After NMDA receptor encephalitis, the psychosis becomes an affective illness - and the memory problems stay

Design

Cross-sectional study with 1:1 propensity-matched healthy controls and normative modelling of structural MRI

Population

70 patients with post-acute NMDA receptor encephalitis (86 per cent female, mean age 28.1 years) and 70 matched controls, Germany

Primary outcome

Morphological brain complexity and its relation to persistent post-acute symptoms

Effect

67 per cent had residual symptoms (memory 61 per cent, psychiatric 36 per cent); greater morphological alteration with persistent psychiatric (t -2.65, P = 0.010) and memory symptoms (t -3.98, P = 0.0002)

Seventy patients referred to Charite in Berlin with post-acute NMDA receptor encephalitis were matched one to one with healthy controls for age and sex, and studied a median 22.2 months after onset (IQR 8.5 to 42.5). They were young - mean age 28.1 years - and 86 per cent were women.

The recovery on paper is dramatic. Median modified Rankin Scale fell from 5 to 1 and the Clinical Assessment Scale for Autoimmune Encephalitis from 11 to 1 (both P < 0.0001). But 47 patients, 67 per cent, still had residual symptoms: memory dysfunction in 61 per cent and psychiatric symptoms in 36 per cent. The character of the psychiatric illness had changed - psychosis-dominant at peak, affective-dominant afterwards. Structural MRI showed reduced morphological complexity in both hippocampi and across a fronto-cingulo-temporal cluster in grey and white matter, and normative modelling found the alterations were greater in patients who still had symptoms, whether psychiatric (t -2.65, P = 0.010), memory (t -3.98, P = 0.0002) or both (t -5.46, P < 0.0001).

The imaging technique is research-stage and nobody should be ordering fractal dimensionality analyses. The clinical message is not. A patient discharged from neurology with a modified Rankin score of 1 has, on these figures, a two-in-three chance of continuing symptoms, and those symptoms will land in a psychiatric clinic looking like a primary affective disorder in a young adult. Knowing the history changes the formulation and the follow-up: this needs structured post-acute care for cognition and mood, not a fresh diagnosis of depression.

• Ask about a history of encephalitis in any young adult presenting with new affective illness and memory complaints.
• Expect the phenotype to have shifted; the absence of psychosis does not mean recovery is complete.
• Test memory formally rather than relying on report; memory dysfunction was the commonest residual problem.
• Arrange structured post-acute follow-up jointly with neurology rather than accepting discharge at a good Rankin score.
• Do not order morphological complexity imaging; it is a research marker with no clinical threshold.

The statistics, in plain English

This is a cross-sectional study with matched controls, so the brain changes and the symptoms were measured at the same moment - the imaging cannot be said to predict outcome, only to accompany it. The comparison of peak-illness and post-acute scales is within the same patients over time and is unambiguous. Normative modelling compares each patient against an expected distribution rather than a group mean, which is more sensitive but also more dependent on the reference sample; with 70 patients, the subgroup t values should be read as consistent direction rather than precise magnitude.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.