Search tips
Search criteria 


Logo of jnnpsycJournal of Neurology, Neurosurgery and PsychiatryVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
J Neurol Neurosurg Psychiatry. 2007 April; 78(4): 332–333.
PMCID: PMC2077794

Limbic encephalitis: extension of the diagnostic armamentarium

Short abstract

Diagnostic spectrum of limbic encephalitis

The highly specific tests for anti‐Hu, anti‐Ma, anti‐amphiphysin and anti‐CV2/CRMP5 antibodies have been an important tool in the diagnosis of paraneoplastic limbic encephalitis (PLE).1 The testing procedure for these antibodies is two tiered. Firstly, patients' sera are incubated with brain slices from a paraformaldehyde‐perfused animal (usually rat or monkey). Specific binding patterns to brain cells allow the identification of the above‐named “well‐characterised” antibodies. After this immunohistochemical test, a confirmatory blotting test is performed on the brain homogenate (in comparison with a positive control) or on the suspected antigen as recombinant protein. For both techniques, testing of serum is more sensitive than testing of cerebrospinal fluid (CSF). The antibodies detected in this way react with intracellular antigens. They portend a poor outcome. A relevant limitation, however, is that 40% of patients with PLE do not harbour any of these antibodies.1

The diagnostic spectrum of limbic encephalitis was considerably broadened by the detection of antibodies against voltage‐gated potassium channels in association with (usually) non‐paraneoplastic limbic encephalitis (NPLE). These antibodies are diagnosed in serum by radioimmunoprecipitation. On standard immunohistochemical analysis, strongly positive sera stain the molecular layer of rat cerebellum. The prognosis of affected patients is good with consequent immunotherapy.2

In the paper by Bataller et al(see p 381),3 the Philadelphia group summarises its most recent experience on autoantibody testing in cases with the clinical syndrome of limbic encephalitis. They report on several patients with PLE found to harbour in their sera and CSF novel antibodies reacting with rat brain neuropil in the molecular layers of the hippocampus and cerebellum on immunohistochemical analysis. Importantly, this reactivity could be picked up only if slices were prepared from a brain that had not been perfused but instead had been immersed in paraformaldehyde for several days. The staining pattern was more easily detected with CSF than with serum. The antigens recognised by these antibodies have not yet been determined, and no consistent banding pattern on western blotting has been identified. Most patients had ovarian teratomata or thymic tumours. They had a mostly favourable outcome after tumour treatment and immunotherapy.

These data on patients harbouring antibodies against cellular membrane structures (ie, voltage‐gated potassium channel antibodies and the newly discovered “neuropil antibodies”) introduce a new borderline into the subclassification of patients with limbic encephalitis. In addition to PLE and NPLE, cases of limbic encephalitis can now be subdivided into poorly and favourably treatable cases. Clinicians will increasingly ask for this novel testing procedure. Future experience alone will show how well it will be to transfer this method as routine practice in other laboratories, especially as no confirmatory test is yet available. Replication studies on the neuropil antibodies will therefore be very welcome.

The neurologist will still make the tentative diagnosis of limbic encephalitis if an adult patient newly develops limbic symptoms and shows an otherwise unexplained mediotemporal magnetic resonance imaging–T2/fluid‐attenuated inversion recovery hyperintensity4 or fluorine‐18 fluorodeoxyglucose positron emission tomography hypermetabolism. In this situation, a tumour search considering the risk factors of the individual patient is mandatory. Ideally, comprehensive antibody testing is performed in a specialised laboratory. If a tumour is found, it should be treated immediately. Thereafter (and in all tumour‐negative cases immediately), immunotherapy should be introduced. In antibody‐positive cases, additional investigation for tumours often associated with this particular antibody may become necessary, and the treatment prognosis can be better pre‐estimated.

Hopefully, diagnostic tools, prognostic data and treatment options will become available for the still insufficiently characterised group of (so far) “antibody negative” cases with NPLE5 (which also stands out in Bataller's paper). Future studies may add to the diagnostic armamentarium for these cases, as has been achieved by the present work for the PLE subgroup.


Competing interests: None declared.


1. Gultekin S H, Rosenfeld M R, Voltz R. et al Paraneoplastic limbic encephalitis: neurological symptoms, immunological findings and tumour association in 50 patients. Brain 2000. 1231481–1494.1494 [PubMed]
2. Vincent A, Buckley C, Schott J M. et al Potassium channel antibody‐associated encephalopathy: a potentially immunotherapy‐responsive form of limbic encephalitis. Brain 2004. 127701–712.712 [PubMed]
3. Bataller L, Kleopa K A, Wu G F. et al Autoimmune limbic encephalitis in 39 patients: immunophenotypes and outcomes. J Neurol Neurosurg Psychiatry 2007. 78381–385.385 [PMC free article] [PubMed]
4. Urbach H, Soeder B M, Jeub M. et al Serial MRI of limbic encephalitis. Neuroradiology 2006. 48380–386.386 [PubMed]
5. Bien C G, Schulze‐Bonhage A, Deckert M. et al Limbic encephalitis not associated with neoplasm as a cause of temporal lobe epilepsy. Neurology 2000. 551823–1828.1828 [PubMed]

Articles from Journal of Neurology, Neurosurgery, and Psychiatry are provided here courtesy of BMJ Publishing Group