fourth ventricles.  It then circulates through the ventricles before exiting the fourth ventricle via the Foramen of Luschka and Foramen of Magendie into the subarachnoid space to flow over the surface of the brain and spinal cord.  CSF flows cranially to caudally due to the pulsatile waves created by cerebral arteries (1). 

The average adult has 125 mL – 150 mL of CSF and in the absence of pathology this is constantly made by the choroid plexus and resorbed by the arachnoid villi at the rate of about 20 mL per hour.  Several disease states can interfere with this process, for instance, the arachnoid villi can be clogged by extra particles in the CSF such as WBCs and bacteria in infection or RBCs in hemorrhage, which leads to decreased reabsorption of CSF and thus, increased pressure (1).  Normal CSF pressure can range from 7 – 18 cm H2O, as measured when the patient is lying down (2).

A Word on the Blood Brain Barrier

The term “Blood Brain Barrier” is used to describe both the blood CSF barrier and the blood brain barrier, which are technically two distinct structures.  The blood brain barrier is made up of endothelial cells and astrocytes that line capillary walls in the brain.  The blood CSF barrier is made up of the choroid epithelial cells in the choroid plexus where CSF is formed.  Both structures serve to maintain the integrity of the central nervous system and prevent entry of cells, pathogens, and other particles into the brain that may have open access to the rest of the body (1).

The Blood Brain Barrier can be violated, most famously by pathogens in the development of meningitis.  The way that pathogens violate the blood brain barrier is not entirely understood. Theories include take over of native receptors on endothelial cells by surface markers on bacteria that lead to transport of bacteria into the CSF, use of bacterial pili to gain access and transport into the CSF, and transport of pathogens into the CSF by hitching a ride inside circulating cells such as monocytes (1).

Now that we have discussed the creation and location of CSF, a few words on some of the common tests obtained on CSF following lumbar puncture.

Color and clarity: CSF should be clear and colorless.  CSF can appear turbid as opposed to clear with as few as 200 WBCs/microL or 400 RBCs/microL.  It takes about 6000 RBCs per microL to make CSF look bloody.  Xanthochromia is the yellowish tinge to CSF that occurs as RBCs break down to bilirubin.  Xanthochromia is classically used to distinguish between traumatic tap and SAH because it does not start to appear until 2 to 4 hours after the RBCs have entered the CSF.  However, absence of xanthochromia does not rule out SAH, especially if the patient presents shortly after onset of symptoms, around 90% of patients have xanthochromia at 12 hours after onset of bleeding (1).  One study showed that 20% of patients had xanthochromia when tapped within 6 hours of onset of symptoms, 65% had xanthochromia when tapped between 6-12 hours post ictus, and 100% had xanthochromia when tapped between 12 hours and 14 days (3, 4).  Xanthochromia can also be caused by elevated CSF protein, or systemic hyperbilirubinemia (1).

Cell Count:  CSF should be acellular, but when looking at a sample obtained by lumbar puncture up to 5 RBCs and up to 5 WBCs is considered normal.

Warning: Cell count may actually be falsely low if measured more then 1 hour after collection because the cells may settle or adhere to the tubes!!

This gets a little hairy in the setting of a traumatic tap, if RBCs have gotten in, WBCs may obviously get in as well.  Assuming the WBC count in the blood is normal, you are allowed 1 WBC for every 500 RBCs (2, 4).  The cell count is also repeated in the fourth tube of CSF collected to check to see if cells from a traumatic tap have “cleared” indicating the SAH or other pathologic source of cells is less likely (1).   

Glucose: CSF glucose levels are typically greater than 0.6x the serum glucose level (roughly two thirds).  Low CSF glucose typically indicates bacterial infection, and interestingly normalizes before cell count or protein levels (1).

Protein: Normal CSF protein in adults typically ranges from 23-38 mg/dL.  When CSF protein is increased due to bleeding such as SAH, it is expected to rise 1 mg/dL for every 1000 RBCs per microL.  Protein is also increased in infection with infection with tuberculosis showing higher levels then bacterial infection, which shows higher levels then viral infection (1).

1.     Tips for Interpreting the CSF Opening Pressure.  Retrieved April 15, 2017 from, https://www.aliem.com/2016/08/tips-for-interpreting-the-csf-opening-pressure/.

2.     Shah, KH. Et al. “Distinguishing traumatic lumbar puncture from true subarachnoid hemorrhage”.  The Journal of Emergency Medicine. 2002;23:67-74.

Now for a couple of cases to demonstrate this information in action

Case 1

It’s a sweltering August day in Cincinnati, and a 25 year old male, previously healthy, presents to the ED complaining of 4 days of headache, fever, and malaise.  On exam he looks uncomfortable but is awake and alert, and maybe cannot range his neck quite as much as you think that he should, thus you determine that he needs an LP to look for signs of meningitis.  CSF analysis shows the following:

  • Opening pressure: normal
  • WBC count: 10 with 7 lymphocytes
  • RBC count tube 1: 0
  • Protein: 65 mg/dL
  • Glucose: 75 mg/dL

When you consider all of the above, the patient most likely has a viral meningitis, particularly an enterovirus given that this illness is most common in the Summer and Fall months, and enteroviruses make up 90% of cases of viral meningitis (1).  However, what components of the history, physical, or lab evaluation suggest that, in the right clinical context, the patient may benefit from further testing of the CSF?

Case 2

A 50 year old female with a history of hypertension presents to the ED with compliant of “worst headache of my life”.  She reports that it was maximal at onset and has been persistent for the last 12 hours.  She indicates that she did not come in soon because she “figured it would just go away”.  Physical exam is unremarkable. Noncontrast head CT was unremarkable thus you proceed to LP.  It was a difficult tap given that the patient had an unfavorable body habitus, thus you are not surprised when results suggest traumatic LP with 5,000 RBCs in the first tube.  However, you note that there are still 700 RBCs in the fourth tube.  You are perplexed as the RBCs are clearing, suggesting traumatic tap, but still elevated in the fourth tube suggesting SAH.

What other studies obtained during LP may help reach the correct diagnosis?

What substance, if found in the CSF, strongly suggests SAH as opposed to traumatic LP?

Now consider the same clinical scenario, however, you are in a rural clinic in Northern Alaska and you have the equipment to perform an LP, but are unable to get lab analysis until the biweekly courier plane picks up the sample.  What common, easy to use bedside test may help you determine if this patient needs to be sent to a higher level of care?

Answers to these cases will be posted in the Grand Rounds Recap on 4/28


  1. Cerebrospinal fluid: Physiology and utility of an examination of disease states. Retrieved April 15, 2017 from, https://www.uptodate.com.
  2. Tips for Interpreting the CSF Opening Pressure.  Retrieved April 15, 2017 from, https://www.aliem.com/2016/08/tips-for-interpreting-the-csf-opening-pressure/.
  3. Shah, KH. Et al. “Distinguishing traumatic lumbar puncture from true subarachnoid hemorrhage”.  The Journal of Emergency Medicine. 2002;23:67-74.
  4. CSF Analysis. Retrieved April 15, 2017 from, https://lifeinthefastlane.com/ccc/csf-analysis/.