Skip to content
PubMed This is a summary of 19 peer-reviewed journal articles Updated
Neurology · Autoimmune Encephalitis

Is Blood or CSF Better for Autoimmune Encephalitis?

At a Glance

For autoimmune encephalitis, neither blood nor spinal fluid is always more accurate. When it is safe, testing both samples gives the clearest result because some antibodies are easier to detect in CSF, others in blood, and negative tests do not rule out disease.

The Bottom Line: A negative blood or spinal fluid antibody test does not by itself rule out autoimmune encephalitis, and a positive result does not automatically prove it. Doctors must combine test results with your symptoms, neurological exams, brain imaging (MRI), brainwave tests (EEG), other spinal fluid findings, and careful checks for infections [1][2].

Neither blood (serum) nor spinal fluid (CSF) testing is universally “better” for diagnosing autoimmune encephalitis. Instead, the medical standard—when safely possible—is paired testing, meaning both blood and spinal fluid are collected and tested at the same time [3][4].

The accuracy of testing depends heavily on the specific antibody being investigated, the laboratory methods used, and the patient’s symptoms [3]. Checking both fluids provides the most complete picture and helps doctors interpret confusing or borderline results [4].

Why Spinal Fluid Testing is Crucial

For certain types of autoimmune encephalitis, spinal fluid is generally more reliable. For example, in anti-NMDAR encephalitis (the most common type), spinal fluid testing is central to diagnosis. In one study, weak blood-only positive results for NMDAR antibodies were often misleading false positives [5]. For other targets like GABA-B antibodies, testing spinal fluid may also improve detection and help clarify uncertain blood results, making paired testing highly recommended [6].

Because the brain and spinal cord are protected by the blood-brain barrier (a filtering mechanism that keeps many substances in the blood out of the brain), finding antibodies directly in the spinal fluid can provide diagnostic support that the immune system is involved in the central nervous system [4]. However, antibodies can sometimes leak from the blood into the spinal fluid, so a positive result must still be evaluated alongside the full clinical picture.

(Note: The decision to perform a lumbar puncture to collect spinal fluid depends on your clinical situation, safety, and the need to exclude infection. It should be guided entirely by your treating medical team).

When Blood Tests Shine

For other subtypes, blood testing can be highly sensitive and very informative. Antibodies such as LGI1 and CASPR2 are often readily detected in blood serum [7][8]. Depending on the patient’s symptoms and the assay used, patients may test positive in their blood while their spinal fluid is negative [7][8].

The Danger of Low-Level, Blood-Only Results

Patients are sometimes told they have autoimmune encephalitis based entirely on a “low-titer” (low-level) blood test. A titer is a measurement of how much antibody is present. These low-level, serum-only results must be interpreted with caution, as they are a recognized diagnostic pitfall [2].

Many false-positive results happen when testing only the blood [9]. Some testing methods, like line blots (a laboratory technique where antibodies stick to strips of proteins), frequently pick up non-specific antibodies in patients with completely different, non-autoimmune conditions [10].

A weak blood result generally provides less diagnostic support if the patient’s clinical syndrome doesn’t match the antibody, or if the MRI, EEG, and spinal fluid are entirely normal [11][12]. However, it is important to know that having a normal MRI or normal spinal fluid does not completely rule out autoimmune encephalitis [13][14].

Another classic example is GAD65. Low levels of GAD65 antibodies can be found in 8% to 13% of perfectly healthy people, depending on the testing method and cutoff used [15]. For a GAD65 result to strongly support a neurological diagnosis, doctors look for very high blood levels and corroboration in the spinal fluid [15].

Antibody Testing Tendencies

(Note: These are general tendencies, not strict rules. Accuracy can vary by laboratory and testing method).

Antibody Often More Reliable In Important Context
NMDAR Spinal Fluid (CSF) Weak blood-only positive results carry a high risk of being false positives [5].
GABA-B Spinal Fluid (CSF) Spinal fluid testing helps clarify uncertain blood results [6].
LGI1 Blood (Serum) Detection depends heavily on the type of test used; blood is often highly sensitive [7].
CASPR2 Blood (Serum) Can be positive in blood while negative in spinal fluid, depending on the specific symptoms [8].
GAD65 Both (High Titers) Low blood levels are common in healthy people; high levels or spinal fluid positivity are more clinically relevant [15].

What if No Antibodies are Found? (Seronegative Disease)

When the standard antibody panels are negative, it does not automatically mean you don’t have autoimmune encephalitis. A negative panel just means the currently recognized or tested antibodies were not detected [16].

Seronegative autoimmune encephalitis is a clinical diagnosis. It is made when a patient has a highly suspicious progression of symptoms, supportive findings on MRI or EEG, and when doctors have rigorously excluded other causes like infections, tumors, or metabolic issues [1][2]. Conflicting or negative test results are frustrating, but “seronegative” does not mean your symptoms are imaginary.

Other Clues in the Spinal Fluid

If classic antibodies aren’t found, doctors look for other markers of inflammation in the spinal fluid. These include:

  • White blood cell count (pleocytosis): An elevated number of immune cells [17].
  • Protein levels: High protein can suggest inflammation or a disruption of the blood-brain barrier [17].
  • Oligoclonal bands (OCBs): When paired with blood testing, this indicates that immunoglobulins (antibodies) are being actively produced within the central nervous system. However, they do not prove the antibodies are specifically targeting brain tissue [18].
  • IgG index: An estimate of antibody production in the spinal fluid relative to the blood [19].

While these markers support the presence of brain inflammation, they are nonspecific [17]. They can occur in infections (like viral encephalitis) or other neurological diseases like multiple sclerosis [18]. Furthermore, perfectly normal spinal fluid does not rule out autoimmune encephalitis—up to 27% of patients in one study of early, active disease had completely normal cell counts and protein levels [14].

Common questions in this guide

Is blood or spinal fluid more accurate for autoimmune encephalitis antibody testing?
Neither sample is always more accurate. The best choice depends on the antibody, laboratory method, and clinical picture; when it is safe, doctors commonly test paired blood and CSF to compare results.
Why can spinal fluid be especially important for NMDAR antibodies?
For anti-NMDAR encephalitis, CSF testing is central because weak blood-only positives can be misleading. A result still needs to match symptoms and other findings rather than proving the diagnosis by itself.
Can autoimmune encephalitis antibodies be found in blood but not CSF?
Yes. LGI1 and CASPR2 antibodies are often detected in serum even when CSF is negative, depending on symptoms and the test used. Doctors interpret this pattern in context rather than treating it as automatically diagnostic.
Does a negative antibody panel rule out autoimmune encephalitis?
No. Seronegative autoimmune encephalitis can be diagnosed when symptoms and supportive MRI or EEG findings strongly suggest the condition after infections, tumors, metabolic problems, and other causes have been excluded. Normal MRI or CSF results also do not completely rule it out.
What does a low-titer positive blood test mean?
A low-titer result means that a small amount of antibody was detected, but it may not be specific to autoimmune encephalitis. Blood-only low-level results, especially from less specific tests, can be false positives and should be checked against the clinical syndrome and, when appropriate, another testing method.
What can spinal fluid show if no autoimmune antibody is found?
Doctors may look for an elevated white blood cell count, increased protein, oligoclonal bands, or an abnormal IgG index as signs of inflammation. These findings are not specific and can occur with infections or other neurologic diseases, while normal results do not exclude autoimmune encephalitis.

Get notified when new evidence is published on Non-specific autoimmune supratentorial encephalitis without characteristic antibodies.

We monitor PubMed for new peer-reviewed studies on this topic and email a short summary when something meaningful changes.

References

References (19)
  1. 1

    Diagnostic criteria for autoimmune encephalitis: utility and pitfalls for antibody-negative disease.

    Dalmau J, Graus F

    The Lancet. Neurology 2023; (22(6)):529-540 doi:10.1016/S1474-4422(23)00083-2.

    PMID: 37210100
  2. 2

    Mimics of Autoimmune Encephalitis: Validation of the 2016 Clinical Autoimmune Encephalitis Criteria.

    Van Steenhoven RW, de Vries JM, Bruijstens AL, et al.

    Neurology(R) neuroimmunology & neuroinflammation 2023; (10(6)) doi:10.1212/NXI.0000000000200148.

    PMID: 37582614
  3. 3

    Neural Antibody Testing in Patients with Suspected Autoimmune Encephalitis.

    Budhram A, Dubey D, Sechi E, et al.

    Clinical chemistry 2020; (66(12)):1496-1509 doi:10.1093/clinchem/hvaa254.

    PMID: 33221892
  4. 4

    Assessing the role of a tissue-based assay in the diagnostic algorithm of autoimmune encephalitis.

    Masi G, Spagni G, Campetella L, et al.

    Journal of neuroimmunology 2021; (356()):577601 doi:10.1016/j.jneuroim.2021.577601.

    PMID: 33975245
  5. 5

    Autoimmune Encephalitis Misdiagnosis in Adults.

    Flanagan EP, Geschwind MD, Lopez-Chiriboga AS, et al.

    JAMA neurology 2023; (80(1)):30-39 doi:10.1001/jamaneurol.2022.4251.

    PMID: 36441519
  6. 6

    Limitations of a Commercial Assay as Diagnostic Test of Autoimmune Encephalitis.

    Ruiz-García R, Muñoz-Sánchez G, Naranjo L, et al.

    Frontiers in immunology 2021; (12()):691536 doi:10.3389/fimmu.2021.691536.

    PMID: 34267758
  7. 7

    The diagnosis of anti-LGI1 encephalitis varies with the type of immunodetection assay and sample examined.

    Muñoz-Sánchez G, Planagumà J, Naranjo L, et al.

    Frontiers in immunology 2022; (13()):1069368 doi:10.3389/fimmu.2022.1069368.

    PMID: 36591253
  8. 8

    Characterization of a Subtype of Autoimmune Encephalitis With Anti-Contactin-Associated Protein-like 2 Antibodies in the Cerebrospinal Fluid, Prominent Limbic Symptoms, and Seizures.

    Joubert B, Saint-Martin M, Noraz N, et al.

    JAMA neurology 2016; (73(9)):1115-24 doi:10.1001/jamaneurol.2016.1585.

    PMID: 27428927
  9. 9

    Indications and Diagnostic Yield of Paraneoplastic and Autoimmune Encephalitis Antibody Testing: A Retrospective Cohort Study.

    Ford H, Seery N, Tan T, et al.

    Brain and behavior 2025; (15(8)):e70779 doi:10.1002/brb3.70779.

    PMID: 40847447
  10. 10

    Seroprevalence of neuronal antibodies in diseases mimicking autoimmune encephalitis.

    Vaisvilas M, Petrosian D, Bagdonaite L, et al.

    Scientific reports 2024; (14(1)):5352 doi:10.1038/s41598-024-55995-6.

    PMID: 38438516
  11. 11

    Neural Autoantibodies in Cerebrospinal Fluid and Serum in Clinical High Risk for Psychosis, First-Episode Psychosis, and Healthy Volunteers.

    Bien CG, Rohleder C, Mueller JK, et al.

    Frontiers in psychiatry 2021; (12()):654602 doi:10.3389/fpsyt.2021.654602.

    PMID: 33841216
  12. 12

    Absence of cerebrospinal fluid antineuronal antibodies in schizophrenia spectrum disorders.

    Oviedo-Salcedo T, de Witte L, Kümpfel T, et al.

    The British journal of psychiatry : the journal of mental science 2018; (212(5)):318-320 doi:10.1192/bjp.2018.24.

    PMID: 29587895
  13. 13

    The Laboratory Diagnosis of Autoimmune Encephalitis.

    Lee SK, Lee ST

    Journal of epilepsy research 2016; (6(2)):45-50 doi:10.14581/jer.16010.

    PMID: 28101474
  14. 14

    Searching for autoimmune encephalitis: Beware of normal CSF.

    Hébert J, Gros P, Lapointe S, et al.

    Journal of neuroimmunology 2020; (345()):577285 doi:10.1016/j.jneuroim.2020.577285.

    PMID: 32563126
  15. 15

    Positive Predictive Value of Anti-GAD65 ELISA Cut-Offs for Neurological Autoimmunity.

    Budhram A, Freeman E, Bhayana V, Yang L

    The Canadian journal of neurological sciences. Le journal canadien des sciences neurologiques 2023; (50(5)):766-768 doi:10.1017/cjn.2022.276.

    PMID: 35861111
  16. 16

    Tissue-based assay-confirmed, antibody-negative autoimmune encephalitis responsive to low-dose rituximab in an elderly patient.

    Han L, Li C, Li L, et al.

    Journal of Alzheimer's disease reports 2025; (9()):25424823251383524 doi:10.1177/25424823251383524.

    PMID: 41018715
  17. 17

    Novel Surrogate Markers of CNS Inflammation in CSF in the Diagnosis of Autoimmune Encephalitis.

    Jiang JX, Fewings N, Dervish S, et al.

    Frontiers in neurology 2019; (10()):1390 doi:10.3389/fneur.2019.01390.

    PMID: 32116981
  18. 18

    CSF oligoclonal bands in neurological diseases besides CNS inflammatory demyelinating disorders.

    Ren J, Wang J, Liu R, et al.

    BMJ neurology open 2025; (7(1)):e000912 doi:10.1136/bmjno-2024-000912.

    PMID: 40051552
  19. 19

    Comparing clinical features, severity and prognosis of autoimmune encephalitis and with and without oligoclonal bands.

    Xue H, Guo X, Jiang Y, et al.

    Frontiers in neurology 2023; (14()):1281276 doi:10.3389/fneur.2023.1281276.

    PMID: 38249729

This page is for informational purposes only and does not constitute medical advice. A neurologist or treating team should interpret your antibody results and decide whether lumbar puncture or treatment is appropriate.