Extracellular water-to-total body water ratio predicts phenoconversion in REM sleep behavior disorder: A bioimpedance-based cohort study

Background REM sleep behavior disorder (RBD) is a prodromal manifestation of α-synucleinopathies, with most patients converting to Parkinson's disease (PD), dementia with Lewy bodies (DLB), or multiple system atrophy (MSA) within 10–15 years. Practical, non-invasive biomarkers for predicting phenoconversion remain limited. Objective To evaluate whether bioelectrical impedance analysis (BIA) parameters, specifically extracellular water-to-total body water ratio (ECW/TBW) and phase angle (PhA), are associated with phenoconversion risk and motor symptom severity in patients with RBD. Methods We retrospectively analyzed 147 patients with polysomnography-confirmed RBD (2016–2024, Samsung Medical Center). Baseline BIA assessed body composition, including ECW/TBW and PhA. Motor symptoms were systematically evaluated at a 24 ± 3-month landmark visit using the Unified Parkinson's Disease Rating Scale (UPDRS). Phenoconversion to PD, DLB, or MSA was confirmed by two neurologists. Cox regression examined predictors of phenoconversion, and linear regression evaluated associations between BIA indices and UPDRS outcomes, adjusting for age, sex, and MoCA. Results During a mean follow-up of 4.46 ± 3.15 years, 31 patients (21.1 %) converted to an α-synucleinopathy. Converters showed higher ECW/TBW (38.9 ± 0.5 % vs. 38.6 ± 0.7 %, p = 0.007) and lower PhA (5.0 ± 0.6 vs. 5.3 ± 0.7, p = 0.037). Elevated ECW/TBW independently predicted phenoconversion (HR 6.56, 95 % CI 1.10–39.30, p = 0.039; cutoff 38.4 %, AUC = 0.69). At the landmark visit, higher ECW/TBW (B = 8.72, p = 0.034) and lower PhA (B = −9.97, p = 0.017) were independently associated with greater UPDRS-III motor burden. Conclusions BIA-derived ECW/TBW and PhA provide complementary insights into prodromal neurodegeneration. ECW/TBW predicted phenoconversion risk, while both indices reflected motor severity. These non-invasive, scalable measures may be useful for risk stratification and disease staging in RBD, warranting prospective validation.