Cerebrospinal Fluid Analysis

Brian Shahan, MD, and Edwin Y. Choi, MD, MS, Womack Army Medical Center, Fort Bragg,
North Carolina; Uniformed Services University of the Health Sciences, Bethesda, Maryland
Gilberto Nieves, MD, Womack Army Medical Center, Fort Bragg, North Carolina

Cerebrospinal fluid (CSF) analysis is a diagnostic tool for many conditions affecting the central nervous
system. Urgent indications for lumbar puncture include suspected central nervous system infection or
subarachnoid hemorrhage. CSF analysis is not necessarily diagnostic but can be useful in the evaluation
of other neurologic conditions, such as spontaneous intracranial hypotension, idiopathic intracranial
hypertension, multiple sclerosis, Guillain-Barré syndrome, and malignancy. Bacterial meningitis has
a high mortality rate and characteristic effects on CSF white blood cell counts, CSF protein levels,
and the CSF:serum glucose ratio. CSF culture can identify causative organisms and antibiotic sensitiv-
ities. Viral meningitis can present similarly to bacterial meningitis but usually has a low mortality rate.
Adjunctive tests such as CSF lactate measurement, latex agglutination, and polymerase chain reaction
testing can help differentiate between bacterial and viral causes of meningitis. Immunocompromised
patients may have meningitis caused by tuberculosis, neurosyphilis, or fungal or parasitic infections.
Subarachnoid hemorrhage has a high mortality rate, and rapid diagnosis is key to improve outcomes.
Computed tomography of the head is nearly 100% sensitive for subarachnoid hemorrhage in the first
six hours after symptom onset, but CSF analysis may be required if there is a delay in presentation or if
imaging findings are equivocal. Xanthochromia and an elevated red blood cell count are characteristic
CSF findings in patients with subarachnoid hemorrhage. Leptomeningeal carcinomatosis can mimic
central nervous system infection. It has a poor prognosis, and large-volume CSF cytology is diagnostic.
(Am Fam Physician. 2021;103(7):422-428. Copyright © 2021 American Academy of Family Physicians.)

Cerebrospinal fluid (CSF) analysis is a diag- CSF analysis is required to identify the causative
nostic tool for many conditions affecting the organism and select appropriate antimicrobial
central nervous system (CNS). The correct selec- coverage. An article on meningitis has been pub-
tion and interpretation of CSF tests can assist lished previously in American Family Physician.7
with diagnosing and treating a variety of con- When available, polymerase chain reaction
ditions. Table 1 lists the normal values of CSF (PCR) testing should be used to rapidly diag-
components.1-4 nose viral meningitis, which allows for early dis-
continuation of antibiotics.8-11 It can also detect
CNS Infections common viral and bacterial etiologies with more
Suspected CNS infection is a medical emergency than 95% sensitivity and specificity,8,12 whereas
and is the most common indication for lumbar viral cultures require days to weeks and have
puncture. Bacterial meningitis has a 14% to 25% lower accuracy.13 However, PCR testing does not
mortality rate; therefore, rapid CSF evaluation and assess drug sensitivities, is not universally avail-
early empiric antibiotic treatment are critical.5,6 able, and does not detect every infectious organ-
ism. Table 2 lists other suggested tests for various
types of CNS infections.1,8,14-25
Additional content at https://www.aafp.org/
afp/2021/0401/p422.html.
BACTERIAL MENINGITIS
CME This clinical content conforms to AAFP
criteria for CME. See CME Quiz on page 401. Opening pressure is the first CSF component
Author disclosure: No relevant financial
assessed when performing a lumbar puncture.
affiliations. It is obtained before draining CSF fluid and
should be measured with the patient in the lateral

422 American
Downloaded Family
from the Physician
American Family Physician website at www.aafp.org/afp. Copyright © 2021 American Academy of Volume
www.aafp.org/afp Family Physicians.
103, NumberFor the
7 private,
◆ Aprilnoncom-
1, 2021
mercial use of one individual user of the website. All other rights reserved. Contact [email protected] for copyright questions and/or permission requests.
 CSF ANALYSIS
SORT: KEY RECOMMENDATIONS FOR PRACTICE

Evidence
Clinical recommendation rating Comments

To minimize unnecessary antibiotic use, polymerase B Consistent results from cohort studies
chain reaction testing should be used to evaluate for viral showing reduced emergency depart-
infections in all patients with suspected meningitis.9-11,28 ment stays and no change in mortality

Gram stain testing of cerebrospinal fluid should be per- C Expert opinion and consensus guide-
formed in all patients with suspected meningitis. 28,29 lines in the absence of clinical trials

Patients with signs and symptoms of subarachnoid C Expert opinion and consensus guide-
hemorrhage who present more than six to 12 hours after line in the absence of clinical trials
symptom onset should undergo cerebrospinal fluid anal-
ysis if computed tomography findings are equivocal.48

A = consistent, good-quality patient-oriented evidence; B = inconsistent or limited-quality patient-oriented evidence;

C = consensus, disease-oriented evidence, usual practice, expert opinion, or case series. For information about the SORT
evidence rating system, go to https://www.aafp.org/afpsort.

decubitus position. Opening pressure is greater stain testing is recommended for all patients
than 300 mm H2O in 39% of patients with bac- with suspected CNS infection.28,29 It has a speci-
terial meningitis.26 The color of the CSF super- ficity of 97% for bacterial meningitis and a sensi-
natant can be cloudy, green, or purulent.8 The tivity of 33% to 90%,8,27,33 depending on the type
white blood cell (WBC) count is often greater of pathogen, cytocentrifugation, infection sever-
than 1,000 per µL, and polymorphonuclear leu- ity, laboratory personnel experience, the number
kocytes are typically predominant.27 However, of slides prepared, and the patient’s exposure
6% of patients with culture-diagnosed bacterial to antibiotics.27,31,34,35 Therefore, a negative CSF
meningitis do not have an elevated
WBC count, and lymphocytes are pre-
dominant in up to 10% of patients.28,29 TABLE 1
WBC elevations can be caused by trau-
matic lumbar punctures rather than Normal Values of CSF Components
infections. The classic method to cor- Component Adults and children Neonates
rect for traumatic WBC elevations has
Color Clear Clear
been to subtract one WBC for every
500 to 1,500 red blood cells (RBCs) ; a
26
CSF:serum 0.44 to 0.90 0.42 to 1.10
more accurate method is to use the for- glucose ratio
mula WBCs (predicted) = CSF RBCs × Differential 70% lymphocytes, 30% mono- PMN count may be
(blood WBCs/blood RBCs).30 cytes, rare PMNs or eosinophils normal
Nearly all patients with bacterial
Gram stain Negative for organisms Negative for organisms
meningitis have an elevated CSF pro-
tein level. The ratio of CSF glucose
8
Lactate level* 11.7 to 21.6 mg per dL (1.3 to 8.1 to 22.5 mg per dL
to serum glucose can be normal to 2.4 mmol per L) (0.9 to 2.5 mmol per L)
significantly decreased depending on Opening Adults and children 8 years and 10 to 100 mm H 2O
the type of pathogen, time since infec- pressure older: 60 to 250 mm H 2O
tion onset, and presence of dextrose Children younger than 8 years:
in any intravenous fluids the patient 10 to 100 mm H 2O
has received.31 Bacterial meningitis Protein level* < 50 mg per dL (500 mg per L) ≤ 150 mg per dL
can be diagnosed with more than 99% (1,500 mg per L)
certainty in patients with any of the
White blood < 5 per µL < 20 per µL
following CSF measurements: glucose cell count*
less than 34 mg per dL (1.89 mmol
per L), protein greater than 220 mg CSF = cerebrospinal fluid; PMN = polymorphonuclear leukocyte.

per dL (2,200 mg per L), WBCs greater *—Normal ranges may differ depending on laboratory calibration.
than 2,000 per µL, or neutrophils Information from references 1-4.
greater than 1,180 per µL.32 Gram

April 1, 2021 ◆ Volume 103, Number 7 www.aafp.org/afp American Family Physician 423
 CSF ANALYSIS
TABLE 2

CSF Characteristics by Infection Type

Infection type Differential Glucose level Opening pressure

Bacterial Usually 80% to 90% PMNs; < 40 mg per dL (2.22 mmol per L) in 50% Adults and children 8 years and older:
(typical)* > 50% lymphocytes possible to 60% of cases; CSF:serum glucose ratio 200 to 500 mm H 2O
< 0.4 is 80% sensitive and 98% specific Children younger than 8 years can have
lower pressures

Cryptococcal Lymphocyte predominance Usually > 40 mg per dL > 250 mm H 2O in severe cases; serial
lumbar punctures or ventriculoperi-
toneal shunt required to drain CSF if
pressure persistently > 250 mm H 2O

Fungal Possible early PMNs progressing Significant decrease possible Variable

(excluding to lymphocyte predominance;
cryptococcal) eosinophils possible

Neurosyphilis Variable Possibly decreased Usually elevated in immunocompetent

patients; may not be elevated in immu-
nocompromised patients

Parasitic Eosinophilia (> 10 eosinophils Usually low normal or normal Variable but can be persistently ele-
per µL or > 10% of total cells) vated, requiring CSF draining

Tuberculosis Early lymphocyte and PMN Median: 40 mg per dL; lower in Variable depending on stage
predominance progressing to advanced stages
lymphocyte predominance

Viral Lymphocyte predominance; Usually normal; decreased in 25% of Usually normal

possible PMN predominance in patients with mumps; mild decrease
early infection possible in patients with HIV infection

CSF = cerebrospinal fluid; PCR = polymerase chain reaction; PMN = polymorphonuclear leukocyte.
Information from references 1, 8, and 14-25.

Gram stain cannot exclude bacterial infection, depending on the organism, and false positives
particularly in patients who have already started have been reported.26,39 Latex agglutination is not
antibiotic therapy. recommended by the European Society of Clini-
Bacterial meningitis may still be present if cal Microbiology and Infectious Diseases, which
results of the CSF analysis are normal or only notes that this method is no more useful than
mildly abnormal. Further testing can help clar- Gram stain testing.29
ify the diagnosis if results of the CSF analysis CSF cultures can provide a definitive diagno-
are equivocal. The Infectious Diseases Society of sis and antimicrobial sensitivities. If antibiotic
America does not recommend routine measure- treatment is initiated before a CSF sample is
ment of CSF lactate in patients with suspected bac- obtained, the culture’s yield will be decreased by
terial meningitis, but two recent meta-analyses 10% to 20%.29
demonstrate its potential value.28,29,36 A CSF lac-
tate level greater than 35.1 mg per dL (3.9 mmol VIRAL MENINGITIS
per L) is 93% to 95% sensitive and 94% to 99% Viral meningitis is the most common type of
specific for diagnosing bacterial meningitis CNS infection.40,41 Opening pressure is typically
if the CSF is obtained before antibiotic treat- normal; the WBC count is generally less than 250
ment begins.37,38 The CSF lactate level is partic- per µL; and lymphocytes are typically predom-
ularly helpful in neurosurgical patients because inant, although polymorphonuclear leukocytes
inflammation from surgery can lead to unreli- may be predominant in the early phase.42 The
able measurements of conventional markers.36 CSF protein level can be normal to mildly ele-
Latex agglutination is another potentially useful vated (less than 150 mg per dL [1,500 mg per L]).31
adjunctive test for detecting bacterial antibodies The CSF glucose level is usually normal but can
or antigens in CSF. However, its sensitivity varies be mildly decreased; mumps infection can lower

424 American Family Physician www.aafp.org/afp Volume 103, Number 7 ◆ April 1, 2021
 CSF ANALYSIS

Protein level White blood cell count Other studies

Almost always elevated Usually 1,000 to 5,000 per µL Gram stain, CSF culture, CSF lactate (> 35.1 mg per dL [3.9 mmol
99% of children have > 100 per µL per L]), PCR testing; latex agglutination if Gram stain is negative
and antibiotics were given before lumbar puncture

Usually < 40 mg per dL Usually mildly elevated; normal CSF culture, CSF cryptococcal antigen test, India ink capsule
(400 mg per L) count possible, especially in patients stain, latex agglutination, enzyme immunoassay, lateral flow
with HIV infection assay, HIV test

50 to 250 mg per dL Usually elevated, up to several hun- CSF (1-3)-beta-D-glucan (elevated level is 95% to 100% sensi-
(500 to 2,500 mg per L) dred per µL tive and 83% to 99% specific), CSF fungal culture, Gram stain
(hyphae); PCR test is only 29% sensitive

> 45 mg per dL Early stage: 10 to 400 per µL HIV test, CSF Venereal Disease Research Laboratory test (30% to
(450 mg per L) Late stage: 5 to 100 per µL 75% sensitive and 100% specific†), CSF fluorescent treponemal
antibody absorption test (100% sensitive and 50% to 70% specific)
Declines over decades

Usually elevated 150 to 2,000 per µL PCR test; enzyme-linked immunosorbent assay for Angiostron-
gylus cantonensis, Gnathostoma spinigerum, and Baylisascaris
procyonis

Usually 100 to 200 mg per Usually 5 to 300 per µL; 500 to Multiple cultures with acid-fast stain; PCR test (56% sensitive and
dL (1,000 to 2,000 mg per L) 1,000 per µL in 20% of cases 98% specific), CSF adenosine deaminase (> 10 U per L [166.67
nkat per L]); “pellicle” appearance of CSF‡

Normal or mildly elevated Usually 100 to 1,000 per µL; higher PCR test preferred; other tests include CSF lactate (low), Gram
in patients with enterovirus infection stain, CSF or serum immunoglobulin M antibodies for arbo-
(elevated red blood cell count possi- viruses, electroencephalography or other neuroimaging for
ble in patients with herpes infection) suspected encephalitis

*—Most commonly Streptococcus pneumoniae, Haemophilus influenzae, Listeria monocytogenes, and Neisseria meningitidis.
†—False-positive results can occur if blood is present in the CSF.
‡—Cobweb-like clot that forms after CSF is allowed to stand for a short time.

the CSF glucose level by 25%.8,31 CSF and serum Subarachnoid Hemorrhage
immunoglobulin M antibody testing are the pre- Subarachnoid hemorrhage has a mortality rate
ferred diagnostic tests for arboviruses such as of 25% to 50% and requires urgent diagnosis.46
West Nile virus.15 Symptoms include sudden onset of severe head-
ache, neck stiffness, photophobia, nausea and
IMMUNOCOMPROMISE vomiting, seizure, and loss of consciousness.
Immunocompromised patients are at increased Most clinically significant subarachnoid hem-
risk of opportunistic infection. Adults older orrhages are apparent on computed tomogra-
than 60 years and people with deficiencies in phy of the head, which is nearly 100% sensitive
cell-mediated immunity have a higher incidence in the first six hours of symptoms46,47; after six to
of Listeria monocytogenes meningitis.43 Elevated 12 hours, the sensitivity wanes and CSF analysis
CSF protein and WBC measurements are sig- may be required for diagnosis.48
nificantly less likely in these patients, and Gram CSF is usually colorless in patients with sub-
stain is negative in 66% of cases.44 arachnoid hemorrhage, but the presence of as
Patients with acute retroviral syndrome due to few as 400 RBCs can change its appearance.47
early or acute HIV infection can present with an Enzymatic degradation of RBCs to produce bili-
aseptic CSF profile (mild lymphocyte elevation, rubin contributes to xanthochromia, a yellowish
normal to mild protein elevation, and a normal appearance of CSF. Xanthochromia begins to
to slightly low glucose level).8 Cryptococcal men- emerge after six hours in approximately 20% of
ingitis, toxoplasmosis, and neurosyphilis should subarachnoid hemorrhages and is present in 90%
be considered in patients with HIV and low CD4 of patients after 12 hours.46 At least 50% of patients
cell counts.45 Table 2 summarizes characteristics who present with a subarachnoid hemorrhage had
of CSF in patients with common infections.1,8,14-25 an initial subclinical sentinel bleed two to three

April 1, 2021 ◆ Volume 103, Number 7 www.aafp.org/afp American Family Physician 425
 CSF ANALYSIS

weeks earlier, which increases the likelihood of monitor CSF pressure, and no fluid should be
xanthochromia.46 Xanthochromia can persist for drained after the pressure decreases by 50%.54
up to two weeks after a bleeding event, and its
presence significantly decreases the probability Demyelinating Disease
of blood being present due to traumatic lumbar CSF analysis can assist in the diagnosis of neuro-
puncture.46 A CSF RBC count of less than 2,000 logic conditions such as multiple sclerosis (MS)
per µL in the absence of xanthochromia effec- and Guillain-Barré syndrome. Patients with
tively rules out subarachnoid hemorrhage.42 Other these conditions typically have an elevated CSF
causes of xanthochromia include a serum biliru- protein level but otherwise normal CSF compo-
bin level greater than 10 to 15 mg per dL (171.04 to sition. However, CSF analysis is only one compo-
256.56 µmol per L), CSF protein level greater than nent of these diagnoses; clinical and radiographic
150 mg per dL, or a traumatic lumbar puncture findings are also important considerations.55-58
with more than 100,000 RBCs per µL.1 Decreas- CSF-restricted oligoclonal bands are the clas-
ing RBC counts in successive CSF collection tubes sic CSF finding in patients with MS. Their high
strongly suggest a traumatic lumbar puncture.49 sensitivity contributed to recent revisions in
diagnostic criteria that allow the presence of
Spontaneous Intracranial Hypotension CSF-restricted oligoclonal bands to substitute
Intracranial hypotension, defined as an opening for the traditional dissemination-in-time cri-
pressure lower than 60 mm H2O in an adult, is terion, particularly in patients younger than 50
an uncommon finding. It can be spontaneous or years who have a clinical presentation otherwise
associated with a CSF leak, usually after a head typical of MS.55 However, oligoclonal bands are
injury.50 Spontaneous intracranial hypotension not necessarily specific for MS and can occur in
has emerged as a significant cause of secondary other inflammatory and infectious conditions.
headache syndrome, and its symptoms usually Elevations in CSF neurofilament light proteins
include a positional component.50 Case stud- correlate with relapsing MS.56
ies suggest that it may contribute to reversible The classic CSF finding in patients with
dementia.51 Although evidence shows that many Guillain-Barré syndrome is an elevated protein
patients with spontaneous intracranial hypo- level with a normal WBC count in the setting of
tension have low-normal opening pressures, less acute ascending areflexic paralysis. It is present in
than 60 mm H2O is considered diagnostic.50 50% of patients in the first week after presentation,
increasing to approximately 75% by three weeks.57
Idiopathic Intracranial Hypertension
Intracranial hypertension, defined as an opening Malignancy
pressure higher than 250 mm H2O in an adult, Leptomeningeal carcinomatosis is a metastatic
can be idiopathic or associated with infectious invasion of the leptomeninges that is difficult to
meningitis or intracranial masses. Idiopathic treat and has a poor prognosis.59 It can present
intracranial hypertension, also called pseudotu- with headache, confusion, and nuchal rigidity,
mor cerebri, should be considered in cisgender mimicking a CNS infection. CSF findings may
women of childbearing age who have obesity and also suggest infection: a decreased glucose level
chronic headaches; it is unclear whether this risk and elevated opening pressure, WBC count,
also pertains to trans men and nonbinary per- and protein level are common.59 Magnetic res-
sons with XX chromosomes. It has an annual onance imaging has limited diagnostic speci-
incidence of 19.3 cases per 100,000 women 20 to ficity, and single-sample CSF cytology has poor
44 years of age who have obesity and can result in sensitivity.60 Obtaining three high-volume lum-
permanent vision loss if untreated.52 Diagnosis is bar punctures (approximately 10 mL) improves
based on the modified Dandy criteria (eTable A), the sensitivity to approximately 90%, and flow
which include an elevated opening pressure and cytometry can improve specificity.59 Recent
normal CSF composition.53 When performing evidence suggests that tumor cell–associated
a lumbar puncture in a patient with an elevated biomarkers in CSF (circulating tumor DNA,
opening pressure, the manometer should be microRNA, and their respective metabolites)
used throughout the procedure to continuously can assist in the diagnosis and treatment of

426 American Family Physician www.aafp.org/afp Volume 103, Number 7 ◆ April 1, 2021
 CSF ANALYSIS

leptomeningeal carcinomatosis; research into 6. Thigpen MC, Whitney CG, Messonnier NE, et al.; Emerg-
their potential use is ongoing.60 ing Infections Programs Network. Bacterial meningitis in
the United States, 1998-2007. N Engl J Med. 2011;364(21):
This article updates a previous article on this topic by 2016-2025.
Seehusen, et al.42 7. Mount HR, Boyle SD. Aseptic and bacterial meningitis:
evaluation, treatment, and prevention. Am Fam Physician.
Data Sources: A literature search was performed in
2017;96(5):314-322. Accessed May 12, 2020. https://www.
PubMed using the terms cerebrospinal fluid and men-
aafp.org/afp/2017/0901/p314.html
ingitis. The search included meta-analyses, random-
8. Tunkel AR. Bacterial Meningitis. Lippincott Williams
ized controlled trials, clinical trials, and reviews. We
& Wilkins; 2001.
also searched Essential Evidence Plus, the Cochrane
9. Rice SK, Heinl RE, Thornton LL, et al. Clinical character-
database, and evidence reports from the Agency
istics, management strategies, and cost implications of
for Healthcare Research and Quality using the same a statewide outbreak of enterovirus meningitis. Clin Infect
terms. Search dates: August 2019 to July 2020. Dis. 1995;20(4):931-937.
The opinions or assertions contained herein are 10. Archimbaud C, Chambon M, Bailly JL, et al. Impact of
the private views of the authors and are not to be rapid enterovirus molecular diagnosis on the manage-
construed as official or as reflecting the views of the ment of infants, children, and adults with aseptic meningi-
Department of the Army, Department of Defense, or tis. J Med Virol. 2009;81(1):42-48.
the U.S. government. 11. Ramers C, Billman G, Hartin M, et al. Impact of a diagnos-
tic cerebrospinal fluid enterovirus polymerase chain reac-
tion test on patient management. JAMA. 2000;283(20):
The Authors 2680-2685.
12. Wu HM, Cordeiro SM, Harcourt BH, et al. Accuracy of
BRIAN SHAHAN, MD, FAAFP, DFPHM, is the pro- real-time PCR, Gram stain and culture for Streptococcus
gram director of the Family Medicine Hospitalist pneumoniae, Neisseria meningitidis and Haemophilus
Fellowship at Womack Army Medical Center, Fort influenzae meningitis diagnosis. BMC Infect Dis. 2013;
Bragg, N.C., and an assistant professor at the Uni- 13:26.
formed Services University of the Health Sciences, 13. Read SJ, Kurtz JB. Laboratory diagnosis of common viral
Bethesda, Md. infections of the central nervous system by using a sin-
gle multiplex PCR screening assay. J Clin Microbiol. 1999;
EDWIN Y. CHOI, MD, MS, FAAFP, is the associate 37(5):1352-1355.
program director of the Family Medicine Hospi- 14. Litvintseva AP, Lindsley MD, Gade L, et al. Utility of
talist Fellowship at Womack Army Medical Center (1-3)-β-D-glucan testing for diagnostics and monitoring
and an assistant professor in the Department of response to treatment during the multistate outbreak of
Family Medicine at the Uniformed Services Uni- fungal meningitis and other infections. Clin Infect Dis.
2014;58(5):622-630.
versity of the Health Sciences.
15. Tyler KL. Acute viral encephalitis. N Engl J Med. 2018;
GILBERTO NIEVES, MD, is a hospitalist at Wom- 379(6):557-566.
ack Army Medical Center. At the time this article 16. Ropper AH. Neurosyphilis [published correction appears
was written, he was a hospitalist fellow at Wom- in N Engl J Med. 2019;381(18):1789]. N Engl J Med. 2019;
ack Army Medical Center. 381(14):1358-1363.
17. Recio R, Perez-Ayala A. Cryptococcus neoformans menin-
Address correspondence to Brian Shahan, MD, goencephalitis. N Engl J Med. 2018;379(3):281.
Womack Army Medical Center, 2817 Reilly Rd., Ft. 18. Darras-Joly C, Chevret S, Wolff M, et al. Cryptococcus
Bragg, NC 28310 (email: brian.t.shahan.mil@mail. neoformans infection in France: epidemiologic features
mil). Reprints are not available from the authors. of and early prognostic parameters for 76 patients who
were infected with human immunodeficiency virus. Clin
Infect Dis. 1996;23(2):369-376.
References 19. Sawanyawisuth K, Sawanyawisuth K, Intapan PM, et al.
1. Fishman RA. Cerebrospinal Fluid in Diseases of the Ner- Specificity of immunoblotting analyses in eosinophilic
vous System. 2nd ed. Saunders; 1992. meningitis. Mem Inst Oswaldo Cruz. 2011;106(5):570-572.
2. Whiteley W, Al-Shahi R, Warlow CP, et al. CSF opening 20. Cinque P, Scarpellini P, Vago L, et al. Diagnosis of central
pressure: reference interval and the effect of body mass nervous system complications in HIV-infected patients:
index. Neurology. 2006;67(9):1690-1691. cerebrospinal fluid analysis by the polymerase chain reac-
3. Ahmed A, Hickey SM, Ehrett S, et al. Cerebrospinal fluid tion. AIDS. 1997;11(1):1-17.
values in the term neonate. Pediatr Infect Dis J. 1996;15(4): 21. Izzat NN, Bartruff JK, Glicksman JM, et al. Validity of the
298-303. VDRL test on cerebrospinal fluid contaminated by blood.
4. Leen WG, Willemsen MA, Wevers RA, et al. Cerebrospinal Br J Vener Dis. 1971;47(3):162-164.
fluid glucose and lactate: age-specific reference values 22. Ismail Y, Arsura EL. Eosinophilic meningitis associated with
and implications for clinical practice. PLoS One. 2012;7(8): coccidioidomycosis. West J Med. 1993;158(3):300-301.
e42745.
23. Dangoudoubiyam S, Vemulapalli R, Ndao M, et al.
5. Durand ML, Calderwood SB, Weber DJ, et al. Acute bacte- Recombinant antigen-based enzyme-linked immunosor-
rial meningitis in adults. A review of 493 episodes. N Engl
J Med. 1993;328(1):21-28.

April 1, 2021 ◆ Volume 103, Number 7 www.aafp.org/afp American Family Physician 427
 CSF ANALYSIS

bent assay for diagnosis of Baylisascaris procyonis larva 42. Seehusen DA, Reeves MM, Fomin DA. Cerebrospi-
migrans. Clin Vaccine Immunol. 2011;18(10):1650-1655. nal fluid analysis. Am Fam Physician. 2003;68(6):1103-
24. Chotmongkol V, Sawanyawisuth K, Thavornpitak Y. Corti- 1108. Accessed May 10, 2020. https://www.aafp.org/
costeroid treatment of eosinophilic meningitis. Clin Infect afp/2003/0915/p1103.html
Dis. 2000;31(3):660-662. 43. Koopmans MM, Brouwer MC, Bijlsma MW, et al. Listeria
25. Chander A, Shrestha CD. Cerebrospinal fluid adenosine monocytogenes sequence type 6 and increased rate of
deaminase levels as a diagnostic marker in tuberculous unfavorable outcome in meningitis: epidemiologic cohort
meningitis in adult Nepalese patients. Asian Pac J Trop Dis. study. Clin Infect Dis. 2013;57(2):247-253.
2013;3(1):16-19. 44. Mylonakis E, Hohmann EL, Calderwood SB. Central ner-
26. Bonadio WA. The cerebrospinal fluid: physiologic aspects vous system infection with Listeria monocytogenes. 33
and alterations associated with bacterial meningitis. Pedi- years’ experience at a general hospital and review of 776
atr Infect Dis J. 1992;11(6):423-431. episodes from the literature. Medicine (Baltimore). 1998;
77(5):313-336.
27. Fitch MT, van de Beek D. Emergency diagnosis and treat-
ment of adult meningitis. Lancet Infect Dis. 2007;7(3): 45. World Health Organization. Guidelines for HIV Diagnosis
191-200. and Monitoring of Antiretroviral Therapy. WHO Regional
Office for South-East Asia; 2004.
28. Tunkel AR, Hartman BJ, Kaplan SL, et al. Practice guide-
lines for the management of bacterial meningitis. Clin 46. Long B, Koyfman A, Runyon MS. Subarachnoid hemor-
Infect Dis. 2004;39(9):1267-1284. rhage: updates in diagnosis and management. Emerg Med
Clin North Am. 2017;35(4):803-824.
29. van de Beek D, Cabellos C, Dzupova O, et al.; ESCMID
Study Group for Infections of the Brain (ESGIB). ESCMID 47. Goyale A, O’Shea J, Marsden J, et al. Analysis of cerebro-
guideline: diagnosis and treatment of acute bacterial men- spinal fluid for xanthochromia versus modern computed
ingitis. Clin Microbiol Infect. 2016;22(suppl 3):S37-S62. tomography scanners in the diagnosis of subarachnoid
haemorrhage: experience at a tertiary trauma referral cen-
30. Mazor SS, McNulty JE, Roosevelt GE. Interpretation of
tre. Ann Clin Biochem. 2016;53(pt 1):150-154.
traumatic lumbar punctures: who can go home? [pub-
lished correction appears in Pediatrics. 2003;111(5):1127]. 48. Edlow JA, Panagos PD, Godwin SA, et al.; American Col-
Pediatrics. 2003;111(3):525-528. lege of Emergency Physicians. Clinical policy: critical
issues in the evaluation and management of adult patients
31. Dougherty JM, Roth RM. Cerebral spinal fluid. Emerg Med
presenting to the emergency department with acute
Clin North Am. 1986;4(2):281-297.
headache. Ann Emerg Med. 2008;52(4):407-436.
32. Spanos A, Harrell FE Jr, Durack DT. Differential diagnosis
49. Edlow JA, Caplan LR. Avoiding pitfalls in the diagnosis of
of acute meningitis. An analysis of the predictive value of
subarachnoid hemorrhage. N Engl J Med. 2000;342(1):
initial observations. JAMA. 1989;262(19):2700-2707.
29-36.
33. van de Beek D, de Gans J, Spanjaard L, et al. Clinical fea-
50. Kranz PG, Malinzak MD, Amrhein TJ, et al. Update on
tures and prognostic factors in adults with bacterial men-
the diagnosis and treatment of spontaneous intracranial
ingitis [published correction appears in N Engl J Med.
hypotension. Curr Pain Headache Rep. 2017;21(8):37.
2005;352(9):950]. N Engl J Med. 2004;351(18):1849-1859.
51. Hong M, Shah GV, Adams KM, et al. Spontaneous intra-
34. Brouwer MC, Thwaites GE, Tunkel AR, et al. Dilemmas
cranial hypotension causing reversible frontotemporal
in the diagnosis of acute community-acquired bacterial
dementia. Neurology. 2002;58(8):1285-1287.
meningitis. Lancet. 2012;380(9854):1684-1692.
52. Wall M, Kupersmith MJ, Kieburtz KD, et al.; NORDIC Idio-
35. Kaplan SL. Clinical presentations, diagnosis, and prognos-
pathic Intracranial Hypertension Study Group. The idio-
tic factors of bacterial meningitis. Infect Dis Clin North
pathic intracranial hypertension treatment trial: clinical
Am. 1999;13(3):579-594.
profile at baseline. JAMA Neurol. 2014;71(6):693-701.
36. Tunkel AR, Hasbun R, Bhimraj A, et al. 2017 Infectious Dis-
53. Burkett JG, Ailani J. An up to date review of pseudotu-
eases Society of America’s clinical practice guidelines for
mor cerebri syndrome. Curr Neurol Neurosci Rep. 2018;
healthcare-associated ventriculitis and meningitis. Clin
18(6):33.
Infect Dis. 2017;64(6):e34-e65.
54. Conly JM, Ronald AR. Cerebrospinal fluid as a diagnostic
37. Huy NT, Thao NT, Diep DT, et al. Cerebrospinal fluid lac-
body fluid. Am J Med. 1983;75(1B):102-108.
tate concentration to distinguish bacterial from aseptic
meningitis: a systemic review and meta-analysis. Crit Care. 55. Solomon AJ, Naismith RT, Cross AH. Misdiagnosis of mul-
2010;14(6):R240. tiple sclerosis: impact of the 2017 McDonald criteria on
clinical practice. Neurology. 2019;92(1):26-33.
38. Sakushima K, Hayashino Y, Kawaguchi T, et al. Diagnostic
accuracy of cerebrospinal fluid lactate for differentiating 56. Novakova L, Zetterberg H, Sundström P, et al. Monitoring
bacterial meningitis from aseptic meningitis: a meta-anal- disease activity in multiple sclerosis using serum neurofila-
ysis. J Infect. 2011;62(4):255-262. ment light protein. Neurology. 2017;89(22):2230-2237.
39. Saha SK, Darmstadt GL, Yamanaka N, et al. Rapid diagnosis 57. Yuki N, Hartung HP. Guillain-Barré syndrome [published
of pneumococcal meningitis: implications for treatment correction appears in N Engl J Med. 2012;367(17):1673].
and measuring disease burden. Pediatr Infect Dis J. 2005; N Engl J Med. 2012;366(24):2294-2304.
24(12):1093-1098. 58. Brettschneider J, Petzold A, Süssmuth S, et al. Cerebrospi-
40. Kupila L, Vuorinen T, Vainionpää R, et al. Etiology of asep- nal fluid biomarkers in Guillain-Barré syndrome—where do
tic meningitis and encephalitis in an adult population. we stand? J Neurol. 2009;256(1):3-12.
Neurology. 2006;66(1):75-80. 59. Wang N, Bertalan MS, Brastianos PK. Leptomeningeal
41. Martin NG, Iro MA, Sadarangani M, et al. Hospital admis- metastasis from systemic cancer: review and update on
sions for viral meningitis in children in England over five management. Cancer. 2018;124(1):21-35.
decades: a population-based observational study. Lancet 60. Zorofchian S, Iqbal F, Rao M, et al. Circulating tumour
Infect Dis. 2016;16(11):1279-1287. DNA, microRNA and metabolites in cerebrospinal fluid

428 American Family Physician www.aafp.org/afp Volume 103, Number 7 ◆ April 1, 2021
 BONUS DIGITAL CONTENT
CSF ANALYSIS

eTABLE A

Modified Dandy Criteria for the Diagnosis

of Idiopathic Intracranial Hypertension
CSF opening pressure > 250 mm H 2O

Neuroimaging shows no etiology for intracranial hypertension

No impaired consciousness

No other cause of intracranial hypertension

No other neurologic abnormalities (except abducens nerve palsy)

Normal CSF composition

Signs and symptoms of increased intracranial pressure (e.g., head-

ache, transient visual obscurations, pulsatile tinnitus, papilledema,
vision loss)

CSF = cerebrospinal fluid.

Information from Wall M, Corbett JJ. The modified Dandy crite-
ria for idiopathic intracranial hypertension, no need to fix what is
not broken. Neurology. January 10, 2014. Accessed May 10, 2020.
https://n.neurology.org/content/modified-dandy-criteria-idiopathic-
intracranial-hypertension-no-need-fix-what-not-broken

April 1, 2021 ◆ Volume 103, Number 7 www.aafp.org/afp American Family Physician 428A
Reproduced with permission of copyright owner. Further reproduction
prohibited without permission.