Conference Coverage

How Accurately Do Current Criteria Identify CBD and PSP?

A retrospective study suggests that the sensitivity and specificity of the criteria need improvement.


LOS ANGELES—Current criteria are comparatively insensitive and nonspecific for distinguishing between corticobasal degeneration (CBD) and progressive supranuclear palsy (PSP), according to research presented at the 70th Annual Meeting of the American Academy of Neurology. Including adjunctive biomarkers with the criteria might improve their sensitivity.

Jessica Weinstein, MD

It is reasonable to ask whether CBD and PSP should be considered one entity instead of two, said Jessica Weinstein, MD, clinical fellow at the University of California, San Francisco, School of Medicine. Such an approach could enrich clinical trials, given that the diseases are uncommon, she added.

Comparing Pathologic and Clinical Diagnoses

Although CBD and PSP have been considered distinct disorders, each is heterogeneous. The two disorders share symptoms such as limb rigidity, akinesia, postural instability, and behavioral changes. They also share symptoms with other clinical syndromes such as behavioral variant frontotemporal dementia. Furthermore, the Armstrong criteria for CBD include criteria for PSP, and the Höglinger criteria for PSP include criteria for CBD.

Dr. Weinstein and colleagues examined data for patients with autopsy-confirmed four-repeat tauopathies to evaluate the sensitivity and specificity of the Armstrong and Höglinger criteria for diagnosing CBD and PSP, respectively. Information was extracted from the Penn Integrated Neurodegenerative Disease Database. Neuropathologic diagnosis for participants in this database was determined using established criteria.

A researcher blinded to pathologic diagnosis coded each patient for the presence or absence of clinical features using data from his or her first clinical visit. The researcher assessed subjects for 34 features associated with CBD, PSP, behavioral-variant frontotemporal dementia, or primary progressive aphasia. Patients with absent or insufficient data were excluded.

Criteria May Need Refinement

The population included 107 autopsy subjects, 37 of whom had a pathologic diagnosis of CBD and 70 of whom had a pathologic diagnosis of PSP. The investigators found no significant differences between the groups in age at death, age at onset, or disease duration. The percentage of females was higher in the CBD group. The percentage of patients evaluated in movement clinics, rather than cognitive clinics, was 8% for patients with CBD and 40% for patients with PSP.

Almost all clinical features were more prevalent in the PSP group than the CBD group, except limb dystonia, myoclonus, and alien limb syndrome. Language impairment (ie, speech apraxia; agrammatism; and impaired naming, single-word comprehension, and grammatical comprehension) was more prevalent in the CBD group, but this difference was not statistically significant. The PSP group had a significantly higher prevalence of falls, being chair bound, postural instability, and vertical saccades. The PSP group had more bradyphrenia than the CBD group, and the CBD group had more executive impairment than the PSP group.

The Armstrong criteria identified probable CBD with a sensitivity of 11% and a specificity of 100%. The specificity result “should be taken with a grain of salt because only four patients met criteria,” said Dr. Weinstein. Armstrong criteria identified possible CBD with 35% sensitivity and 34% specificity.

The Höglinger criteria identified probable PSP with 66% sensitivity and 70% specificity. They identified possible PSP with 63% sensitivity and 65% specificity. A post hoc analysis suggested that including grammar comprehension in the Höglinger criteria improved their sensitivity.

The study’s limitations include its retrospective design, potential for selection bias, and the uncertain generalizability of its results. Only one researcher coded patients’ clinical features, and many data were missing. Nevertheless, “it is studies like these that use the gold standard autopsy data and look backward that provide the heart of clinical diagnostic criteria,” said Dr. Weinstein.

Longitudinal, prospective studies could validate these results and potentially improve the Armstrong and Höglinger criteria.

Erik Greb

Suggested Reading

Armstrong MJ, Litvan I, Lang AE, et al. Criteria for the diagnosis of corticobasal degeneration. Neurology. 2013;80(5):496-503.

Höglinger GU, Respondek G, Stamelou M, et al. Clinical diagnosis of progressive supranuclear palsy: The movement disorder society criteria. Mov Disord. 2017;32(6):853-864.

Next Article: