Kaplan: Clinical Chemistry, 5th Edition

Clinical References - Methods of Analysis

Cerebrospinal Fluid (CSF) Protein Quantitation

Danyel H. Tacker and Anthony O. Okorodudu i

Name: CSF Proteins

Clinical significance: Central nervous system (CNS) disease, trauma
Molecular formula: N/A – See information for individual proteins
Molecular mass: N/A – See information for individual proteins
Merck Index: N/A
Chemical class: Protein
Refer to Chapter 47, Nervous System, in the 5th edition of Clinical Chemistry: Theory, Analysis,
Correlation.

Students’ Quick Hyperlink Review

• Principles of analysis and current usage
• Reference and preferred methods
• Specimen
• Interferences
• CSF proteins reference intervals
• Interpretation
• CSF proteins assay performance goals
• References
• Table of CSF protein reference intervals
• CSF protein method

Principles of Analysis and Current Usage

Cerebrospinal fluid (CSF) is the fluid in the ventricles of the brain, in the space between the
arachnoid and the pia mater, and surrounding the spinal cord. It is an ultrafiltrate of plasma that
passes through the blood-brain barrier (BBB) in the choroid plexus [1,2]. Selective transport and
diffusion of proteins, electrolytes, and other molecules and bloodborne factors occurs at the level
of this highly selective form of capillary endothelium. Thus CSF has protein constituents and an
electrophoretic pattern similar to that of plasma/serum but at vastly lower concentrations in

i
CSF Proteins
Previous and current authors of this method:
First edition: Gayle Birkbeck
Methods edition: Gayle Jackson
Second edition: Not updated
Third edition: Not updated
Fourth edition: Gayle Jackson
Fifth edition: Danyel H. Tacker, Anthony O. Okorodudu

Methods of Analysis © 2010 by Lawrence A. Kaplan and Amadeo J. Pesce.

normal health states. In disease states, the permeability and function of the BBB can be
compromised, increasing the protein concentrations of CSF relative to plasma/serum.
The choroid plexus is also the site of in-situ production and secretion of various proteins,
peptides, and hormones in healthy and disease states. Thus alterations in the protein
concentration of CSF relative to plasma/serum, as well as the appearance of proteins unique to
the CSF fluid, can give an indication of the general health of the central nervous system (CNS)
[2]. Finally, the appearance of “CSF-specific” proteins in non-CSF body fluids can indicate CNS
trauma, which can be helpful in diagnosis.

Like serum, CSF is a complex solution including ions, nutrients, and proteins. The entry of ions
present in CSF (H+, K+, Ca2+, Mg2+, bicarbonate, etc.) is specifically regulated by ion channels in
the membranes composing the BBB. However, nutrients and wastes (i.e., glucose, urea,
creatinine) diffuse freely through the BBB. Bilirubin is typically not present in CSF in
measurable quantities [1,2].

The proteins present in CSF, or which are measured in CSF, include:

• Albumin, which passively diffuses through the BBB. Albumin is the predominant protein
in the plasma, as well as the CSF, and has a primary function in CSF as a transporter of a
variety of compounds [1-4].
• Immunoglobulins, which are usually excluded from the CSF by the BBB.
Immunoglobulins that appear in CSF have two potential sources: (1) massive failure of
the BBB and “leakage” into the CSF from the plasma or (2) intrathecal production in
cases of CNS infection [1-3,5-8].
• Transthyretin/prealbumin, which is synthesized and secreted in situ. Transthyretin is a
shuttle protein and transporter of thyroid hormones; it complexes with retinol-binding
protein to aid in transport of vitamin A/retinols [1-3,5].
• β 2 -Transferrin, which is a desialated form of transferrin found only in the CNS and
functions in iron complexation and maintenance of metal ion redox status. β 2 -Transferrin
can be detected long (months and sometimes years) after CNS trauma in collected body
fluids [1-3,5,9,10].
• C-Reactive protein (CRP), which is usually found in very low concentrations or is
altogether absent from CSF. CRP is an acute-phase reactant and appears to have a role in
modulating inflammatory response [1,2,5,11].
• α 2 -Macroglobulin, which is normally excluded owing to its size. α 2 -Macroglobulin is a
protease inhibitor and an acute-phase protein. Presence of this protein in the CSF usually
only occurs when the integrity of the BBB has been heavily compromised [1-3,5].
• β-Amyloid and tau (τ) proteins, which are normally absent from CSF. β-Amyloid is a
cleavage product of amyloid precursor protein (function is not currently described),
which forms extracellular amyloid plaques when not properly removed from sites of
production. τ Protein normally associates with tubulin to stabilize microtubule structures
present in neurons; hyperphosphorylation of τ protein causes it to form massive
complexes, which result in pathological plaques that are neurotoxic. Both of these
proteins are associated with neurodegeneration [2].
• β 2 -Microglobulin, which is normally present in low concentrations and is largely an
investigational marker in the CSF. β 2 -Microglobulin composes part of the major

Methods of Analysis © 2010 by Lawrence A. Kaplan and Amadeo J. Pesce.

 histocompatibility complex I protein complex, which is responsible for “non-self”
recognition (primarily of viruses and tumor antigens). β 2 -Microglobulin is amyloidogenic
[1-3,5,12,13].
• Protein 14-3-3, which is normally absent from CSF. Protein 14-3-3 is a general name for
a highly conserved class of intracellular signaling and regulatory proteins. Presence of
14-3-3 protein in the CSF is observed in spongiform encephalopathy (e. g., Creutzfeldt-
Jakob disease) [14].
• Myelin basic protein, which is normally absent from CSF. MBP was originally isolated
from myelin sheaths. Its presence in the CSF is associated with multiple sclerosis
[5,7,15,16].
• Fibronectin, which is normally present in low concentrations. Fibronectin complexes with
integrins to stabilize the extracellular matrix of all tissues. Presence of increased
fibronectin in CSF is associated with trauma and disruption of the extracellular matrix of
the CNS [1,2,5,17].

CSF has a number of functions in the CNS [1,2]. Firstly, CSF surrounds the brain in the
subarachnoid space, fills the ventricles, and surrounds the spinal cord, providing a fluid layer of
insulation against mechanical trauma to the fragile tissues of the CNS. Since it completely
surrounds these tissues, CSF also gives tissues ready access to nutrients, hormones, and other
factors within the CNS compartment. Further, the ions present in CSF allow for maintenance of
the ambient ionic charge needed for efficient CNS function. Finally, because there is no lymph
drainage in the CNS, the CSF provides an important route for the excretion of CNS waste.

The constant production of CSF in the CNS also aids in the maintenance of intracranial pressure
and homeostasis and delays alterations of these when acute changes in peripheral pressure and
homeostasis occur. In short, CSF provides protection against abrupt pathological change because
of its isolated location and highly regulated composition.

Reference and Preferred Methods

Reference Method: N/A – Method used is dependent on the protein and type of analysis.

Preferred & Other Methods: Specific for each protein; see below.

Total Protein: [1-3,5,18-21]

Accurate measurement of CSF protein is difficult because CSF proteins are present in small
quantities and because qualitative differences exist between the amounts of immunoglobulin and
the amounts of albumin present in patient specimens. Methods used for CSF total protein
measurements can be sensitive but can give inaccurate results when the albumin-to-gamma
globulin ratio changes.

A. Preferred Method: Trichloroacetic Acid (TCA) + Nephelometric or Turbidimetric Analysis.

The TCA precipitates the protein, which is then read using nephelometric or turbidimetric
techniques. Light scatter or absorbance is proportional to protein concentration. Comments:
Simple and fast, but requires the largest sample volume. Alternate precipitating agents
include benzalkonium chloride and benzethonium chloride.

Methods of Analysis © 2010 by Lawrence A. Kaplan and Amadeo J. Pesce.

B. Preferred Method: Sulfosalicylic Acid (SSA): 3% Sulfosalicylic and 7% Sodium Sulfate;
Total Protein, Turbidimetric. The assay principle is based on precipitation of protein, with
the resultant turbidity measured in a spectrophotometer. Comments: CSF (500 μL),
commonly used; sulfosalicylic acid alone results in greater turbidity with albumin than with
globulin. The same phenomenon is seen with sulfosalicylic acid-sodium sulfate in
combination.
C. Historical Method: Biuret Method/Colorimetric Analysis. Copper ions in the Biuret solution
interact with amide groups (e.g., lysine) on amino acid chains. Albumin has the best
binding/detection. Comments: Rare; low color yield restricts usage to very sensitive
instruments with rate-measurement capability.
D. Historical Preferred Method: Lowry Method. Two-step assay that incorporates the Biuret
reaction with a second, Folin-Ciocalteu reagent step (whereby molybdate and tungstate bind
to tryptophan and tyrosine residues on the protein). This dual reaction gives greater
sensitivity but preferentially binds to albumin. Comments: CSF (200 μL), rarely used; time
consuming; influenced by endogenous phenols and drugs.
E. Other Method: Coomassie Brilliant Blue Dye Binding; Total Protein, Chemical,
Spectrophotometric. In the assay, Coomassie Brilliant Blue G-250 binds to protein, with a
resultant color change from brownish orange to an intense blue color, and the absorbance of
the dye shifts from 465 to 595 nm. Comments: CSF (25 to 100 μL), infrequently used;
problem with standardization; different proteins show variability in sensitivity and variation
in standard curves.

Specific Proteins: [1-8,11-19,22-37]

A. Immunonephelometry – light scatter is proportional to protein-antibody complex
concentration. Typically, the target protein is selectively bound by a specific antibody, and
the resulting antigen-antibody complex increases the light scatter of the sample. The
scattered light is detected at an angle >180 degrees from the incident light beam passing
through the sample.

B. Immunoassay (IA) – protein-specific antibodies bind target proteins in the CSF sample.
Depending on the type of IA, tagging of the detection antibody, and other analytical
specifications, the signal-to-concentration relationships vary. Examples of IA used in the
above-referenced publications include:
1. IA after high-performance affinity chromatography (for ultra-low, specific IgG
detections)
2. Enzyme IA – an example is fibronectin, but any enzyme-linked specific antibody used
for detection falls in this category. Enzyme-linked detection antibodies react with a
substrate added after the detection antibody has complexed with the target protein and
excess has been washed away. The reaction with the substrate produces a signal
(typically colorimetric) that is proportional to the target protein concentration and rate-
limited by the presence of the enzyme after the washing steps.
3. Radio IA – an example is β 2 microglobulin, but new and emerging assays often start
with RIA and then progress to other IA techniques. RIA is typically competitive in
nature, whereby a radioactive, tagged target protein is added to the sample. The tagged
protein and the native protein compete for antibody binding (typically to a solid
substrate). After thorough washing, the radioactivity on the substrate is measured,

Methods of Analysis © 2010 by Lawrence A. Kaplan and Amadeo J. Pesce.

 usually via scintillation counting. Owing to the nature of competitive IA, concentration
of the target protein in the sample is typically inversely proportional to the
signal/scintillation detected.
4. Electrochemiluminescence (ECLIA) and related immunoassay technologies are used for
measuring specific proteins. A concentration step is usually necessary for CSF protein
quantitation. Depending on the size of the target, the detection can either be
proportional/indirect, or inversely proportional/competitive. For most proteins, the target
is large enough that the assay is indirect. Using this approach, a trapping antibody is
bound to a substrate (often a flat detection surface or a paramagnetic bead), and the
target protein in the sample is first incubated with the trapping antibody. A second
antibody (the detection antibody, which is tagged with a chemiluminescent compound)
is added to the mixture for a second incubation. Excess sample is washed away or
removed in a flow cell, and a substrate for the chemiluminescent compound is applied.
After electrical stimulation of the sample, the chemiluminescent compound is excited
and gives off photons of light. These photons are counted, and the amount of signal is
proportional to target protein concentration.

C. High-performance liquid chromatography (HPLC) with coulometric or other detection

method is typically used for peptides and hormones present in very low concentrations.
With HPLC, processed samples are injected into a column under high pressure and
carried by a liquid phase. Depending on the column, the liquid phase, and the protein
chemistry, fractions of proteins in the sample will elute at different times. These fractions
can be subsequently analyzed in a variety of ways. Use of internal standards and in-run
calibrators can allow for quantitation.

Electrophoretic Patterns (Semiquantitative): [3,7-11,25,36,38-40]

A concentration step is necessary for this procedure. Polyacrylamide and other gel techniques are
used for electrophoretic separation of protein bands in serum and CSF. In short, CSF is applied
to an agarose gel and separated in an electric field through a liquid buffer phase. Dye-binding
techniques (utilizing bromocresol purple, Coomassie Brilliant Blue, and other dyes) are used for
visualization and densitometric quantitation. Densities of bands are used to express protein
fraction concentration based on their percentages of total protein, usually measured via
nephelometric or turbidimetric detection (above).

The most common application of CSF electrophoresis is for the visualization of “oligoclonal”
bands in the gamma region of the gel. Banding in the gamma region suggests that in-situ
expression of monoclonal antibodies is taking place and occurs in a variety of CNS diseases,
including multiple sclerosis and malignancy.

Specialized/Emerging/Research Techniques: [15,34,38]

A. Enzyme-linked immunosorbent assay (ELISA) and related methods are used for specialized
testing of specific and uncommon proteins. ELISA is a common “first step” in research
analysis and often expands to automated IA. IAs were discussed previously.

B. Specialized detections using Western blot are usually limited to research but are possible as
long as specific antibodies are available. In Western blotting, proteins are extracted from the

Methods of Analysis © 2010 by Lawrence A. Kaplan and Amadeo J. Pesce.

 sample and concentrated. The concentrated protein extract is electrophoretically separated
using an agarose gel preparation then incubated with specific antibodies (which are typically
linked to enzymes). Colorimetric detection is then possible, as is semiquantitation using
densitometric analysis of protein bands.

Specimen
Preparation: [2]
There are many side effects associated with the collection of CSF (typically via lumbar puncture,
also called a spinal tap) that are undesirable and even life-threatening. The risks associated with
CSF collection must be weighed against the need for diagnostic information, and volume of
collection must be kept to an absolute minimum for the testing required. Patients should remain
calm. Opening pressure should be between 90 and 180 mm of water (for normal adults) but is
dependent on postural changes and elevations, as well as patient physiological status and age [2].
An opening pressure greater than 200 mm of water in a relaxed patient suggests serious disease
and limits CSF collection volume to 2.0 mL.

Draw/Collection: [2,5]
When opening pressure allows for full collection, three tubes are typically drawn. The first tube
is used for chemistry and immunology (unless traumatic; bloody specimens will affect protein
measurements and determinations), the second tube is for culturing, and the third tube is for cell
counting/differentials. Tubes may be substituted or added, but volumes must be kept to a
minimum. Maximal volume drawn is 20 mL in an adult (total CSF volume typically 90 to 150
mL), and about 3 mL in a neonate (total CSF volume typically 10 to 60 mL). Collection of blood
samples for comparisons and index calculations is optimal; when collected 2 hours before the
CSF collection, the best comparisons are obtained (because of the delayed equilibrium of the
CSF).

Storage: [2]
CSF degradation begins within 1 hour, so all efforts should be made to promptly store the sample
tubes properly. All samples should be centrifuged upon receipt. If analysis cannot take place
immediately, store CSF at 4°C for < 72 hours. Freezing CSF at −20°C keeps constituents stable
for up to 6 months. Freezing CSF at −70°C keeps constituents stable indefinitely. Blood samples
should be handled and stored normally.

Patient Risks Involved: [2]

The most common untoward effect of lumbar puncture is headache. Infections, meningitis, and
death have occurred as a result of lumbar puncture procedures.

Interferences [1,2]
The location and specific functions of CSF make it particularly susceptible to diurnal variation
and hormonal factors, which can affect specific protein levels. Such influences should be
considered accordingly when analyzing CSF.

Depending on the assay used for CSF analysis, a number of factors could cause interference. For
example, colorimetric techniques employing spectrophotometric detection can be susceptible to
interference from hemolysis, icterus, uremia, drugs with spectral activity, and high ammonium

Methods of Analysis © 2010 by Lawrence A. Kaplan and Amadeo J. Pesce.

concentrations (for Biuret and Lowry reactions). Also, other conditions or therapies (i.e., dextran,
contrast media, and high concentrations of immunoglobulins) that increase turbidity of the CSF
can affect nephelometric and turbidimetric analysis.

CSF Proteins Reference Intervals

Reference Intervals: Please consult Table 1: Available Reference Intervals for CSF Proteins

Calculations: [1-3,20]

A. Albumin Index

Albumin Index = CSF albumin (mg/dL)__

Serum albumin (mg/dL)

Interpretation: For adults, an albumin index < 9 is normal. An index of 9 to 14 reflects

slight impairment of the BBB. An index of 14 to 30 reflects moderate impairment of the
BBB, and an index > 30 reflects severe impairment of the BBB.

B. IgG Ratio

IgG Ratio = CSF IgG (mg/dL)__

Serum IgG (mg/dL)

Interpretation: For adults, an IgG ratio of 3.0 to 8.7 is normal. IgG ratios > 8.7 reflect
BBB impairment.

C. CSF IgG Index

CSF IgG Index = [CSF IgG (mg/dL) / serum IgG (g/dL)]

[CSF Alb (mg/dL) / serum albumin (g/dL)]

Interpretation: For adults, a CSF IgG index < 0.8 is normal. A CSF IgG index > 0.8
suggests intrathecal IgG production.
NOTE: With additional time-sensitive information about the samples used, IgG synthesis
rates may be calculated.

Interpretation and Clinical Utility [1,2,14,22,38]

Interpretation of CSF Electrophoresis:
Normally, CSF samples show an evident transthyretin/prealbumin band (typically 2% to 7% of
total sample protein content), followed by bands for albumin (56% to 76%), α 1 (2% to 7%), α 2
(4% to 12%), and β 1 /β 2 (8% to 18%) globulins. The γ region (3% to 12% of total sample protein
content) should be faintly staining and diffuse.

If the γ region shows multiple distinct bands, this is called an oligoclonal banding pattern.
Oligoclonal bands are typically associated with demyelinating disease. However, if the γ region

Methods of Analysis © 2010 by Lawrence A. Kaplan and Amadeo J. Pesce.

shows only a single band, this pattern is typically associated with infectious disease, such as
meningitis.

When CSF electrophoresis is performed alongside a serum sample from the same patient,
collected at generally the same time as the CSF sample, then an assessment of BBB integrity can
be made. Typically, the CSF lane on the electrophoresis will be faint when compared to the
serum lane. If the CSF lane appears to be equal in density or darker than the serum lane, then the
BBB is probably compromised. Calculation of the CSF IgG index and/or the albumin index
should always accompany CSF electrophoretic analysis; the calculations give a more exact
assessment of BBB integrity than the density of sample staining.

Disease-Specific Interpretation:
Infection:
In CNS infection, the BBB is compromised, and increased CSF protein levels relative to plasma
protein/serum levels are observed. Other indicators, such as elevation of CRP and
immunoglobulins, aid in diagnosis. Electrophoresis of CSF can reveal oligoclonal banding
patterns of intrathecally produced immunoglobulins.

CNS Trauma:
Subarachnoid hemorrhage and other CNS trauma increases BBB permeability. Total protein will
be elevated in the CSF specimen relative to plasma protein concentrations, and typically absent
components such as bilirubin and methemoglobin are detectable.

In cases of rhinorrhea or otorrhea, the presence of β 2 -transferrin is suggestive of CSF leakage

outside the CNS. Transthyretin detection is only useful in this type of analysis if present in very
high concentrations, since transthyretin is also seen in plasma.

Demyelinating Disease:
The intrathecal production of immunoglobulins in the choroid plexus occurs in demyelinating
diseases such as multiple sclerosis and produces oligoclonal banding patterns on electrophoresis.
The ratio of IgG in the CSF relative to that of the plasma will be elevated, and specific studies
for myelin basic protein in the CSF are typically positive.

Neoplastic Disease:
Different neoplasms produce different alterations of CSF protein content, the description of
which is beyond the scope of this summary. Markers used in diagnosing neoplasm in CSF fluid
include β 2 -microglobulin, fibronectin, myelin basic protein, and others.

Other CNS Disease:

Terminal spongiform encephalopathies are typically associated with elevations in Protein 14-3-3.
Imaging and case-relevant symptomatic information are usually used for diagnosis.

Alzheimer’s disease diagnosis is made with imaging results, case-relevant symptomatic

information, and prediction of early risk of disease by determining the ratio of τ protein to β 2 -
amyloid protein in CSF.

Methods of Analysis © 2010 by Lawrence A. Kaplan and Amadeo J. Pesce.

CSF Proteins Assay Performance Goals
Total protein analysis of CSF employs techniques also used for serum and plasma protein
quantitation. This is a benefit to CSF protein quantitation in that the available assays are largely
automated and come with reliable control materials. Typically, automated versions of the SSA
and TCA techniques show coefficients of variation (CVs) ≤ 5% and wide measuring ranges.

Immunoassay performance will vary according to the detection technique measured. For
automated IA assays, CVs tend to be slightly wider than the serum/plasma version of the assay.
Depending on the availability of a reference material for a given protein, degrees of assay
standardization, and thus the resulting assay performance, will vary.

Electrophoretic assay performance is largely based on satisfactory visualization of all relevant

bands when compared to control samples and serum samples accompanying the CSF sample on
the gel. When analysis is semiquantitative and tied to band density as a percent of total sample
protein concentration, then cumulative statistics can be generated and performance assessed.
CVs ≤ 10% for major bands and ≤ 20% for minor bands are generally acceptable.

HPLC technique performance, when made semiquantitative through the use of internal and assay
controls, varies according to the targets being detected and the degree of automation used in
sample preparation and assay.

References
1 Burtis CA, Ashwood ER, Bruns DE, eds. Tietz Textbook of Clinical Chemistry and
Molecular Diagnostics. 4th ed. Philadelphia: Saunders; 2006:577-595.
2 McPherson R, Pincus M, eds. Henry’s Clinical Diagnosis and Management by Laboratory
Methods. 21st ed. Philadelphia/China: Saunders; 2007:426-434.
3 Wise BL. The quantitation and fractionation of proteins in cerebrospinal fluid. Am J Med
Technol 1982;48:821-7.
4 Bonate PL. Quantification of albumin in cerebrospinal fluid. Anal Biochem 1988;175:300-
4.
5 Wu AHB, ed. Tietz Clinical Guide to Laboratory Tests. 4th ed. Philadelphia: Saunders;
2006.
6 Arseneault JJ. An enzyme immunoassay for the quantitation of IgG in serum and
cerebrospinal fluid. Clin Chim Acta 1980;107:73-84.
7 Gerson B, Cohen SR, Gerson IM, Guest GH. Myelin basic protein, oligoclonal bands, and
IgG in cerebrospinal fluid as indicators of multiple sclerosis. Clin Chem 1981;27:1974-7.
8 Matikainen MT. Solid-phase immunoassay methods for quantitation of IgG and viral
antibodies in cerebrospinal fluid and its electrophoretic fractions. J Neurosci Methods
1981;4:277-86.
9 Rouah E, Rogers BB, Buffone GJ. Transferrin analysis by immunofixation as an aid in the
diagnosis of cerebrospinal fluid otorrhea. Arch Pathol Lab Med 1987;111:756-7.
10 Zaret DL, Morrison N, Gulbranson R, Keren DF. Immunofixation to quantify beta 2-
transferrin in cerebrospinal fluid to detect leakage of cerebrospinal fluid from skull injury.
Clin Chem 1992;38:1908-12.

Methods of Analysis © 2010 by Lawrence A. Kaplan and Amadeo J. Pesce.

11 Hansson LO, Lindquist L, Linne T, Sego E. Quantitation of C-reactive protein in
cerebrospinal fluid and serum by zone immunoelectrophoresis assay (ZIA). J Immunol
Methods 1987;100:191-5.
12 Mavlight GM, Stuckey SE, Cabanillas FF, Keating MJ, Tourtellotte WW, Schold SC,
Freireich EJ. Diagnosis of leukemia or lymphoma in the central nervous system by beta 2-
microglobulin determination. N Engl J Med 1980;303:718-22.
13 Lutz CT, Cornell SH, Goeken JA. Establishment of a reference interval for beta 2-
microglobulin in cerebrospinal fluid with use of two commercial assays. Clin Chem
1991;37:104-7.
14 Conly JM, Ronald AR. Cerebrospinal fluid as a diagnostic body fluid. Am J Med
1983;75:102-8.
15 Chamczuk AJ, Ursell M, O’Connor P, Jackowski G, Moscarello MA. A rapid ELISA-
based serum assay for myelin basic protein in multiple sclerosis. J Immunol Methods
2002;262:21-7.
16 Cohen SR, Herndon RM, McKhann GM. Radioimmunoassay of myelin basic protein in
spinal fluid. An index of active demyelination. N Engl J Med 1976;295:1455-7.
17 Gressner AM, Wallraff P. Laser nephelometric quantitation of fibronectin (cold-insoluble
globulin) in human cerebrospinal fluid and seminal plasma. Clin Chim Acta
1981;113:207-12.
18 Deaton CD, Maxwell KW, Smith RS, Creveling RL. Use of laser nephelometry in the
measurement of serum proteins. Clin Chem 1976;22:1465-71.
19 Liu TZ, Sun CF. Laser nephelometric quantitation of protein in urine and cerebrospinal
fluid. J Formos Med Assoc 1990;89:1067-71.
20 Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin
phenol reagent. J Biol Chem 1951;193:265-75.
21 Mauck JC. The quantitation of urinary and spinal fluid proteins by a modified biuret
method. Clin Biochem 1976;9:188-91.
22 Aksamit AJ. Cerebrospinal fluid 14-3-3 protein: variability of sporadic Creutzfeldt-Jakob
disease, laboratory standards, and quantitation. Arch Neurol 2003;60:803-4.
23 Coore J, Ambler J. Potential problem with standardization of prealbumin (transthyretin)
results in immunoturbidimetry. Clin Chem 1991;37:588-9.
24 Croci D, Nespolo A, Bosoni MA, Tarenghi G. A simple immunoturbidimetric method for
IgG and albumin quantitation in cerebrospinal fluid and serum. J Clin Chem Clin Biochem
1989;27:863-8.
25 Dinesen B, Nielsen S. A sensitive enzyme immunoassay for the quantitation of human
IgM. Scand J Clin Lab Invest 1979;39:449-54.
26 Hirohata S, Inoue T, Yamada A, Hirose S, Miyamoto T. Quantitation of IgG, IgA and IgM
in the cerebrospinal fluid by a solid-phase enzyme-immunoassay. Establishment of normal
control values. J Neurol Sci 1984;63:101-10.
27 Javors MA, Bowden CL, Maas JW. 3-methoxy-4-hydroxyphenylglycol, 5-
hydroxyindoleacetic acid, and homovanillic acid in human cerebrospinal fluid. Storage and
measurement by reversed-phase high-performance liquid chromatography and coulometric
detection using 3-methoxy-4-hydroxyphenyllactic acid as an internal standard. J
Chromatogr 1984;336:259-69.

Methods of Analysis © 2010 by Lawrence A. Kaplan and Amadeo J. Pesce.

28 Meucci G, Rossi G, Bettini R, Montanaro D, Gironelli L, Voci L, Bianchi F. Laser
nephelometric evaluation of albumin, IgG and alpha 2-macroglobulin: applications to the
study of alterations of the blood-brain barrier. J Neurol Sci 1993;118:73-8.
29 Mingioli ES, Strober W, Tourtellotte WW, Whitaker JN, McFarlin DE. Quantitation of
IgG, IgA and IgM in the CSF by radioimmunoassay. Neurology 1978;28:991-5.
30 Nerenberg ST, Prasad R. Radioimmunoassays for Ig classes G, A, M, D, and E in spinal
fluids: normal values of different age groups. J Lab Clin Med 1975;86:887-98.
31 Phillips TM, More NS, Queen WD, Holohan TV, Kramer NC, Thompson AM. High-
performance affinity chromatography: a rapid technique for the isolation and quantitation
of IgG from cerebral spinal fluid. J Chromatogr 1984;317:173-9.
32 Poulik MD, Perry DJ, Vokac E, Sekine T. Post-gamma globulin. II. Radioimmunoassay
determination of levels of post-gamma globulin and beta 2-microglobulin. Clin Chim Acta
1983;128:249-60.
33 Pressac M, Vignoli L, Aymard P. Immunoturbidimetric assay of transthyretin by random-
access analysis. Clin Chem 1988;34:176.
34 Schwerer B, Bach M, Bernheimer H. ELISA for determination of albumin in the nanogram
range: assay in cerebrospinal fluid and comparison with radial immunodiffusion. Clin
Chim Acta 1987;163:237-44.
35 Tourtellotte WW, Tavolato B, Parker JA, Comiso P. Cerebrospinal fluid
electroimmunodiffusion. An easy, rapid, sensitive, reliable, and valid method for the
simultaneous determination of immunoglobulin-G and albumin. Arch Neurol
1971;25:345-50.
36 Tourtellotte WW, Tavolato BA, Parker JA, Walker JE, Comiso PC.
Electroimmunodiffusion: a simple, rapid, reliable, and valid method for the simultaneous
determination of cerebrospinal fluid and serum immunoglobulin-G and albumin. Trans
Am Neurol Assoc 1970;95:323-5.
37 Vatassery GT, Krezowski AM, Sheridan MA. Comparison of manual methods of
determination of albumin in human cerebrospinal fluid by the bromcresol green and
immuno-precipitation methods. Clin Biochem 1980;13:78-80.
38 Mashige F, Shimizu T, Iijima S, Ohkubo A. Analysis for cerebrospinal fluid proteins by
sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Clin Chem 1992;38:2008-12.
39 Sorjonen DC, Cox NR, Swango LJ. Electrophoretic determination of albumin and gamma
globulin concentrations in the cerebrospinal fluid of dogs with encephalomyelitis
attributable to canine distemper virus infection: 13 cases (1980-1987). J Am Vet Med
Assoc 1989;195:977-80.
40 Bichsel P, Vandevelde M, Vandevelde E, Affolter U, Pfister H. Immunoelectrophoretic
determination of albumin and IgG in serum and cerebrospinal fluid in dogs with
neurological diseases. Res Vet Sci 1984;37:101-7.
41 Soldin SJ, Brugnara C, Wong EC, eds. Pediatric Reference Intervals. 7th ed. Washington
DC: AACC Press; 2007.

Methods of Analysis © 2010 by Lawrence A. Kaplan and Amadeo J. Pesce.

Tables and Figures

Table 1: Available Reference Intervals for CSF Proteins

Analyte Method Range (mg/dL)

α 1 -Acid glycoprotein Radial immunodiffusion (RID) 0.28-0.54

Albumin [4,18,24,28,34] IA techniques, turbidimetric, 0-45 (3 mo – 4 y)

Biuret/colorimetric 10-30 (>4 y)
α 2 -HS-glycoprotein Enzyme immunodiffusion (EID) 0.17

IgA [30] Radioimmunoassay (RIA) (in mg/dL by age):

15-20 y: 0.07 ± 0.04
21-40 y: 0.07 ± 0.03
41-60 y: 0.10 ± 0.03
61-87 y: 0.11 ± 0.06
IgD RIA (in U/mL by age):
15-20 y: 3.56 ± 2.0
21-40 y: 3.02 ± 1.3
41-60 y: 2.96 ± 0.88
61-87 y: 3.20 ± 0.9
IgG RIA (in mg/dL by age):
15-20 y: 3.5 ± 2.0
21-40 y: 4.2 ± 1.4
41-60 y: 4.7 ± 1.0
61-87 y: 5.8 ± 1.6
IgM RIA (in mg/dL by age):
15-20 y: 0.020 ± 0.009
21-40 y: 0.016 ± 0.003
41-60 y: 0.017 ± 0.004
61-87 y: 0.017 ± 0.005
α 2 -Macro-globulin Laser nephelometry 0.11-0.45 mg/dL

β 2 -Microglobulin [12,13] Nephelometry 1.5 ±0.2 mg/L

β 2 -Transferrin [9,10] Electrophoresis, immunofixation CSF, positive

electrophoresis, Western blot
Myelin basic protein RIA, ELISA <2.5 ng/mL
[7,15,16]
Total protein [1,2,41] Turbidimetry, nephelometry, (in mg/dL)
Biuret/colorimetric Adult 18-32
Pediatric:
Premature: 15-130
Full-term at birth: 40-
120
<1 month: 20-80
>1 month: 15-40

Methods of Analysis © 2010 by Lawrence A. Kaplan and Amadeo J. Pesce.

Transthyretin/prealbumin Nephelometry, turbidimetry, ~2% of total protein
[23,33] densitometry concentration
NOTE: Reference intervals will vary by the method used and should be established in-house. In newborns, CSF
proteins are higher, owing to high permeability of the newborn’s BBB. Reference intervals for specific proteins
have been published and should be established or verified accordingly.

Procedure: Sulfosalicylic Acid Turbidimetric Assay

Principle: Protein is precipitated as a fine, white precipitate by the addition of sulfosalicylic acid.
The resulting turbidity is determined spectrophotometrically at 430 nm.
Reagents
1. Sulfosalicylic acid, 30 g/L (118 mmol/L). Dissolve 30 g of sulfosalicylic acid in 800 mL
of distilled water, and bring to 1 L. Store in a brown bottle. Stable for 1 year at room
temperature.
2. Control. Ortho Diagnostics CSF control (Ortho Diagnostics, Raritan, NJ).
Assay
Equipment: Spectrophotometer capable of reading at 430 nm.
1. Add 1.0 mL of 3% sulfosalicylic acid to three “test” tubes; one for an instrument blank, one
for the Ortho standard, and one for the patient.
2. Add 0.2 mL of Ortho Diagnostics CSF control to the standard tube (click here).
3. Add 0.2 mL of CSF to the patient’s tube. Add 0.2 mL of saline solution to blank tube.
4. If the spinal fluid is strongly colored, prepare a patient blank using 0.2 mL of CSF and 1.0
mL of saline.
5. Cover all test tubes with plastic cups or Parafilm, and gently invert several times.
6. Let stand 10 min at room temperature.
7. Set absorbance at zero on the spectrophotometer at 430 nm, using the instrument blank of
sulfosalicylic acid plus saline.
8. Invert test tube gently to mix, introduce sample into cuvette, and read absorbance (A) of all
standards and patient samples.
9. Dilute samples 1:2 with saline, and rerun if concentration is greater than 2000 mg/L.
Calculations
1. Correct absorbance of samples with blank:

Apatient – Apatient blank = Acorrected patient

2. Read the corrected absorbance from the standard curve (see below) to obtain the protein
concentration.
Controls and Standards
1. Since sulfosalicylic acid produces different degrees of turbidity with equal concentrations of
different types of CSF proteins, it is necessary to construct a curve using a standard that has
the same albumin-globulin ratio as the spinal fluid being tested. Usually spinal fluid and
serum have a very similar albumin-globulin ratio, so a normal serum control with the proper
dilutions can be used to construct the curve. The albumin-globulin ratio of the standard
should be between 1.0 and 1.5.
2. To prepare the standard curve, dilute a control serum of 70 g/L of total protein with saline
so that 5 dilutions with protein concentration between 100 and 1500 mg/L are prepared as

Methods of Analysis © 2010 by Lawrence A. Kaplan and Amadeo J. Pesce.

Clinical References - Methods of Analysis 35-14

follows:

Stock Volume Total Volume (mL) Concentration (mg/L)

2.0 100 1400
1.5 100 1050
1.0 100 700
0.5 100 350
0.2 100 140
0 100 0

Do not use a serum control with elevated lipid or bilirubin content.

3. Standard curve. Construct the standard curve, plotting absorbance at 430 nm versus
protein concentration. This should be a straight line that passes through the origin.
4. The standard curve needs to be run only when there is a change in the spectrophotometer
(i.e., bulb change) or that lot of quality control material or the quality control sample does
not give the expected value.
Notes
1. The first drop of spinal fluid should be added slowly to the sulfosalicylic acid reagent. If a
dense cloud forms immediately, the protein concentration is very high, and dilution will be
necessary. In this case, the rest of the 0.2-mL sample should not be added to the acid but
instead returned to the original tube. The spinal fluid should be diluted with saline.
2. If red blood cells are present in the spinal fluid, the fluid must be centrifuged before protein
analysis.

Methods of Analysis © 2010 by Lawrence A. Kaplan and Amadeo J. Pesce.