Lewy body dementia

Dementia with Lewy bodies (DLB) is the preferred term for entities previously called the Lewy body variant of AD, Lewy body dementia, senile dementia of Lewy body type and cortical Lewy body disease.

DLB and Parkinson’s disease dementia (PDD) are two ends of one Lewy body disease spectrum. The “one-year rule” applies: PDD describes dementia developing in established Parkinson’s disease, while DLB is diagnosed when dementia precedes or coincides within one year of motor symptom onset. By the end stage the two are often indistinguishable clinically and neuropathologically.

DLB is the second commonest cause of neurodegenerative dementia in older people after AD, accounting for ~15–20% of autopsy series and 10–15% of dementia cases in population studies. Incidence is 32–112 per 100,000 person-years in those over 65.

Onset is between 50 and 90 years, and illness duration varies between 6 and 10 years.

The defining lesion is the α-synuclein-containing Lewy body and Lewy neurite. Recent evidence indicates that presynaptic α-synuclein aggregates, rather than the Lewy body itself, are synapto-toxic and cause neurodegeneration.

Three patterns of Lewy body distribution: (i) brainstem-predominant — the PDD phenotype; (ii) cerebral cortex-predominant — the DLB phenotype; (iii) transitional, intermediate between the two. The likelihood of presenting as typical DLB is directly related to the distribution of Lewy body pathology and inversely related to the severity of AD pathology.

Concomitant AD pathology is present in 90% of Lewy body disease cases. Patients with marked concurrent neurofibrillary tangle pathology present with an insidious amnestic syndrome more suggestive of AD, with fewer cognitive fluctuations, fewer neuropsychiatric symptoms and less parkinsonism.

Most cases are sporadic, but rare autosomal dominant inheritance occurs, with SNCA and LRRK2 mutations, which may manifest as any Lewy body phenotype — supporting a single underlying genetic-pathological entity.

Fluctuating cognition with pronounced variations in attention and alertness — waxing and waning behavioural inconsistency, incoherent speech, variable attention, staring or “zoning out”. The marked amplitude between best and worst performance distinguishes it from minor day-to-day variation. Endorsement of at least three of four composite features — daytime drowsiness and lethargy, daytime sleep of ≥2 hours, staring into space for long periods, episodes of disorganised speech — gives a positive predictive value of 83% for DLB over AD. Fluctuations best predict DLB when present early.

Recurrent vivid visual hallucinations occur in up to 80% — typically well-formed, detailed, three-dimensional mute images of people, children or animals. Caregivers may under-report them, so patient self-report is useful. Their presence marks cortical Lewy body pathology and greater cortical cholinergic deficit, and predicts good response to cholinergic therapy. Misidentification delusions are the commonest delusional type, followed by persecutory/paranoid, phantom boarder and abandonment delusions.

REM sleep behaviour disorder (RBD) — a parasomnia with loss of REM atonia, so patients act out vivid, frightening dreams: talking or shouting in sleep, striking the bed partner, falling out of bed. History must come from the bed partner. It precedes motor, cognitive and neuropsychiatric manifestations by years or decades and curiously improves as disease progresses. It is highly specific for synucleinopathy (DLB, PD, MSA) and rare in AD and primary tauopathies.

Spontaneous parkinsonism occurs in about 75%, is less pronounced and less levodopa-responsive than idiopathic PD, so only one cardinal feature (bradykinesia, rigidity or resting tremor) is required for diagnosis.

Supportive clinical features: severe sensitivity to antipsychotics, postural instability, repeated falls, syncope or transient unresponsiveness, severe autonomic dysfunction (orthostatic hypotension, urinary incontinence, constipation), hypersomnia, systematised delusions, hallucinations in other modalities, and apathy, anxiety and depression.

Neuroleptic sensitivity: about 50% of DLB patients do not react adversely, so tolerance does not exclude the diagnosis, but a positive history of severe sensitivity is strongly suggestive. Reactions range from sedation, increased confusion and worsening parkinsonism to irreversible parkinsonism, impaired consciousness and marked autonomic disturbance.

Cognitive profile: a “dysexecutive” syndrome with prominent executive, attentional and visuospatial dysfunction, with visuospatial impairment prominent even early. Amnesia may not be prominent initially; memory impairment does not preclude DLB, and patients with pure Lewy body disease have similar free recall but better delayed recognition memory than AD, indicating a retrieval rather than storage deficit.

Essential: progressive cognitive decline with reduced social and occupational functioning. Probable DLB = ≥2 core clinical features, or 1 core feature plus ≥1 indicative biomarker. Possible DLB = 1 core feature with no indicative biomarker, or ≥1 indicative biomarker with no core feature. DLB is less likely if there is a history of stroke, another illness or brain disorder sufficient to explain the picture, or if parkinsonism is the only core feature and appears first at a severe stage of dementia.

Indicative biomarkers: low dopamine transporter uptake in basal ganglia on SPECT/PET; abnormal (low uptake) 123I-MIBG myocardial scintigraphy, which quantifies postganglionic sympathetic cardiac innervation and discriminates DLB from AD especially in milder cases (MMSE >21); and polysomnographic confirmation of REM sleep without atonia. Supportive biomarkers: relative preservation of medial temporal lobe structures on CT/MRI; generalised low uptake with reduced occipital activity ± the cingulate island sign on FDG-PET; and prominent posterior slow-wave EEG activity with periodic pre-alpha/theta fluctuations.

Earlier criteria had high specificity (80–100%) but limited sensitivity (35–80%); the third consensus criteria (2005) improved sensitivity, but detection remains suboptimal, with many cases missed or misdiagnosed as AD. The revised fourth consensus criteria clearly separate clinical features from diagnostic biomarkers, with increased weighting given to RBD and MIBG scintigraphy. CSF α-synuclein may be low in DLB versus AD, while Aβ, tau and p-tau help determine concomitant AD pathology; amyloid deposition has little value in differentiating DLB from AD, though greater amyloid load predicts faster decline in DLB.

Management addresses motor, cognitive, psychiatric and autonomic dysfunction, and the clinician must be mindful of the tension between improving one symptom at the expense of worsening another — for example aggravating hallucinations with levodopa. Conventional neuroleptics are best avoided; quetiapine or clozapine may be used judiciously. Cholinesterase inhibitors are effective and relatively safe for psychiatric and cognitive symptoms in DLB, with mainly gastrointestinal side effects, although DLB patients may be more susceptible to bradyarrhythmic effects because of autonomic dysfunction.

Memantine produces cognitive and global benefits in DLB but without significant improvement in psychiatric symptoms, and case reports describe worsening delusions and hallucinations with memantine in DLB — caution is warranted. Preliminary subgroup analysis suggests a more pronounced global response in PDD than DLB. Autonomic management uses fludrocortisone or midodrine for symptomatic postural hypotension; RBD is managed with a safe sleep environment, melatonin and, where necessary, clonazepam; and falls prevention and dysphagia surveillance are essential. Latest Updates (2023–Present) α-Synuclein seed amplification assays (SAA/RT-QuIC) have transformed diagnosis. In CSF they achieve >90% sensitivity and specificity for Lewy body pathology, and skin biopsy SAA (2024 JAMA) detected phosphorylated α-synuclein in around 95% of patients with synucleinopathies — moving the field toward a biological definition of Lewy body disease paralleling the biological definition of AD, and enabling detection at the prodromal RBD stage. Isolated RBD is now formally recognised as prodromal synucleinopathy, with around 75–90% phenoconverting to DLB, PD or MSA over 10–15 years; this has made RBD clinics the principal recruitment route for neuroprotection trials in Lewy body disease. Research criteria for prodromal DLB (McKeith et al.) define three syndromes — MCI with Lewy bodies, delirium-onset DLB and psychiatric-onset DLB — allowing diagnosis before dementia, with MCI-LB criteria now operationalised. Therapeutics remain symptomatic. Neflamapimod (p38α kinase inhibitor) showed cognitive benefit in the RewinD-LB Phase 2b trial (2025) in DLB patients without amyloid co-pathology and is progressing to Phase 3. Ambroxol (glucocerebrosidase chaperone) and α-synuclein immunotherapies are in trials. No disease-modifying therapy is yet licensed. Antipsychotic caution has been reinforced — pimavanserin remains licensed only for Parkinson’s disease psychosis; the HARMONY trial showed relapse prevention in dementia-related psychosis but did not yield a DLB indication, and all antipsychotics carry the boxed warning for increased mortality in elderly patients with dementia-related psychosis. Practical Indian points: DLB is systematically under-diagnosed and frequently mislabelled as AD or late-onset psychosis, exposing patients to conventional antipsychotics such as haloperidol with potentially catastrophic neuroleptic sensitivity reactions. DAT-SPECT and MIBG scintigraphy are available only in selected tertiary centres, so clinical recognition of the four core features — especially a careful bed-partner history of RBD — remains the single most important diagnostic step in Indian practice.