Atrophy measurement of the anterior commissure and substantia innominata with 3T high-resolution MR imaging: does the measurement differ for patients with frontotemporal lobar degeneration and Alzheimer disease and for healthy subjects?

Abstract

Background and purpose: The anterior commissure (AC) and substantia innominata (SI) can be clearly demonstrated at 3T high-resolution MR imaging. Our aim was to investigate if atrophy of the AC and SI on 3T MR imaging differs among patients with frontotemporal lobar degeneration (FTLD) and Alzheimer dementia (AD) and healthy subjects.

Materials and methods: Seven consecutive patients with FTLD, 20 patients with AD, and 16 age-matched control subjects were enrolled. MR imaging was performed at 3T. The AC thickness as well as the SI thickness was measured on a thin-section coronal T2-weighted image, and the AC area was measured on a sagittal T1-weighted image. The measurement differences among the participants were analyzed with the Kruskal-Wallis test. A correlation of the measurement with the Mini-Mental State Examination (MMSE) score was obtained with the Spearman rank correlation test.

Results: Thinning of the AC was significantly more prominent in FTLD than in AD (P < .001). Although the right SI thickness was significantly decreased in patients with AD as compared with control subjects (P < .05), there was no significant difference, with a substantial overlap of the average SI thickness among the 3 groups. The thickness and the area of the AC were positively correlated with the MMSE score (rho = 0.612, P < .001, and rho = 0.659, P < .001, respectively). In contrast, the average SI thickness showed a weak positive correlation with the MMSE score (rho = 0.325, P < .05).

Conclusion: Measurement of AC atrophy with 3T MR imaging may provide additional diagnostic clues for FTLD and AD. Conversely, SI atrophy measurement does not provide an additional benefit in the evaluation of FTLD and AD, owing to a considerable overlap in the average thickness of bilateral SI.

Fig 1.

Schema of the MR imaging measurements. A, A sagittal T1-weighted inversion-recovery image shows the AC area (dotted circle and arrow) in a control subject. The corpus callosum (CC), fornix (f), and mamillary body (m) are identified. B, A coronal thin-sectional T2-weighted image shows the AC (a), the SI (i), the diagonal gyrus (d), and the ventral pallidum (v). AC (bidirectional arrow) and the SI thicknesses (arrows) are depicted. LV indicates left ventricle.

Fig 2.

Measurement of the AC and SI thicknesses. Coronal T2-weighted fast spin-echo (TR/TE, 3000/125.4 ms; section thickness, 2 mm) MR images show the AC thickness (bidirectional arrow) in a control subject (A; AC thickness, 4.30 mm), a patient with AD (B; AC thickness, 3.01 mm), and a patient with FTLD (C; AC thickness, 2.15 mm). The arrows indicate the measured part of the thickness of the SI. Thinning of the SI is noted in the patient with AD and to a lesser degree in the patient with FTLD.

Fig 3.

Boxplot of the AC area in patients with FTLD and AD and control subjects. The line across the box represents the median value. The box ends represent the first and third quartiles. The end points of each graph represent the smallest and largest values. The median AC area is the lowest in the FTLD group of patients compared with the patients with AD and control subjects. None of the AC cross-sectional area measurements in the FTLD group of patients reached the lowest value obtained for the control subjects.

Fig 4.

Boxplot of the AC thickness in patients with FTLD and AD and control subjects. The line across the box represents the median value. The box ends represent the first and third quartiles. The end points of each graph represent the smallest and largest values. The median AC thickness is the lowest in the FTLD group of patients compared with the patients with AD and control subjects. Although considerable overlap of the values between the FTLD and AD groups of patients exists, none of the AC measurements in the FTLD group of patients reached the lowest value obtained for the control subjects.

Fig 5.

Boxplot of the SI thickness in patients with FTLD and AD and control subjects. The line across the box represents the median value. The box ends represent the first and third quartiles. The end points of each graph represent the smallest and largest values. The median SI thickness tends to be lower in the AD group of patients compared with the FTLD group of patients and control subjects. Note the considerable overlap of the values between the patients with FTLD and control subjects.