Evaluating agreement with a gold standard in method comparison studies.

We develop a statistical model for method comparison studies where a gold standard is present and propose a measure of agreement. This measure can be interpreted as a population correlation coefficient in a constrained bivariate model. An estimator of this coefficient is proposed and its statistical properties explored. Applications of the new methodology to data from the medical literature are presented.

Benzodiazepine receptor binding in cerebellar degenerations studied with positron emission tomography.

We used positron emission tomography with [11C]flumazenil to study gamma-aminobutyric acid type A/benzodiazepine receptor binding quantitatively in the cerebral hemispheres, basal ganglia, thalamus, cerebellum, and brainstem of 72 subjects, including 14 with multiple system atrophy of the ataxic (olivopontocerebellar atrophy) type, 5 with multiple system atrophy of the extrapyramidal/autonomic (Shy-Drager syndrome) type, 18 with sporadic olivopontocerebellar atrophy, 15 with dominantly inherited olivopontocerebellar atrophy, and 20 normal control subjects with similar age and sex distributions. In comparison with data obtained from the normal control subjects, we found significantly decreased ligand influx in the cerebellum and brainstem of multiple system atrophy patients of the olivopontocerebellar atrophy type and in patients with sporadic olivopontocerebellar atrophy, but not in patients with multiple system atrophy of the Shy-Drager syndrome type. Despite these differences in ligand influx, benzodiazepine binding was largely preserved in the cerebral hemispheres, basal ganglia, thalamus, cerebellum, and brainstem in patients with multiple system atrophy of both types as well as those with sporadic or dominantly inherited olivopontocerebellar atrophy as compared with normal control subjects. The finding of relative preservation of benzodiazepine receptors indicates that these sites are available for pharmacological therapy in these disorders.

A comparison of cerebral blood flow and glucose metabolism in olivopontocerebellar atrophy using PET.

OBJECTIVE: In sporadic cases of olivopontocerebellar atrophy (OPCA), to determine whether local cerebral blood flow (lCBF) is reduced, whether lCBF is coupled to local cerebral metabolic rate for glucose (lCMRglc), and whether lCBF measurements are potentially useful in diagnosing OPCA. DESIGN: Positron emission tomography was used with [15O]H2O to measure lCBF and with [18F]fluorodeoxyglucose to measure lCMRglc in 17 patients with OPCA and 21 normal control subjects. RESULTS: In OPCA patients, lCBF was significantly decreased in the cerebellum, but not in the cerebral cortex, basal ganglia, thalamus, or brainstem. In the same patients, lCMRglc was significantly decreased in the cerebellum and brainstem, where the largest changes were observed, and also in the cerebral cortex, basal ganglia, and thalamus. The ratio of lCBF to lCMRglc, an indicator of the coupling of blood flow to metabolism, was similar in OPCA patients and normal subjects for all regions except the brainstem, where the ratio was marginally decreased in OPCA patients. Using logistic discriminant analysis to assess the ability of lCBF and lCMRglc to differentiate OPCA patients from normal subjects, we found the cross-validated sensitivity of absolute lCMRglc as a predictor of OPCA was 82% with a corresponding specificity of 71%; the sensitivity of absolute lCBF was 71% and the specificity 76%. CONCLUSIONS: In sporadic cases of OPCA, lCBF is reduced in the cerebellum, CBF remains coupled to lCMRglc, and the lCBF pattern is a useful predictor of the diagnosis.

Patterns of cerebral glucose metabolism detected with positron emission tomography differ in multiple system atrophy and olivopontocerebellar atrophy.

We used positron emission tomography with [18F]fluorodeoxyglucose to study local cerebral metabolic rates for glucose (ICMRglc) in patients with multiple system atrophy (MSA), sporadic olivopontocerebellar atrophy (sOPCA), and dominantly inherited olivopontocerebellar atrophy (dOPCA) in comparison with normal control subjects. IN MSA, absolute lCMRglc was significantly decreased in the brainstem, cerebellum, putamen, thalamus, and cerebral cortex. In sOPCA, absolute lCMRglc was significantly decreased in the brainstem, cerebellum, putamen, thalamus, and cerebral cortex. In dOPCA, absolute lCMRglc was significantly decreased in the brainstem and cerebellum but not in the other structures. Examination of lCMRglc normalized to the cerebral cortex in comparison with normal controls revealed in MSA significant decreases in the brainstem, cerebellum, and putamen but, in both sOPCA and dOPCA, significant decreases only in the brainstem and cerebellum. The findings indicate that these three disorders all show a marked decrease of lCMRglc in the brainstem and cerebellum but differ in the degree of hypometabolism in forebrain and cerebral cortical structures. The results are consistent with the possibility that, in many cases, sOPCA will evolve into MSA. Moreover, positron emission tomography may provide helpful diagnostic information in these neurodegenerative diseases.

Alternative calculation of the global clinical dementia rating.

The Clinical Dementia Rating (CDR) is a widely used instrument for rating the global severity of dementia, with scores in six independently rated categories used as a basis for computing a global score. We have found that the algorithms currently used to calculate the global score from the six component scores produce results that are sometimes inconsistent with the goals of the rating scale. We systematically analyzed these incongruities and developed an alternative algorithm that eliminates them while retaining the fundamental features of the original method.