Elsevier

NeuroImage

Volume 105, 15 January 2015, Pages 486-492
NeuroImage

Validation of quantitative susceptibility mapping with Perls' iron staining for subcortical gray matter

https://doi.org/10.1016/j.neuroimage.2014.11.010Get rights and content

Highlights

  • In situ or in vivo QSM is compared with Perls' iron stain in subcortical gray matter.

  • Full slice Perls' iron staining enabled spatial maps of ferric iron to relate directly to MRI.

  • High linear correlations are found between QSM and Perls' iron stain in postmortem subjects with multiple sclerosis.

  • High linear correlations are also found between QSM and R2* for both postmortem and in vivo subjects.

Abstract

Quantitative susceptibility mapping (QSM) measures bulk susceptibilities in the brain, which can arise from many sources. In iron-rich subcortical gray matter (GM), non-heme iron is a dominant susceptibility source. We evaluated the use of QSM for iron mapping in subcortical GM by direct comparison to tissue iron staining. We performed in situ or in vivo QSM at 4.7 T combined with Perls' ferric iron staining on the corresponding extracted subcortical GM regions. This histochemical process enabled examination of ferric iron in complete slices that could be related to susceptibility measurements. Correlation analyses were performed on an individual-by-individual basis and high linear correlations between susceptibility and Perls' iron stain were found for the three multiple sclerosis (MS) subjects studied (R2 = 0.75, 0.62, 0.86). In addition, high linear correlations between susceptibility and transverse relaxation rate (R2*) were found (R2 = 0.88, 0.88, 0.87) which matched in vivo healthy subjects (R2 = 0.87). This work validates the accuracy of QSM for brain iron mapping and also confirms ferric iron as the dominant susceptibility source in subcortical GM, by demonstrating high linear correlation of QSM to Perls' ferric iron staining.

Introduction

Iron accumulation in subcortical gray matter (GM) may serve as an important biomarker of normal aging (Aquino et al., 2009, Cherubini et al., 2009, Hallgren and Sourander, 1958, Schenck and Zimmerman, 2004), and of neurological diseases including Alzheimer's disease, Parkinson's disease, Huntington's disease and multiple sclerosis (MS) (Berg and Youdim, 2006, Chen et al., 1993, Dexter et al., 1991, Khalil et al., 2011, LeVine, 1997, Williams et al., 2012). The mechanisms behind iron accumulation are not yet fully understood, although iron may accumulate through inflammatory and destructive processes (Stephenson et al., 2014), and may relate to the presence and extent of neurodegeneration. Measuring the state of brain iron metabolism may provide important information on aging and neurological diseases.

MRI provides a variety of contrast mechanisms that are sensitive to brain iron (Haacke et al., 2005) including transverse relaxation rates R2 and R2*, and susceptibility methods such as phase and susceptibility-weighted imaging. Previous studies in healthy subjects have shown that R2 and R2* increase in iron-rich brain regions and correlate strongly with iron concentration (Drayer et al., 1986, Gelman et al., 1999, Langkammer et al., 2010, Li et al., 2009, Peran et al., 2007, Thomas et al., 1993). While sensitive to iron, R2 and R2* may be affected by other sources such as macromolecular and water content changes (Mitsumori et al., 2012), which makes them not specific to brain iron. The introduction of phase imaging minimizes the influence of changes in macromolecular and water content, and is able to distinguish between negative and positive susceptibility sources (Duyn et al., 2007, Haacke et al., 2004, Rauscher et al., 2005). In addition, phase imaging has demonstrated good correlation to brain iron in subcortical GM (Haacke et al., 2007, Ogg et al., 1999, Yao et al., 2009). However, the non-local field properties of phase imaging cause it to be dependent on the shape and orientation of the object to the main magnetic field (Li and Leigh, 2004, Marques et al., 2009), which complicates interpretation.

The developing field of quantitative susceptibility mapping (QSM) inherits the iron sensitivity from phase imaging while eliminating the problem of non-locality. Derived from a deconvolution process from phase images, QSM unveils the local tissue susceptibility directly (De Rochefort et al., 2010, Kressler et al., 2010, Li et al., 2011, Liu et al., 2009, Liu et al., 2011, Reichenbach, 2012, Schweser et al., 2011, Shmueli et al., 2009, Wharton and Bowtell, 2010). A number of in vivo susceptibility maps have shown good correlations with subcortical GM iron concentrations (Bilgic et al., 2012, Schweser et al., 2011, Wu et al., 2012) as estimated from the hallmark study on brain iron by Hallgren and Sourander (1958). Nevertheless, validation of QSM for brain iron mapping requires postmortem studies that make a direct comparison between MRI and histochemistry. Only two human postmortem studies have been performed to date that compare QSM to histochemically measured iron content in subcortical GM. These studies used mass spectrometry (Langkammer et al., 2012) or X-ray emission and fluorescence (Zheng et al., 2013). The Langkammer et al. (2012) study provided absolute iron values but in small samples that do not provide a full spatial map of the tissue to relate to the susceptibility map, while the work by Zheng et al. (2013) used previously frozen formalin fixed tissue for MRI rather than in situ imaging. Furthermore, both studies examined total iron (ferrous and ferric). Thus to further validate QSM for subcortical GM iron mapping and to verify ferric iron as the main susceptibility source, there remains a need to compare in situ and in vivo susceptibility maps directly to spatial maps of ferric iron. In this study, we make use of Perls' iron staining (Meguro et al., 2007) to obtain full slice spatial maps of relative ferric iron content and compare to in situ and in vivo QSM in subcortical GM.

Section snippets

Subjects

In situ or in vivo QSM followed by Perls' iron staining was performed on three subjects who have been previously studied for phase, R2, and R2* mapping (Walsh et al., 2013). Subject 1 was a 63 year old male imaged in situ 28 h after death. Subject 2 was a 60 year old male imaged in situ 7 h after death. Subject 3 was a 45 year old male imaged in vivo one year before death. Subjects 1 and 2 had secondary progressive MS with Expanded Disability Status Scale (EDSS) scores of 8.5 before death, and

Results

Fig. 3 illustrates three coronal brain images from Subject 3 (in vivo) including field, susceptibility, and R2* maps and the Perls' iron stains. The field maps suffer from strong dipole effects which are resolved in the susceptibility maps, providing clear delineation between iron-rich regions. Subcortical GM hyperintensities in susceptibility and R2* maps correspond well to hypointensities in Perls' iron stains.

The resulting correlations of susceptibility to Perls' iron stain are shown in

Discussion

To compare susceptibility directly to ferric iron, we performed whole slice Perls' iron staining after in vivo or in situ QSM. This process enabled similar large ROI analysis on both MRI and Perls' stains, rather than highly localized samples. Furthermore, we performed in situ MRI shortly after death, to avoid extraction and fixation which can substantially alter MRI properties (Dawe et al., 2009, Van Duijn et al., 2011). Our approach yielded high correlations between susceptibility and ferric

Acknowledgments

Grant support from Canadian Institute of Health Research (MOP102582) and the Multiple Sclerosis Society of Canada (EGID1619) is acknowledged.

References (54)

  • S. Wharton et al.

    Whole-brain susceptibility mapping at high field: a comparison of multiple- and single-orientation methods

    Neuroimage

    (2010)
  • B. Yao et al.

    Susceptibility contrast in high field MRI of human brain as a function of tissue iron content

    Neuroimage

    (2009)
  • W. Zheng et al.

    Measuring iron in the brain using quantitative susceptibility mapping and X-ray fluorescence imaging

    Neuroimage

    (2013)
  • L.M. Al-Alousi et al.

    Multiple-probe thermography for estimating the postmortem interval: II. Practical versions of the Triple-Exponential Formulae (TEF) for estimating the time of death in the field

    J. Forensic Sci.

    (2001)
  • D. Aquino et al.

    Age-related iron deposition in the basal ganglia: quantitative analysis in healthy subjects

    Radiology

    (2009)
  • D. Berg et al.

    Role of iron in neurodegenerative disorders

    Top. Magn. Reson. Imaging

    (2006)
  • C. Birkl et al.

    Iron mapping using the temperature dependency of the magnetic susceptibility

    Magn. Reson. Med.

    (2014)
  • A. Bizzi et al.

    Role of iron and ferritin in MR imaging of the brain: a study in primates at different field strengths

    Radiology

    (1990)
  • J.C. Chen et al.

    MR of human postmortem brain tissue: correlative study between T2 and assays of iron and ferritin in Parkinson and Huntington disease

    AJNR Am. J. Neuroradiol.

    (1993)
  • R.J. Dawe et al.

    Postmortem MRI of human brain hemispheres: T2 relaxation times during formaldehyde fixation

    Magn. Reson. Med.

    (2009)
  • L. De Rochefort et al.

    Quantitative susceptibility map reconstruction from MR phase data using bayesian regularization: validation and application to brain imaging

    Magn. Reson. Med.

    (2010)
  • D.T. Dexter et al.

    Alterations in the levels of iron, ferritin and other trace metals in Parkinson's disease and other neurodegenerative diseases affecting the basal ganglia

    Brain

    (1991)
  • B. Drayer et al.

    MRI of brain iron

    AJR Am. J. Roentgenol.

    (1986)
  • J.H. Duyn et al.

    High-field MRI of brain cortical substructure based on signal phase

    Proc. Natl. Acad. Sci. U. S. A.

    (2007)
  • N. Gelman et al.

    MR imaging of human brain at 3.0 T: preliminary report on transverse relaxation rates and relation to estimated iron content

    Radiology

    (1999)
  • E.M. Haacke et al.

    Susceptibility weighted imaging (SWI)

    Magn. Reson. Med.

    (2004)
  • E.M. Haacke et al.

    Establishing a baseline phase behavior in magnetic resonance imaging to determine normal vs. abnormal iron content in the brain

    J. Magn. Reson. Imaging

    (2007)
  • Cited by (126)

    View all citing articles on Scopus
    View full text