Accuracy of clinical diagnosis of progressive supranuclear palsy

Mov Disord. 2004 Feb;19(2):181-9. doi: 10.1002/mds.10680.


We assessed the accuracy of clinical diagnosis of progressive supranuclear palsy (PSP, Steele-Richardson-Olszewski disease) and the validity of existing sets of clinical diagnostic criteria for PSP (see Appendix) using neuropathologically examined cases from the Queen Square Brain Bank for Neurological Disorders. Diagnosis of PSP was made by 40 different physicians, and 60 cases clinically diagnosed as PSP when last assessed in life were studied. In 47 cases (78%), the diagnosis of PSP was confirmed pathologically. False-positive diagnoses included Parkinson's disease with significant additional cortical Lewy body (n = 3) or Alzheimer (n = 1) pathology, multiple system atrophy (n = 4), and corticobasal degeneration, Pick's disease, motor neurone disease, cerebrovascular disease, and a sporadic case of frontotemporal dementia and parkinsonism linked to chromosome 17 (1 case each). Most cases of PSP were diagnosed accurately by neurologists at the final assessment. Although application of National Institute of Neurological Disorders and the Society for PSP possible category marginally improved the accuracy of initial clinical diagnosis, none of the existing operational criteria could significantly improve accuracy of the final clinical diagnosis.

MeSH terms

  • Adult
  • Aged
  • Aged, 80 and over
  • Brain / pathology
  • Diagnosis, Differential
  • Female
  • Humans
  • Male
  • Middle Aged
  • Neurologic Examination / statistics & numerical data*
  • Reproducibility of Results
  • Retrospective Studies
  • Supranuclear Palsy, Progressive / diagnosis*
  • Supranuclear Palsy, Progressive / pathology