Development of novel clinical examination scales for the measurement of disease severity in Creutzfeldt-Jakob disease

Abstract

AbstractObjectiveSporadic Creutzfeldt-Jakob disease (sCJD) causes rapidly-progressive dementia and complex abnormalities of motor systems with striking phenotypic heterogeneity, but no tools are available for the clinician to determine disease severity from bedside cognitive and neurological assessments. We used a robust statistical methodology and routinely-collected examination data to develop and validate short clinical rating scales quantifying longitudinal motor and cognitive dysfunction in sCJD.MethodsWe undertook a retrospective analysis of clinical examination data from the prospective National Prion Monitoring Cohort study, October 2008 – December 2016. Rasch analysis was used to iteratively construct interval scales measuring composite cognitive and motor dysfunction from pooled bedside neurological and cognitive examination tests.A longitudinal clinical examination dataset was constructed from a total of 528 patients with sCJD, comprising 1030 Motor Scale and 757 Cognitive Scale scores, over 130 patient-years of study, and used to demonstrate scale utility.ResultsThe Rasch-derived Motor Scale consists of 8 items, including examination items reliant on pyramidal, extrapyramidal and cerebellar systems. The Cognitive Scale comprises 6 items, and includes measures of executive function, language, visual perception and memory. Both scales are unidimensional, perform consistently regardless of age or gender and have excellent inter-rater reliability. Each scale can be completed in a few minutes at the bedside, as part of a normal neurocognitive examination. Several uses of the scales, in measuring longitudinal change, prognosis, and phenotypic heterogeneity are illustrated.InterpretationThese two novel scales measuring motor and cognitive dysfunction in sCJD should prove useful to objectively measure phenotypic and clinical change in future clinical trials and for patient stratification. This statistical approach can help to overcome obstacles to assessing clinical change in rapidly-progressive, multisystem conditions with limited longitudinal follow-up.