[en] Objective: The purpose of this study was to investigate the effect of age on the iron concentration of the human brain. Methods: The brain iron level was evaluated in vivo in 78 healthy adult volunteers using a noninvasive magnetic resonance method termed susceptibility weighted imaging. The subjects were divided intothree groups due to different ages: young (22-35 years old, n=27), middle- aged (36-55 years old, n=35), and aged (56-78 years old, n=16). The phase values were measured on the corrected phase images in the globus pallidus, putamen, caudate, substantia nigra, red nucleus, thalamus and frontal white matter. The phase values of those regions measured from the subjects over than 30 years old were correlated with published values of brain iron concentration in normal adults to check the validity of the data. Then, the phase values of the three groups were tested for significant age-related differences using one-way ANOVA, followed by post hoc testing using least significant difference (LSD) procedure. Regression analysis was used to further examine age-related effects revealed by group comparisons, and to estimate the rates of age-related changes. Results: A strong negative correlation was found between the phase values and the published values of the brain iron concentration (r=-0.796, P= 0.032), which indicated that the higher the iron deposition level, the greater the negative phase values. In the putamen (F=20.115, P<0.01) and frontal white matter (F=3.536, P=0.034), significant differences were detected in the phase values of the three age groups. Linear regression analysis showed that phase values of the putamen, frontal white matter, and red nucleus decreased with age (The regression coefficients were -0.001, -0.001, and < -0.001 respectively, and the P value were all < 0.05), which indicated that the iron concentration of those brain structures increased with age. No significant age- related changes of the iron concentration were found in the globus pallidus, caudate, substantia nigra, and thalamus. Conclusions: These findings extended our knowledge of the patterns of the brain iron accumulation in normal aging. Such information is necessary to understand disease-related changes that involve the brain iron deposition. (authors)