Objective: Compare the association of classic prodromal features with subsequent development of Parkinson’s Disease (PD) versus Dementia with Lewy Bodies (DLB).

Background: DLB shares pathology with PD, but symptomatology and progression can vary dramatically. Ascertainment of DLB and key prodromal disorders like REM-behavior sleep disorder (RBD) are fraught with low positive predictive value. Large-scale comparative studies with accurate DLB and PD case ascertainment are limited.

Method: A retrospective case-control study was performed using electronic health record data from the VA under IRB approval (MIRB 04744). Ascertainment of DLB was iteratively performed to achieve a positive predictive value (PPV) of 79%. 10:1 control subjects were matched by sex, ethnicity, race, smoking status, and birth year.  PD subjects were identified using a previously validated approach (PPV = 78%), a subset matched to DLB by the same covariates, and then 10:1 controls selected. Prodromal exposures were defined by ICD-9 and -10 codes except for RBD which uses analysis of free-text (PPV = 86%). Association was calculated using conditional logistic regression adjusted for race, sex, smoking status, ethnicity, and birth year.

Results: 3,149 patients fulfilled the DLB case definition, 33,846 controls were matched, and 16,006 patients with PD were selected. Veterans were mostly male and aged > 60 years. Prodromal disorders at year -5 with stronger association with DLB versus PD included RBD (OR 8.5 [6.6,11.0] vs 4.3 [3.5,5.2]), orthostatic hypotension (OR 2.1 [1.7,2.6] vs. 1.7 [1.5,2.0]), and mild increases for constipation and chronic pain. Increased association with PD versus DLB was seen for smell/taste loss (OR 3.2 [2.4,4.2] vs. 1.8 [1.1,3.0]) and mild increases for erectile dysfunction. No differences were seen for hypersomnia, urinary dysfunction, joint pain, or anxiety. As expected, mild cognitive impairment (OR 12.9 [9.4,17.8] vs 5.1 [3.5,7.4]), cognitive decline (OR 11.3 [8.7,14.8] vs. OR 4.7 [3.7,6.0], and visual hallucinations (OR 12.4 [6.2, 24.7] vs 0.8 [0.3,1.9]) were enriched in DLB versus PD.

Conclusion: These data expand our knowledge about differences between prodromal DLB and PD using a large cohort with more accurate case definitions and longer data durations. Future work is needed to identify the most salient limited subset of clinical data that can discriminate DLB from PD up to two decades before diagnosis.

« Back to 2025 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/comparison-of-prodromal-features-between-parkinsons-disease-and-dementia-with-lewy-bodies-over-15-years/