Previous studies with 1.5 T or 3.0 T magnetic resonance imaging (MRI) have produced mixed results regarding the structural changes of the hippocampus in major depressive disorder (MDD). Subtle region-specific hippocampal tissue changes might be more sensitively detected by measuring the T2* relaxation time (T2*-RT) by ultra-high-field (UHF) MRI, as it provides much higher contrast and sensitivity and consequently greater resolution. We assessed the T2*-RTs of hippocampal sub-regions in 16 MDD patients (9 with recurrent MDD) and 16 control subjects using an UHF 7.0 T MRI system. T2*-RTs of CA1, CA2, CA3, CA4, and subiculum were calculated for both left and right hippocampus. MDD patients had significantly longer T2*-RTs in the right CA1 and subiculum than control subjects. Patients with recurrent MDD had significantly longer T2*-RTs in the right subiculum than those experiencing a first depressive episode, and longer T2*-RTs in the right CA1, CA3, and subiculum than control subjects. Values for T2*-RTs of the right CA3 were significantly correlated with illness duration. In conclusion, we report that T2*-RTs in the right subiculum and CA1 were increased in patients with MDD, especially in cases of recurrent MDD. These findings suggest that region-specific hippocampal damage may be occurring in recurrent depression.
Numerous studies have reported reduced hippocampal volume in major depressive disorder (MDD). However, this MDD-induced reduction in hippocampal volume has not been consistently demonstrated (Campbell et al., 2004, Eker and Gonul, 2009 and McKinnon et al., 2009). The inhomogeneity of depressed subjects has been suggested as one probable reason for the discrepancy. The most prevailing view is that hippocampal reduction can be found mainly in subjects with longer illness duration (Sheline et al., 1999 and Colla et al., 2007). In addition, differences in disease severity (Vakili et al., 2000), age of onset (Lloyd et al., 2004 and Janssen et al., 2007), experience of stressful events (Gianaros et al., 2007 and Vythilingam et al., 2002), antidepressant medication (Sheline et al., 2003), treatment-resistance (Frodl, Meisenzahl, Zetzsche et al., 2004, Frodl T et al., 2008 and Kronmuller, Pantel, Kohler et al., 2008), or genotype (Frodl, Meisenzahl, Zill et al., 2004 and Taylor et al., 2005) have been suggested to be related to the discrepancy. A recent review on structural imaging studies of the hippocampus in MDD patients suggested that a smaller hippocampus was prominent in patients of old age or with severe or recurrent episodes (Eker and Gonul, 2009).
Another plausible cause for this controversy may lie in the imaging methods used in prior studies. Measurements of the total hippocampal volume using conventional imaging resolutions had inevitable limitations in detecting subtle hippocampal changes. Moreover, specific hippocampal sub-regions may be predominantly affected by depression, while others may remain relatively intact. To overcome this limitation, some researchers have conducted shape analysis of the hippocampus (Posener et al., 2003 and Ballmaier et al., 2008); however, even with this post-processing technique, the hippocampal sub-regions could not be well differentiated in detail using previous imaging resolutions widely available to most clinics.
T2 or T2* relaxometry may be a more sensitive measure for hippocampal tissue composition compared to volumetric measurements. T2* relaxation refers to the decay of transverse magnetization caused by a combination of spin–spin relaxation. In addition, it also refers to additional spin dephasing which causes the magnetic field inhomogeneities and the differences in magnetic susceptibility between tissues. T2* images are commonly used for evaluating the water content of tissue. T2* relaxation also correlates with iron content (Gelman et al., 1999) or steady-state regional blood volume (Anderson et al., 2005) in the tissue. Therefore, T2 or T2* relaxation time (T2*-RT) has been shown to correlate with brain tissue damage (Cross et al., 2008) or resting neuronal activity (Teicher et al., 2000 and Small et al., 2000).
It has been reported that depressive symptoms were related to hippocampal T2-RT in whole-brain wise measurements of T2-RT in sleep apnea patients (Cross et al., 2008). The relationship between depressive symptoms and hippocampal T2-RT has also been reported in patients with epilepsy (Nees et al., 2001). However, hippocampal T2*-RT in depression has not been extensively investigated, mainly due to the limited resolution of 1.5 T and 3.0 T MR imaging.
Recently, ultra-high-field (UHF) MRI systems have become available for in-vivo human research. The UHF MRI such as 7.0 T has the potential of providing substantially better images in resolution and contrast (Li et al., 2006, Duyn et al., 2007 and Cho et al., 2009). Especially, T2* contrast at UHF has reveal a number of brain structures which were not visible in 1.5 T or 3.0 T MR images (Li et al., 2006 and Duyn et al., 2007). The hippocampal boundaries are well-demarcated in UHF T2* contrast images clearly-demonstrating hippocampal sub-regions. In addition to the contrast, the iron concentration and the resting oxygen levels can change signal intensity in MR images, especially at high field (Schenck and Zimmerman, 2004). Thus using T2* relaxometry with UHF MRI may be more sensitive in the detection of abnormalities compared to T2*-weighted image obtained utilizing conventional scanners.
Based on prior studies suggesting hippocampal damage in patients with MDD, we hypothesized that T2*-RT may be increased in the hippocampus of MDD patients. The aim of the current study was to assess the T2*-RT of hippocampal sub-regions in MDD patients using a T2* contrast images from UHF 7.0 T MRI. We have also investigated the relationship between the T2*-RT of hippocampal sub-regions and the clinical characteristics of MDD.
