I. Introduction

Since the early 1990's, many hospitals in Japan have been offering brain imaging directly to the public, referred to as a "Brain Dock" or a "Brain checkup", primarily to identify aneurysms before they rupture.

As well, magnetic resonance imaging (MRI) has been increasingly used to detect vascular diseases, such as silent minor hemorrhages and lacunar strokes ^[1,2]. Since national medical insurance does not cover these checkup examinations, patients who choose to have these MRI must pay a substantial fee. The prevalence of these diseases among patients having MRI without insurance coverage has been previously reported ^[1,3].

We were requested by a branch office of a large company to do brain checkups for their employees because three employees had died of apoplexy within the last year. In Japan, it is a legal obligation that the employer pay for the medical checkups of their employees. Although such medical checkups do not usually include brain MRI because of their high cost, this company decided to includes MRI in their employees' routine checkups. We reviewed the employees' brain MRI examination, and it appeared that their T2-weighted images had more T2-elongated spots than those of self-referred subjects.

T2-elongated spots include various lesions, such as leukoencephalopathy ^[4], cystic infarction ^[5], Binswanger's disease ^[6], white matter lesions ^[3], and gliosis. Hypertension and aging are the reported risk factors for these lesions ^[3,7-15].

The purpose of this study was to evaluate the difference in T2-elongated spots (T2ES) between self-referred subjects and the company employees, referred to as third party-referred subjects.

II. Materials and methods

Subjects

In this retrospective study, MRI studies of 814 healthy adults obtained from February 2006 to August 2007 at the Kameda Medical Center, Chiba, Japan, were assessed. The subjects were categorized into two groups, based on the reason for the MRI study: Group A included 312 self-referred subjects (214 males, 98 females; age range, 49 to 65 years; mean age, 56.5 years) ; Group B included 502 third party-referred subjects (442 males, 60 females; age range, 49 to 65 years; mean age, 54.3 years) ^{(Table 1)}. All subjects provided their written informed consent and completed a questionnaire dealing with current and past diseases.

Brain MRI

Magnetic resonance imaging scans were performed on 1.0-Tesla MRI scanners (MAGNETOM EXPERT, Siemens). The MRI protocol consisted of axial T2-weighted spin echo (TR=40000, TE=99.0, image matrix was 1024 x 242 voxels, 6 mm thick, 1 mm gap, about 20 images per case.), coronal T1-weighted image (TR=520, TE=16.0, image matrix 512 x 256 voxels, 6 mm thick, 1.8 mm gap), axial FLAIR (TR=8002, TE=146.8, TI=2000, image matrix 256 x 192 voxels), and MRA.

Analysis of brain lesions

MRI T2-weighted image findings were used since they can be more clearly interpreted than other sequences and they provide good resolution. Furthermore, the differences between the two groups were difficult to detect using the other sequences.

Two radiologists interpreted the MRI scans independently, blinded to the subjects' past and present disease histories. They counted the number of T2-elongated spots (T2ES) of any size on all axial T2-weighted images, except in the perivascular spaces (Virchow-Robin spaces) ^[16,17].

If a spot was located adjacent to cerebrospinal fluid, it was evaluated with the assistance of axial FLAIR images. When the two radiologists differed, they reviewed the MRI together and reached a consensus on the correct number of spots for each such case.

The cases were graded from Grade 0 to 5 based on the grading system that was developed ^{(Table 2)}. As has already been described, T2ES increased with age. Therefore, age was taken into consideration, so that the number of T2ES could be evaluated relative to the subjects' age.

First, the base number of T2ES was calculated by subtracting 49 from the subjects' age, since 49 was the age of the youngest subjects studied.

Next, each case was categorized based on the difference between the base number and the actual number of T2ES. If the actual number of T2ES was equal to or up to 4 more than the base number, the subject was categorized as Grade 2; if the actual number ranged from 5 to 9 more than the base, the subject was categorized as Grade 3 ^{(Fig. 2)} ; if the actual number ranged from 10 to 14 more than the base, the subject was categorized as Grade 4; if the actual number was 15 more than the base number, or if the number of lesions was not countable the subject was categorized as Grade 5 ^{(Fig. 3)}. If the actual number was less than the base number, the subject was categorized as Grade 1, and if there were no T2ES, then the subject was categorized as Grade 0 ^{(Fig. 1)}. Thus, the grade was assigned in a manner that took into account the subjects' age.

Statistical Analysis

The prevalence of each T2ES Grade and the risk factors was calculated using the chi-square test with independence with Yates' correction. JSTAT-for Windows version 11.0 (Shareware; http://www.vector.co.jp/soft/win95/business/se030917.html; in Japanese) was used for statistical testing.

III. Results

Prevalence of each grade

No T2ES were found (Grade 0) in 36.2% of Group A and 34.4% of Group B. ^{(Table 3)}

The prevalence of Grade 4 was 20.0% in Group B and 13.5% in Group A (χ2=5.25; P=0.022) ^{(Fig. 4)}. Among males, the prevalence of Grade 3 was 18.8% in Group A and 11.8% in Group B (χ2=5.19; P=0.0228) ^{(Fig. 5)}. Among females, the prevalence of Grade 1 was 25.5% in Grade A and 6.8% in Group B (χ2=7.38; P=0.006) ^{(Fig. 6)}. The prevalence of Grade 4 was 25.4% in Group B and 11.2% in Group A (χ2= 4.39; P=0.036) ^{(Fig. 6)}.

Risk factors in each group

The prevalences of hypertension, diabetes mellitus, hyperlipidemia, and other diseases that could affect the MRI results are shown in Table 4.

The prevalence of hypertension was 22% in Group A and 20% in Group B (χ2=0.8; P=0.37; NS). The prevalence of diabetes mellitus was 15.0% in Group A and 9.6% in Group B (χ2=4.86; P=0.027). The prevalence of hyperlipidemia was 18.0% in Group A and 9.0% in Group B (χ2=13.29; P=0.0003). The prevalence of other diseases that could affect the MRI findings was 5.8% in Group A and 6.2% in Group B (χ2=0.002; P=0.9680; NS).

Table 5 shows the prevalence of other risk factors; there were no significant differences between Group A and Group B in these risk factors.

IV. Discussion

Relative Grading System

Kobayashi et al.reported the incidence of silent lacunar leison in Normal adults. They reported that the incidence of these lesion were correlated with their age. And they reported the subjects who had no silent lacunar lesion were all under age 49 ^[18]. Also, other researcher reported correlation its incidence with their age. ^[3,12,19]

As already mentioned, T2ES contained such silent lacunar lesion. and other lesions were also age related.

Our grading system was developed from these facts.

Using this grading system, significant differences between the two groups were identified; thus, this relative grading system was useful for doing this study.

However, this system has several limitations, Grade 0, which was defined as no T2ES on MRI, was found in approximately 39% of subjects; this finding contradicts the assumption that the number of T2ES is related to age.

In the future, the absolute number of T2ES will need to be evaluated.

Discrepancies between the risk factors and the T2ES results

The number of T2ES was greater in Group B than in Group A. However, Group A had more risk factors than Group B. The prevalence of diabetes mellitus and hyperlipidemia was greater in Group A than in Group B, while there was no difference between the groups in the prevalence of hypertension. Of note, the severity of the risk factors and the presence of other risk factors, such as increased creatinine levels and carotid artery stenosis ^[12], were not assessed; these other factors may have contributed to the discrepant findings.

It is possible that Group A's risk factors were mild and well controlled; the fact that the patients sought MRI examinations on their own initiative supports this hypothesis.

The utility of T2ES as a screening tool for cerebral vascular accidents

Several studies have reported on the significance of T2ES lesions. Golomb et al. reported that T2ES lesions were characterized by extensive arteriosclerosis ^[4]. Takao et al. reported that T2ES lesions were part of the normal aging process ^[7]. Kasahara et al. and Matsubayashi et al. reported that T2ES lesion had an effect on higher order cerebral function ^[10,11,20]. However, Scarpelli et al. cautioned that T2ES lesions on MRI were sensitive but not very specific ^[5].

Therefore, the utility of T2ES in disease prevention and determining patient prognosis is unclear. Further studies are warranted.

V. Conclusion

The presence of T2-elongated spots on brain MRI was assessed in self-referred and third party-referred subjects.

An ‘Age-related Grading System' was developed to compare the groups and was shown to be effective.

Grade 4, which was defined as the presence of 10 to 14 more T2-elongated spots than age minus 49, was significantly more common in the third party-referred group than in the self-referred group, while the self-referred group had a high prevalence of well controlled diabetes mellitus and hyperlipidemia. The findings suggest that the self-referred subjects were more health-conscious than the third party-referred subjects.

V. Acknowledgments

The author would like to thank Professor Hisao Ito of the Department of Radiology Postgraduate School of Chiba University for helpful comments and editorial support during manuscript presentation.

The author would also like to thank Director-General Toshihiro O'uchi of the Department of Radiology, Kameda Medical Center, and my colleague Asako Azuma for their efforts and help with the study.