Longitudinal Psychiatric Symptoms in Prodromal Huntington's Disease: A Decade of Data

Abstract

Objective: Psychiatric symptoms are a significant aspect of Huntington's disease, an inherited neurodegenerative illness. The presentation of these symptoms is highly variable, and their course does not fully correlate with motor or cognitive disease progression. The authors sought to better understand the development and longitudinal course of psychiatric manifestations in individuals who carry the Huntington's disease mutation, starting from the prodromal period prior to motor diagnosis.

Method: Longitudinal measures for up to 10 years of psychiatric symptoms from the Symptom Checklist-90-Revised were obtained from 1,305 participants (1,007 carrying the Huntington's disease mutation and 298 without [classified as controls]) and 1,235 companions enrolled in the Neurobiological Predictors of Huntington's Disease (PREDICT-HD) study. Participants with the mutation were stratified into three groups according to probability of motor diagnosis within 5 years. Using linear mixed-effects regression models, differences in psychiatric symptoms at baseline and over time between the mutation-positive groups and the controls were compared, as well as between ratings by mutation-positive participants and their companions.

Results: Nineteen of 24 psychiatric measures (12 participant ratings and 12 companion ratings) were significantly higher at baseline and showed significant increases longitudinally in the individuals with the Huntington's disease mutation compared with controls. The differences were greatest in comparisons of symptom reports from companions compared with self-reports, especially in participants who were closest to motor diagnosis.

Conclusions: The results indicate that psychiatric manifestations develop more often than previously thought in the Huntington's disease prodrome. Symptoms also increase with progression of disease severity. Greater symptom ratings by companions than by mutation-positive participants suggest decreasing awareness in those affected.

Trial registration: ClinicalTrials.gov NCT00051324.

Figure 1

Fitted Linear Mixed-Effects Regression (LMER) curves by group for participant and companion Symptom Checklist-90-Revised (SCL-90-R) ratings. All model coefficients were estimated adjusting for gender, years of education, and age at entry. The plots show the SCL-90-R score as a function of duration, person (participant or companion) and group. Low, Med, High indicate low, medium, or high probability of diagnosis within five years as estimated by CAG-Age Product (CAP) score.