Autoimmune limbic encephalitis

Editor–We read with interest the recent article by Derry et al (Clin Med October 2011 pp 479–82) concerning limbic encephalitis (LE). There are important additional issues leading on from this succinct article that may be useful to the general physician when considering a LE diagnosis.

As reflected in the article, the diagnosis of autoimmune encephalitis (AE) and, more specifically, LE is a difficult one with no causative agent identified in up to two-thirds of encephalitic patients referred to specialist centres.¹

As mentioned in the article, recent work has indicated potassium channels are not the antigenic target in LE and it is in fact Anti-LGI1 (Leucine-rich glioma inactivate 1).^2,3 In addition, other studies suggest that another auto-antigen ‘CASPR2 - contactin-associated protein-like 2’ (expressed in hippocampal neurons) is associated with illnesses previously attributed to anti-VGKC antibodies such as drug-refractory epilepsy, encephalitis, peripheral nerve dysfunction, or a combination of both: Morvan syndrome or neuromyotonia.^4–7 Therefore, the term ‘anti-VGKC’ encephalitis is no longer in routine use.

The authors rightly illustrate the cardinal symptoms of LE as being severe short-term memory impairment with psychiatric symptoms such as personality change, depression, anxiety, hallucinations, confusion, and complex partial (often temporal) and generalised seizures.^8,9 It must be recalled that auto-antibodies may take some weeks to process and often LE treatment will have to be initiated prior to definitive diagnosis. With this in mind, there are other clinical features that may assist in pointing the clinician towards the diagnosis of LE. Features classically associated with anti-LGI1 encephalitis include faciobrachial tonic seizures⁷ and, in 40% of patients, myoclonus.² With regards to initial blood tests, hyponatraemia is a common finding in patients with anti-LGI1 encephalitis being found in patients both with and without underlying malignancy, indicating that it is not purely a paraneoplastic phenomenon.¹⁰ Autonomic instability is also described.

Once LE has been identified, it is essential to exclude paraneoplastic aetiology such as anti-Hu or anti-Ma2 associated LE as tumour treatment and, if possible, removal is associated with better outcomes than immunomodulatory therapy alone.³ It is important to note that the encephalitic presentation will precede the identification of the neoplasm in up to three quarters of patients.^9,11,12

Thankfully, as the article points out, non-paraneoplastic LE is associated with good outcomes if recognised early and treated aggressively. However, the clinician must be alert to the fact that residual neurological deficits may be subtle, difficult to elicit and, in some cases, sufficient to preclude the patient's return to work.

It is clear that the earlier AE or LE is diagnosed and therapy started, the better the outcome. We recently published an AE review containing a clinical algorithm for work-up, diagnosis, and treatment of autoimmune encephalitis¹³ that–in conjunction with important articles such as Derry et al's–we hope will assist the neurologist and generalist alike in early AE diagnosis and treatment.

• © 2012 Royal College of Physicians