Bi-phase age-related brain gray matter magnetic resonance T1ρ relaxation time change in adults

Abstract

Objectives

To investigate normative value and age-related change of brain magnetic resonance T1ρ relaxation at 1.5 T.

Methods

This study was approved by the local ethical committee with participants' written consent obtained. There were 42 adults healthy volunteers, including 20 males (age: 41 ± 16 (mean ± standard deviation) years, range: 22–68 years,) and 22 females (age: 39 ± 15 years, range: 21–62 years). MRI was performed at 1.5 T using 3D fluid suppressed turbo spin echo sequence. Regions-of-interests (ROIs) were obtained by atlas-based tissue segmentation and T1ρ was calculated by fitting the mean value to mono-exponential model. Correlation between T1ρ relaxation of brain gray matter regions and age was investigated.

Results

A regional difference among individual gray matter areas was noted; the highest values were observed in the hippocampus (98.37 ± 5.37 ms, median: 97.88 ms) and amygdala (94.95 ± 4.34 ms, median: 94.73 ms), while the lowest values were observed in the pallidum (83.81 ± 5.49 ms, median: 83.77 ms) and putamen (83.93 ± 4.76 ms, median: 83.99 ms). Gray matter T1ρ values decreased slowly (mean slope: − 0.256) and significantly (p < 0.05) with age in gray matter for subjects younger than 40 years old, while for subjects older than 40 years old there was no apparent correlation between T1ρ relaxation and age. Global white matter measured T1ρ value of 88.65 ± 3.47 ms (median: 87.86 ms), and the correlation with age was not significant (p = 0.18).

Conclusion

Gray matter T1ρ relaxation demonstrates a bi-phase change with age in adults of 22–68 years.

Introduction

Magnetic T1ρ relaxation has the potential to provide information about the low frequency motions (100 Hz to a few kilohertz) in biological systems [1], [2], [3]. T1ρ relaxation depends on T1 and T2 relaxation times as well as contributions from several MR interactions such as chemical exchange, dipolar interaction and J-coupling [1], [2]. Depending upon the tissue type, more than one mechanism may be operative simultaneously but with different relative contributions. During recent years, T1ρ relaxation has been increasingly used to explore the pathophysiology or predictive diagnostics of a number of neurological conditions [4], [5], [6], [7], [8], [9], [10]. Previous studies suggested that neurodegeneration may contribute to the increased T1ρ value in brain regions [11], [12], [13]. For example, cerebral atrophy, which reflects underlying neuronal loss, has been reported to be associated with increased T1ρ value in the hippocampus and medial temporal lobe of Alzheimer's disease [11]. Although the biophysical/biochemical mechanism remains to be further investigated, novel MRI techniques such as T1ρ, T1ρ dispersion, chemical exchange saturation transfer and its variant chemical exchange imaging with spin-lock technique may provide early imaging biomarkers for neurodegeneration diseases including Alzheimer's disease, Parkinson's disease and dementia [14], [15], [16], [17], [18]. These techniques have been refined to be increasingly more time-efficient, faster and more robust in recently years [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30].

The T1ρ and T2 values are correlated with T2 can be regarded as a special case of T1ρ with a spin lock frequency of 0 Hz. T2 relaxation has been explored as one of the contrast mechanism in brain disease characterization [31], [32], however, the relationship between T2 relaxation and brain physiological ageing remains not confirmed till now. Siemonsen et al. [33] studied 50 subjects (age: 12–91 years) and found an increase in T2 that linearly correlated with age in the thalamus and three white matter (WM) structures, but not in the caudate nucleus and lentiform nucleus. Ding et al. [34] studied 70 normal subjects (age: 3 weeks–31 years) and showed that T2 decreased with increasing age; the rate of decrease was greater at a younger age and slower in the years after, indicating a nonlinear relationship with age. Hasan et al. [35] studied 130 healthy subjects (age: 15–59 years) and reported the relation between T2 and age in whole brain gray and white matter, caudate nucleus, and the anterior limb of internal capsule followed a quadratic, U-shaped curve. More recently, Wang et al. [36] studied 77 normal subjects (age: 9–85 years) and reported brain tissue R2 (1/T2)-age correlations followed various time courses with both linear and nonlinear characteristics depending on the particular brain structure.

The relationship between brain tissue T1ρ and age has not been studied as extensively as T2 relaxation. Borthakur et al. [37] found no relationship between T1ρ vs. age in 16 elderly subjects (age: 70–91 years). Recently Watts et al. [38] studied 41 subjects (age: 18–76 years) and reported T1ρ values significantly decrease with age in cortical gray matter (GM), left and right caudate, putamen, hippocampus, amygdala, and nucleus accumbens; while increases with age were observed in white matter tracts. Information on age-related change in T1ρ relaxation is not only needed to gain a deeper understanding of brain ageing but also useful as a normative data set for examinations performed with T1ρ MRI in patients. The primary goal of this study was to further clarify the discrepancy seen in the reported literature [37], [38].