Background: Although cerebrospinal fluid (CSF) biomarkers and medial temporal lobe atrophy (MTA) contribute to the diagnosis of Alzheimer disease (AD), they may not be specific. Relatively little is known about how they correlate with each other.
Purpose: To identify the validity of the radiological linear measurements of brain atrophy in AD diagnosis by examining the correlation with CSF biomarkers and by examining if specificity could be improved in classification of AD from controls, when the linear measurements are combined with the CSF biomarkers.
Material and Methods: 59 controls (20 male/39 female, age 73±8 years), 162 pure AD patients (49/113, 74±7 years), and 86 AD patients with minor cerebrovascular changes (CVC) (31/55, 77±5 years), aged between 52 and 94 years, were recruited from the Malmö Alzheimer Study. AD patients were subgrouped into “pure AD” and “AD + CVC” in order to clarify the possible influence of CVC on atrophy or CSF biomarkers in AD patients. Abeta42, T-tau, and P-tau in CSF were examined. Computed tomography (CT) linear measurements were performed, which included temporal horn ratio and suprasellar cistern ratio that reflect MTA.
Results: Compared with the 14 significant correlations between the CT measurements and three CSF biomarkers in the pure AD group, there was only one significant correlation in the AD + CVC group and one in the control group. In particular, P-tau correlates with temporal horn ratio only in pure AD. When the CT measurements were added with CSF biomarkers as independent variables in discriminant analysis, the percentage of correct classification of AD + CVC from controls increased from 79.5% (only CSF biomarkers) to 84.6% (combined CT measurements with CSF biomarkers). However, little was changed in the pure AD group.
Conclusion: P-tau correlates with the linear CT measure of MTA only in pure AD without CVC. Combined with the measure of MTA, the specificity of CSF biomarkers can be increased, but only in AD + CVC. The linear measurements of MTA are of value in AD diagnosis.
