ARTICLEJournal of Neuroimaging Vol 13 No 2 April 2003Imaizumi et al: Dynamics of Hemosiderin Spots

Dynamics of Dotlike HemosiderinSpots Associated WithIntracerebral Hemorrhage

Toshio Imaizumi, MD, PhD

Masahiko Chiba, MD, PhD

Toshimi Honma, MD

Junpei Yoshikawa, MD, PhD

Jun Niwa, MD, PhD

A B S T R A C T

The authors observed dotlike, low-intensity spots in T2*-weighted mag-

netic resonance imaging (MRI), subsequently diagnosed histologically

as previous microbleeds associated with lipofibrohyalinosis, amyloid

angiopathy, and small vessel disease. The nature of dotlike hemosiderin

spots (dotHSs), however, is still unknown. This case report seeks to

demonstrate the dynamics of dotHSs associated with an intracerebral

hematoma (ICH). T2*-weighted MRI of a 72-year-old man with a history

of hypertension demonstrated 4 dotHSs 24 months after a left

putaminal hemorrhage. Follow-up T2*-weighted MRI 40 months after

the acute event demonstrated the asymptomatic formation of 3 more

dotHSs, even with good control of blood pressure. Fifty months after the

stroke, T2*-weighted MRI showed that 2 of the new dotHSs had

become fainter, whereas the hemosiderin associated with the ICH scar

remained detectable. To the authors’ knowledge, this is the first

description of dotHS dynamics associated with ICH.

Key words: Microbleed, intracerebral hemorrhage, hemosiderin,

magnetic resonance imaging, dynamics.

Imaizumi T, Chiba M, Honma T, Yoshikawa J, Niwa J.

Dynamics of dotlike hemosiderin spots

associated with intracerebral hemorrhage.

J Neuroimaging 2003;13:155-157

DOI: 10.1177/1051228403251250

Ruptured small or medium-sized arteries, frequently damagedby hypertension-induced lipofibrohyalinosis, are the primarycause of intracerebral hemorrhage (ICH).1-3 Small, dotlike areasof signal loss—“dotlike hemosiderin spots” (dotHSs)—visualizedby gradient-echo T2*-weighted magnetic resonance imaging(MRI), which is especially sensitive for hemosiderin detection,4,5

indicate old extravasations of blood. These spots originate from

hemosiderin stored in macrophages.1,6,7 A dotHS significantlycorrelates with microangiopathy related to a small vessel disease(SVD), including ICH and lacunar infarction,1,6 amyloidangiopathy,1,7 and congenital angiopathy, such as cerebralautosomal dominant arteriopathy with subcortical infarct andleukoencephalopathy.8 Follow-up T2*-weighted examinationsof a patient with ICH demonstrated the dynamics of dotHSs inthis case study. To our knowledge, this is the first report explor-ing the dynamics of dotHS formation and absorption associatedwith ICH.

Case Presentation

A 69-year-old man with a long history of untreated hypertensionwas admitted to our hospital on May 12, 1998. A neurologicalexamination revealed right hemiparesis and dysarthria. Com-puted tomography (CT) showed a left focal putaminalhematoma on admission (Fig 1). Laboratory investigations, in-cluding HgA1c, bleeding time, and the standard biological ex-amination, were normal. The patient was neither a smoker noran alcohol drinker. Conservative treatment for hypertensionwas performed. The patient was discharged on June 17, 1998.His blood pressure was under good control continuously. InitialT2*-weighted MRI 24 months after the hemorrhage (May 22,2000) revealed 4 dotHSs, including 2 in the pons, 1 in the rightoccipital area (not shown), 1 in the right putamen, and a hemor-rhagic scar of the putaminal hematoma (Fig 2). Follow-up T2*-weighted MRI 40 months after the original hemorrhage (Sep-tember 20, 2001) demonstrated the formation of 3 more dotHSsin the right internal capsule, the right corpus callosum near thesplenium, and the left thalamus on an axial slice at the level ofthe pineal body (Fig 2); the patient was asymptomatic, and a CT

Copyright © 2003 by the American Society of Neuroimaging 155

Received September 19, 2002, and in revised form November 20,2002. Accepted for publication November 30, 2002.

From the Department of Neurosurgery, Hakodate MunicipalHospital, Japan.

Address correspondence to Toshio Imaizumi, MD, PhD, De-partment of Neurosurgery, Hakodate Municipal Hospital, 1-10-1 Minato-cho, Hakodate 041-0821, Hokkaido, Japan. E-mail:timaizum@d9.dion.ne.jp. Fig 1. Computed tomography of a 69-year-old man exhibiting a

left putaminal hemorrhage on admission (May 12, 1998).

scan could not detect the new dotHSs (not shown) (Fig 2). Fiftymonths after the hemorrhage ( July 25, 2002), additional T2*-weighted MRI showed that the new dotHSs in the right internalcapsule and right corpus callosum had become fainter. Hemo-siderin deposition from the putaminal hematoma was still visi-ble but appeared a bit fainter. On the other hand, the 2 dotHSs inthe pons and 1 in the right occipital area had not changed notice-ably.

We carefully made the radiographs in same manner by usingthe same 1.5-T scanner and the same developer. We obtainedT2*-weighted examinations in the axial plane with the following

parameters: a TR of 450 ms, a TE of 26 ms, 2 excitations, a flipangle of 20°, a section thickness of 10 mm without gaps, and amatrix of 256 × 205. We checked the corresponding areas of lowintensity and signal loss and ruled out cerebral calcification on aCT scan.

Discussion

The T2*-weighted technique is used for the detection of hemo-siderin in MRI.4,5 T2*-weighted MRI demonstrated a variety of

156 Journal of Neuroimaging Vol 13 No 2 April 2003

Fig 2. Follow-up T2*-weighted magnetic resonance imaging (MRI) demonstrating the dynamics of dotlike hemosiderin spots (dotHSs). (a),(b), and (c) (May 22, 2000): Initial T2*-weighted MRI, 24 months after a putaminal hemorrhage, revealed several dotHSs and an old hemorrhagicscar in the left putamen (b). (d), (e), and (f) (September 20, 2001):Follow-up T2*-weighted MRI, 40 months after the first stroke, demonstrated theformation of new dotHSs in the right internal capsule, the right corpus callosum near the splenium, and the left thalamus at the level of the pinealbody (e). The upper and lower slices were simultaneously investigated to compare hemosiderin deposition. (g), (h), and (i) (July 25, 2002): Addi-tional T2*-weighted MRI, 50 months after the original hemorrhage, demonstrated that dotHSs in the right internal capsule and the right corpus cal-losum had become fainter (h).

d fe

a cb

g ih

dense and thin, small and large dotHSs in this case. The diame-ters and densities of dotHSs, however, are likely to depend on aferromagnetic effect related to the amount of hemosiderin andthe parameters used for MRI. Therefore, we carefully per-formed T2*-weighted MRI in the same manner. Although ex-actly identical T2*-weighted MRI slices were not made, wesimultaneously investigated the dotHSs from upper and lowerslices to compare the condition and number of dotHSs (Fig 2).

The putaminal hematoma remained apparent > 50 monthsafter onset, and dotHSs remained visible for > 26 months.Although dotHSs are too small to be detected by CT, they mayremain detectable in T2*-weighted MRI for several years, as isthe case for ICH-associated hemosiderin.

DotHSs are significantly associated with hypertension.1,9,10

The number of dotHSs increased in this case, however, evenwith good blood pressure control. We guess that other factorsaffect the formation of new dotHSs or that the microangiopathyassociated with dotHS can progress in the absence ofhypertension.

This case, in which 7 dotHSs, including nascent ones, wereidentified, may represent > 7 episodes of intracerebralmicrobleeding, which fortunately did not enlarge to symptom-atic ICH. Some mechanisms may affect the enlargement of amicrobleed. DotHSs may be a potential risk factor for early cere-bral bleeding after ischemic stroke.11 Furthermore, dotHSs maybe markers of increased risk for hemorrhagic transformation inpatients receiving thrombolitic therapy for acute ischemicstroke.12 Our preliminary study demonstrates that a large num-ber of dotHSs may be associated with the severity of angiopathyand related to the recurrence of SVD.13 Further study, ideallywith pathologically verified diagnoses, must confirm this rela-tionship between large or increasing numbers of dotHSs and aprognosis of SVD.

References

1. Fazekas F, Kleinert R, Roob G, et al. Histopathologic analy-sis of foci of signal loss on gradient-echo T2*-weighted MRimages in patients with spontaneous intracerebral hemor-rhage: evidence of microangiopathy-related microbleeds.Am J Neuroradiol 1999;20:637-642.

2. Kalimo H, Kaste M, Haltia M. Hypertensive angiopathy. In:Graham DI, Lantos PL, eds. Greenfield’s Neuropathology. 6thed. London: Arnold; 1997:325-331.

3. Lammie GA, Brannan F, Slattery J, Warlow C.Nonhypertensive cerebral small-vessel disease. An autopsystudy. Stroke 1997;28:2222-2229.

4. Atlas SW, Thulborn KR. MR detection of hyperacuteparenchymal hemorrhage of the brain. Am J Neuroradiol1998;19:1471-1507.

5. Wardlaw JM, Dennis MS, Warlow CP, Sandercock PA.Imaging appearance of the symptomatic perforating arteryin patients with lacuner infarction: occlusion or other vascu-lar pathology? Ann Neurol 2001;50:208-215.

6. Tanaka A, Ueno Y, Nakayama Y, Takano K, Takebayashi S.Small chronic hemorrhages and ischemic lesions in associa-tion with spontaneous intracerebral hematomas. Stroke1999;30:1637-1642.

7. Good CD, Ng VW, Clifton A, Britton JA, Hart Y, Wilkins P.Amyloid angiopathy causing widespread miliary haemor-rhage within the brain evident on MRI. Neuroradiol1998;40:308-311.

8. Dichgans M, Holtmannspotter M, Herzog J, Peters N,Bergmann M, Yoursry TA. Cerebral microbleeds inCADASIL. A gradient-echo magnetic resonance imagingand autopsy study. Stroke 2002;33:67-71.

9. Kinosita T, Okudera T, Tamura H, Ogawa T, Hatazawa J.Assessment of lacunar hemorrhage associated with hyper-tensive stroke by echo-planar gradient-echo T2*-weightedMRI. Stroke 2000;31:1646-1650.

10. Kwa VIH, Franke CL, Verbeeten B Jr, Stam J. Silentintracerebral microhemorrhage in patients with ischemicstroke. Ann Neurol 1998;44:372-377.

11. Nighoghossian N, Hermier M, Adeleine P, et al. Oldmicrobleeds are a potential risk factor for cerebral bleedingafter ischemic stroke. A gradient-echo T2*-weighted brainMRI study. Stroke 2002;33:735-742.

12. Kidwell CS, Saver JL, Villablanca JP, et al. Magnetic reso-nance imaging detection of microbleeds beforethrombolisis. An emerging application. Stroke 2002;33:95-98.

13. Imaizumi T, Honma T, Chiba M, Niwa J, Miyazaki Y. Riskfactor for cerebral hemorrhage for the patient with lacunarinfarction: investigation of 5 cases associated with both ofsymptomatic hemorrhagic and ischemic strokes. No ToShinkei 2001;53:737-741.

Imaizumi et al: Dynamics of Hemosiderin Spots 157