SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Aim  Wallerian degeneration is a radiological finding thought to reflect corticospinal tract degeneration. This finding on magnetic resonance imaging (MRI) is routinely used as a predictor of poor prognosis in childhood stroke. However, its validity has never been established. Our objective was to correlate Wallerian degeneration seen on MRI with histopathology.

Method  We searched the databases of the Department of Pathology and Children’s Stroke registry at the Hospital for Sick Children, Toronto for autopsy specimens exhibiting focal infarcts from children born at term who underwent MRI after a stroke. The specimens were examined for Wallerian degeneration and then correlated with the pre-mortem MRI findings.

Results  Seven children (four females, three males) with a median age of 11 years (1–17y) at the time of stroke met the inclusion criteria for this study. Of the seven children included in the study with ischaemic or haemorrhagic infarcts, six had concordant Wallerian degeneration findings on both MRI and post-mortem histopathological examination. The median time between stroke and death was 20 days (3–1825d).

Interpretation  Our results show for the first time that the radiographic finding of Wallerian degeneration is a valid biomarker of corticospinal tract degeneration in children who have had ischaemic or haemorrhagic stroke.


Abbreviations
DCST

Descending corticospinal tract

DWI

Diffusion-weighted imaging

What this paper adds

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  •  Corticospinal tract MRI signal abnormality after stroke represents true loss of neuronal processes known as Wallerian degeneration.
  •  The first to validate the MRI finding of Wallerian degeneration in children with corresponding histopathological changes.
  •  Wallerian degeneration occurs in both hemorrhagic and ischemic strokes in children.
  •  This study provides the clinician with a reliable prognostic marker in childhood stroke.

Wallerian degeneration is a progressive antegrade demyelination and disintegration of the distal axons following injury to the cell body or proximal axon. The early stage of Wallerian degeneration is characterized by axonal swelling and breakdown of the myelin sheath. Loss of myelinated fibres and infiltration of foamy macrophages occur in the subacute stage. The degenerating fibres are then replaced by gliosis in the chronic stage.1 In arterial ischaemic stroke, Wallerian degeneration appears as an abnormal magnetic resonance imaging (MRI) signal within neuronal tracts emanating from the zone of primary insult beyond the borders of the affected vascular territory.2,3 Wallerian degeneration as a radiological finding has been correlated with poor prognosis in childhood4–7 and adult8 arterial stroke.

Wallerian degeneration is a well-documented sequela of arterial ischaemic stroke in adults.2,3 Kuhn et al.2 divided Wallerian degeneration into four radiographic stages. In stage 1, less than 4 weeks post injury, there is a minimal change in axonal chemical integrity, resulting in absence of signal abnormality on MRI. In stage 2, 4 to 10 weeks post injury, hypointense signal abnormalities of the descending corticospinal tract (DCST) are seen on T2-weighted MRI (acute Wallerian degeneration). In stage 3, after 10 to 14 weeks, the signal becomes hyperintense on T2-weighted MRI. Stage 4, several years later, is characterized by volume loss, seen as unilateral brainstem atrophy on MRI. The authors attributed these changes in signal intensity to the sequential breakdown of protein and lipid content of myelin and the subsequent changes in water content. However, there was no pathological association with these changes in MRI signal intensity.

Signal changes representing acute Wallerian degeneration in the DCST have now been reported following neonatal and childhood arterial ischaemic stroke and correlate with poor motor outcome.5–7 A recent study9 has shown that imaging changes in the DCST are indicative of axonal degeneration. This was detected by neurofilament staining in a neonatal rat model of hypoxic–ischaemic injury. However, the validity and true nature of this MRI finding have not been established in the human brain. The aim of this study was to ascertain whether the findings of Wallerian degeneration on MRI in post childhood arterial stroke indeed correlate with post-mortem histopathological changes of Wallerian degeneration.

Method

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The databases of the Department of Pathology and Children’s Stroke registry at the Hospital for Sick Children, Toronto, Canada were searched for autopsy specimens from children registered between January 1990 and June 2010. Specimens with a diagnosis of focal cerebral ischaemic or haemorrhagic infarct were included if (1) the patient was born at term, (2) the cerebral infarct had well-defined borders, and (3) the patient had undergone MRI after the stroke.

The presence of oedema and high signal intensity within the DCST and cerebral peduncles on diffusion-weighted imaging (DWI) were the criteria for acute Wallerian degeneration. Atrophy in cerebral peduncles was indicative of chronic or end-stage Wallerian degeneration. Pathology specimens were stained with haematoxylin and eosin or Luxol Fast Blue. Wallerian degeneration was considered to be present if, in recent cases of infarction, there was evidence of myelin breakdown and/or axonal swellings, or in older cases there was atrophy, gliosis, and myelin loss. The pathology specimens and the MRI images (T1-weighted axial, T2-weighted axial, and DWI) were examined for features of Wallerian degeneration by the study neuropathologist (CH) and neuroradiologist (DA) respectively, both of whom were blinded. The study neuropathologist was also blinded to the side of the lesion clinically and radiologically when examining the sections. The study was approved by the Research Ethics Board of the Hospital for Sick Children, Toronto.

Results

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Seven children (four females, three males; median age 11y, age range 1–17y) met the inclusion criteria. Demographic details and causes of death are shown in Table I. In six out of seven children, the MRI findings correlated with the pathological findings of the autopsy specimens, that is Wallerian degeneration was either present or absent in both modalities (Table II). In three children, MRI signal changes within the DCST corresponded with pathological changes (Fig. 1). In the other three children, in whom MRI did not reveal signal changes, histological changes in the white matter were also absent (Fig. 2). Only one child out of seven showed pathological changes consistent with Wallerian degeneration in the DCST with no corresponding MRI signal changes (false negative on MRI). This patient underwent serial MRI imaging at 1, 6, and 16 weeks and 5 years after stroke and all images were negative for Wallerian degeneration. Wallerian degeneration occurred as a result of ischaemic and haemorrhagic childhood stroke (Fig. 1). Two out of four children with pathological evidence of Wallerian degeneration experienced haemorrhagic stroke.

Table I. Relevant clinical data for patients included in the study
Patient no./sexAge at stroke, yStroke typeVascular territoryInfarcted areaCause of death
  1. M, male; F, female; MCA, middle cerebral artery.

1 M16IschaemicLeft MCALeft frontal lobe, basal gangliaAcute ischaemic stroke
2 M17HaemorrhagicMultiple, left MCALeft frontoparietal lobes, left cerebellumMultiorgan failure
3 F10IschaemicRight MCARight frontal, parietal, temporal, occipital lobesAcute ischaemic stroke
4 F1IschaemicLeft MCALeft basal gangliaRuptured MCA aneurysm
5 F17IschaemicRight MCARight frontal lobeXanthoastrocytoma
6 F9HaemorrhagicRight MCARight insula, globus pallidusAcute viral hepatitis
7 M11IschaemicLeft MCALeft cerebral hemisphereCardiac failure
Table II. Association of radiographic white matter signal abnormalities with pathological changes
Patient no.Time from stroke to MRITime from stroke to deathWD on MRI ipsilateral to strokeWD on pathology
  1. MRI, magnetic resonance imaging; WD, Wallerian degeneration; CST, corticospinal tract.

11d6dNo WDNo WD
21d20dWD of corpus callosum diffusion restrictionVacuolation of corpus callosum
31d4dNo WDNo WD
41d3dNo WDNo WD
55d, 12wks15wksWD CST, T2 hyperintensityMyelin loss and axonal swellings of CST
61wk, 6wks, 16wks, 5y5yNo WDMyelin loss and gliosis of CST
75y5yWD of left CST, corpus callosum, cerebellar peduncle with atrophyMyelin loss and gliosis of CST
image

Figure 1.  The three children in whom white matter signal changes were detected on magnetic resonance imaging (MRI) had corresponding histopathological changes consistent with Wallerian degeneration. Patient 2: Axial diffusion weighted imaging (a) during the first week after left middle cerebral artery (MCA) haemorrhagic stroke showed a hyperintense signal within the corpus callosum (arrow). Histological sections from corresponding regions of the corpus callosum showed vacuolation of white matter (b). Patient 5: Axial T2 fluid-attenuated inversion recovery (FLAIR) MRI (c) revealed hyperintense signal of the corticospinal tract at the level of the pons (arrow) 4 weeks after right MCA ischaemic stroke that was consistent with features of Wallerian degeneration. The tumour extends to the ventral border of the pons without infiltrating the pons (arrowhead). Histological sections at the level of the pons and the medulla (d) showed myelin loss, vacuolation, and axonal swellings. Patient 7: Axial T2 MRI (e) showed left MCA ischaemic stroke with encephalomalacia (arrow), local ventriculomegaly (long arrow), and atrophy of the corticospinal tracts at the level of the pons (arrowhead) more than 15 weeks after the stroke. Histological sections of the medulla (f) showed myelin loss and gliosis of the ipsilateral corticospinal tract. All histological sections are stained with haematoxylin and eosin or Luxol Fast Blue.

Download figure to PowerPoint

image

Figure 2.  Representative images from the three children. in whom the absence of magnetic resonance imaging (MRI)I signal abnormalities in the descending corticospinal tract was confirmed by intact myelin on histology. Patient 4: Axial diffusion-weighted imaging (a) and axial T2 fluid-attenuated inversion recovery MRI sequences (b) during the first week of presentation show a large left middle cerebral artery ischaemic stroke with no change in signal intensity in the corpus callosum or cerebral peduncles respectively. This was correlated with intact myelin histology at the midbrain/cerebral peduncles level (c).

Download figure to PowerPoint

Signal changes resulting from Wallerian degeneration were evident on MRI acutely in patient 2, but not in patient 5, within 1 month post stroke (Fig. 1). Histologically, there was vacuolation and axonal swelling in the corpus callosum and DCST (Fig. 1, patients 2 and 5), consistent with the early stages of Wallerian degeneration, in both patients.1 However, MRI changes were delayed in patient 5, being detected only at the 12-week MRI, compared with MRI day 1 post stroke in patient 2. Differences in the timing of Wallerian degeneration detection on MRI post stroke may be due to differences in the pathophysiology between ischaemic and haemorrhagic stroke. The other two patients with histologically evident Wallerian degeneration were patients 6 and 7. Both of these patients died years after the stroke. Their autopsy specimens showed significant axonal loss and gliosis in the DCST consistent with the chronic stages of Wallerian degeneration.1

Discussion

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Our results indicate that MRI (T1, T2, and/or DWI) signal changes, when present in areas representing axonal projections from infarcted brain tissue, are appropriately diagnosed as Wallerian degeneration. They do indeed correlate with pathological changes reflecting loss of neuronal processes. The absence of MRI changes in these areas was also correlated with intact axonal projections in histological specimens. To our knowledge, this is the first study to report this correlation in children. Moreover, we have shown that histopathological findings of Wallerian degeneration occurred in both ischaemic and haemorrhagic childhood stroke types.

Magnetic resonance imaging signal changes of DCST have been shown in the adult8,10,11 and developing brain5,6,12 post ischaemic injury. There is controversy about the early (up to the first 3 months after ischaemic insult) detection of Wallerian degeneration on MRI in the adult brain. Kuhn et al.2 reported the detection of Wallerian degeneration 4 weeks after the insult, whereas Inoue et al.13 could not detect early signal changes before 5 weeks after the ictus. Transient early Wallerian degeneration has also been reported in the adult brain using DWI.14,15 The discrepancies in these reports may be attributed to differences in MRI sequences such as proton density compared with T2-weighted MRI or DWI, and whether 0.5T or 1.5T was used in the study.10 MRI studies on the developing brain have shown that abnormal signal intensity on DWI may be evident within 1 week of a neonatal6 or childhood5 stroke. Out of our three patients with Wallerian degeneration, MRI changes were detected in two, in one case within the first week after the stroke (patient 2) and in the other in the first 12 weeks (patient 5). This abnormal signal intensity on MRI correlated with histopathological changes consistent with the early changes of Wallerian degeneration in both patients. These results histologically confirm for the first time that Wallerian degeneration can be detected by diffusion restriction as early as 1 to 3 weeks after a stroke in a child. DWI was more sensitive than T2 signal changes in depicting Wallerian degeneration in the first week after a stroke. It is important to note that patient 2 had a history of acute myeloid leukaemia and had previously been treated with chemotherapy, which may also contribute to vacuolation of myelin.

In our third patient (patient 7) with evidence of Wallerian degeneration, MRI was first carried out 5 years after the stroke. This is consistent with a previous report of chronic changes, in the form of increased signal intensity on T2-weighted MRI and cerebral peduncle atrophy, 6 months or more after the initial insult in the adult brain.11

Studies of the association of MRI signal abnormalities with histopathological changes consistent with Wallerian degeneration are limited to one study in the adult brain16 that included two patients, making our study the largest in the human brain and the first in the immature brain. In their study, Matsusue et al.16 showed that acute post-mortem T2-weighted MRI hyperintensity correlated histologically with vacuolation of myelin and mild loss of axons. These histological changes were seen in patient 2 from our cohort less than 1 month after the stroke and correlated with signal changes on DWI. However, the T2 hyperintense signal reported to occur beyond 14 weeks in adults2 occurred before 14 weeks in our cohort (patients 2 and 5), suggesting that the process of Wallerian degeneration may occur at a faster rate in children.

In our cohort, four out of seven children did not have MRI white matter signal abnormalities. This is consistent with previous reports that not all childhood arterial ischaemic stroke result in Wallerian degeneration.5,6 Lack of pathological changes correlated with the negative MRI studies in three of these four children. Absence of Wallerian degeneration on MRI suggests a more favourable motor outcome.5 Our results indicate that absence of axonal signal abnormalities on MRI indeed represents intact DCST and is a reliable predictor of a more favourable motor outcome.

In only one patient (patient 6), out of the seven children included in the study, was there no association between MRI and pathology findings. Serial MRI studies performed at all three stages showed no signs of Wallerian degeneration but had features of chronic Wallerian degeneration on histopathology. In this patient, T2-weighted MRI performed 6 and 16 weeks after the stroke did not show hypointense or hyperintense signal changes that would suggest Wallerian degeneration. DWI was performed more than 4 years after the stroke. It may have been possible to detect changes of Wallerian degeneration during the acute stages had DWI been performed at that time. The reason for the lack of sensitivity of T2-weighted MRI is unclear.

The MRI finding of DWI signal change has been shown to be a predictor of motor outcome in paediatric4–7 and, more recently, in adult8 stroke. Early evolution of MRI signal abnormalities in the DCST is associated with poor motor outcome in both patient populations. Therefore, correlation of this MRI finding with corresponding histopathological evidence of Wallerian degeneration is important for its utility as an accurate prognostic determinant to provide information to families, facilitate rehabilitation, and select patients for therapeutic interventions.

Study limitations

This case study was limited by the small sample size related to the infrequency of autopsy in cases of paediatric stroke. This was despite collecting data over a 10-year period from the stroke and pathology databases of a tertiary paediatric stroke referral centre. It is a retrospective study so the lack of standardized MRI protocols limits our ability to compare serial Wallerian degeneration MRI changes with existing adult studies.

Future research is needed to refine Wallerian degeneration changes on MRI in childhood stroke as a prognostic marker to select patients for therapeutic interventions. The correct selection of patients for stroke intervention is required so that the patients at greatest risk of a poor outcome receive the most appropriate treatment.

Acknowledgements

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The authors thank Lois Child and Susan Cromwell from the Division of Pathology for their assistance in data collecting; and Jeffrey Swaluk, Anne Marie Pontigon, and Elisa Wilson from the Pediatric Stroke Research team for their assistance with the Research Ethics Board approval.

References

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    Duchen LW. General pathology of neurons and neuroglia. In: Adams JH, Allen IV, Bruton CJ, editors. Greenfield’s Neuropathology, 5th edn. London: Edward Arnold, 1992: 1620.
  • 2
    Kuhn MJ, Mikulis J, Ayoub DM, Kosofsky BE, Davis KR, Traveras JM. Wallerian degeneration after cerebral infarction: evaluation with sequential MR imaging. Radiology1989; 172: 17982.
  • 3
    Sawlani V, Gupta RK, Singh MJ, Kohli A. MRI demonstration of Wallerian degeneration in various intracranial lesions and its clinical implications. J Neurol Sci1997; 146: 1038.
  • 4
    Mazumdar A, Mukherjee P, Miller JH, Malde H, McKinstry RC. Diffusion-weighted imaging of acute corticospinal tract injury preceding Wallerian degeneration in the maturing human brain. AJNR Am J Neuroradiol2003; 24: 105766.
  • 5
    Domi T, deVeber G, Shroff M, Kouzmitcheva E, MacGregor DL, Kirton A. Corticospinal tract pre-Wallerian degeneration: a novel outcome predictor for pediatric stroke on acute MRI. Stroke2009; 40: 7807.
  • 6
    Kirton A, Shroff M, Visvanathan T, deVeber G. Quantified corticospinal tract diffusion restriction predicts neonatal stroke outcome. Stroke2007; 38: 97480.
  • 7
    Groenendaal F, Benders MJ, De Vries LS. Pre-Wallerian degeneration in the neonatal brain following perinatal cerebral hypoxia–ischemia demonstrated with MRI. Semin Perinatol2006; 30: 14650.
  • 8
    DeVetten G, Coutts SB, Hill MD, et al.Acute corticospinal tract Wallerian degeneration is associated with stroke outcome. Stroke2010; 41: 75156.
  • 9
    Lama S, Qiao M, Kirton A, et al.Imaging corticospinal degeneration in neonatal rats with unilateral cerebral infarction. Exp Neurol2011; 228: 1929.
  • 10
    Orita T, Tsurutani T, Izumihara A, Kajiwara K. Early, evolving Wallerian degeneration of the pyramidal tract in cerebrovascular diseases: MR study. J Comput Assist Tomogr1994; 18: 94346.
  • 11
    Uchino A, Imada H, Ohno M. MR imaging of Wallerian degeneration in the human brain stem after ictus. Neuroradiology1990; 32: 1915.
  • 12
    De Vries LS, Van der Grond J, Van Haastert IC, Groenendaal F. Prediction of outcome in new-born infants with arterial ischaemic stroke using diffusion-weighted magnetic resonance imaging. Neuropediatrics2005; 36: 1220.
  • 13
    Inoue Y, Matsumura Y, Fukuda T, et al.MR imaging of Wallerian degeneration in the brainstem: temporal relationships. AJNR Am J Neuroradiol1990; 11: 897902.
  • 14
    Kang DW, Chu K, Yoon BW, Song IC, Chang KH, Roh JK. Diffusion-weighted imaging in Wallerian degeneration. J Neurol Sci2000; 178: 1679.
  • 15
    Uchino A, Sawada A, Takase Y, Egashira R, Kudo S. Transient detection of early Wallerian degeneration on diffusion-weighted MRI after an acute cerebrovascular accident. Neuroradiology2004; 46: 1838.
  • 16
    Matsusue E, Sugihara S, Fujii S, Kinoshita T, Ohama E, Ogawa T. Wallerian degeneration of the corticospinal tracts: postmortem MR–pathologic correlations. Acta Radiol2007; 48: 6904.