alexander disease: diagnosis with mr imagingings of alexander disease. conclusion: the deﬁned...

541

AJNR Am J Neuroradiol 22:541–552, March 2001

Alexander Disease: Diagnosis with MR Imaging

Marjo S. van der Knaap, Sakkubai Naidu, Steven N. Breiter, Susan Blaser, Hans Stroink, Stephan Springer,Jacobus C. Begeer, Rudy van Coster, Peter G. Barth, Neil H. Thomas, Jacob Valk, and James M. Powers

BACKGROUND AND PURPOSE: To date, the demonstration of Rosenthal fibers on brainbiopsy or autopsy specimens is considered a prerequisite for a definitive diagnosis of Alexanderdisease. We initiated a multiinstitutional survey of MR abnormalities in both presumed andconfirmed cases of Alexander disease to assess the possibility of an MR-based diagnosis.

METHODS: MR imaging studies in three patients with an autopsy-based diagnosis of Al-exander disease were analyzed to define MR criteria for the diagnosis. These criteria were thenapplied to 217 children with leukoencephalopathy of unknown origin.

RESULTS: Five MR imaging criteria were defined: extensive cerebral white matter changeswith frontal predominance, a periventricular rim with high signal on T1-weighted images andlow signal on T2-weighted images, abnormalities of basal ganglia and thalami, brain stemabnormalities, and contrast enhancement of particular gray and white matter structures. Fourof the five criteria had to be met for an MR imaging-based diagnosis. In a retrospective analysisof the MR studies of the 217 patients, 19 were found who fulfilled these criteria. No otheressentially new MR abnormalities were found in these patients. In four of the 19 patients,subsequent histologic confirmation was obtained. The clinical symptomatology was the samein the patients with and without histologic confirmation and correlated well with the MRabnormalities. MR abnormalities were in close agreement with the known histopathologic find-ings of Alexander disease.

CONCLUSION: The defined criteria are sufficient for an in vivo MR imaging diagnosis ofAlexander disease; only in atypical cases is a brain biopsy still necessary for a definitivediagnosis.

Alexander disease is a rare, nonfamilial leukoen-cephalopathy that typically presents with frontalpreponderance of white matter abnormalities andmacrencephaly. Since the first description of this

Received June 26, 2000; accepted after revision August 23.From the Departments of Child Neurology (M.S.v.d.K.) and

Radiology (J.V.), Free University Hospital, Amsterdam, theNetherlands; the Department of Neurogenetics, Kennedy Krie-ger Institute, Baltimore, MD (S.N.); the Department of Radi-ology, Johns Hopkins Medical Institute, Baltimore, MD(S.N.B.); the Department of Neuroradiology, Hospital for SickChildren, Toronto, Canada (S.B.); the Department of ChildNeurology, St. Elisabeth Hospital, Tilburg, the Netherlands(H.S.); the Department of Pediatrics, Technical University,Munich, Germany (S.S.); the Department of Child Neurology,University Hospital, Groningen, the Netherlands (J.C.B.); theDepartment of Child Neurology, University Hospital, Gent,Belgium (R.v.C.); the Department of Child Neurology, Aca-demic Medical Center, Amsterdam, the Netherlands (P.G.B.);the Department of Pediatric Neurology, Southampton GeneralHospital, Southampton, UK (N.H.T.); and the Departments ofPathology and Neurology, University of Rochester MedicalSchool, Rochester NY (J.M.P.).

Address reprint requests to M. S. van der Knaap, MD, De-partment of Child Neurology, Free University Hospital, P.O.Box 7057, 1007 MB Amsterdam, the Netherlands.

q American Society of Neuroradiology

disease by Alexander in 1949 (1), different clinicalsubtypes have been recognized. Most of the re-ported cases have been the infantile variant, withearly onset of macrocephaly and rapid neurologicdeterioration leading to early death (1–14). A neo-natal variant has been distinguished that is evenmore rapidly fatal (15). A less frequently reportedsubtype is the juvenile variant, in which macro-cephaly is a less consistent feature and onset ofneurologic deterioration occurs later in childhoodand is less rapid (16–21). Several instances of anadult variant have also been described (19, 22–24).

Laboratory investigations are not helpful in es-tablishing the diagnosis of Alexander disease. Thehallmark of the disease is the presence of Rosenthalfibers throughout the CNS. Rosenthal fibers are ab-normal intracytoplasmic proteinaceous inclusionsin fibrous astrocytes that have a distinctive hyalineappearance under the light microscope (1–3) and agranular electron-dense quality under the electronmicroscope (5, 19). Demonstration of Rosenthal fi-bers on histologic examination is considered a pre-requisite for a definitive diagnosis. There are ob-vious disadvantages associated with a diagnosisthat is limited to histologic confirmation. A brainbiopsy, which is an invasive procedure, is a diag-

AJNR: 22, March 2001542 VAN DER KNAAP

TABLE 1: Clinical signs and symptoms in patients with a histo-logically confirmed diagnosis

Infantile VariantJuvenileVariant

No. of patientsAge at first problemsDelayed motor developmentHighest motor milestoneDelayed mental developmentBehavioral problemsEpilepsyFeeding problemsVomitingDifficulty swallowing/chokingInsufficient gain in weightSpeech problemsMacrocephalyPoor eye contact

2Birth to 6 wkSevere (2)None (2)Severe (2)0Severe (2)2122Not applicable22

1,2 yModerateWalkingMild1Mild (1)1111100

Hypotonia Generalized (1)axial (1)

0

Hypertonia of the limbsHyperreflexiaCerebellar ataxiaExtrapyramidal signsDeteriorationAge at death

12Not testable0Rapid10 and 11½ mo,

respectively

1111Slow10 y

Note.—Numbers indicate the number of patients in whom a partic-ular feature was observed.

nostic necessity, especially in cases of slow pro-gression. The only other option is to wait for post-mortem examination.

MR imaging is known for its high sensitivity andspecificity in identifying white matter disorders(25, 26). We initiated a multiinstitutional survey ofMR imaging patterns in both presumed and con-firmed cases of Alexander disease to assess the pos-sibility of providing criteria for an MR imaging-based diagnosis.

MethodsMR studies of three patients with a histopathologically con-

firmed diagnosis of Alexander disease were available andformed the basis of the defined MR imaging criteria. Thesecriteria were then applied to 217 children with leukoencepha-lopathy of unknown origin whose MR studies we had receivedfor review. In these patients, well-known causes of childhoodleukoencephalopathy had been excluded by means of the fol-lowing laboratory examinations: urine organic acids, includingN-acetyl aspartate; plasma amino acids; serum lactate and py-ruvate; very long-chain fatty acids; and arylsulfatase A, gal-actocerebrosidase, b-galactosidase, and b-hexosaminidase ac-tivity. Patients who fulfilled the MR imaging criteria wereselected for further evaluation.

All MR studies were examined retrospectively using a listof many anatomic white and gray matter structures (27), whichwere scored as either normal or abnormal. In addition, othercharacteristics were scored (27), including swelling and atro-phy of white and gray matter structures, presence of cysts, andcontrast enhancement. The criterion for swelling of cerebralwhite matter was volume increase with broadening of the gyri.The criterion for swelling of the basal ganglia and thalamuswas volume increase with compression of the ventricles. Thecriterion for atrophy was volume decrease with widening ofthe ventricles and subarachnoid spaces. By comparing the im-ages of the patients, we determined the relative severity ofcharacteristics, such as swelling, atrophy, signal change, andcontrast enhancement, and distinguished two grades: 1) clearlypresent, and 2) slight to mild. In the analysis, we divided theMR studies into early and late, based on whether they wereobtained before or soon after the onset of neurologic deterio-ration or late in the course of the disease. All images werereviewed independently by two investigators, and consensuswas reached when there was a disagreement between theirinterpretations.

The medical histories and clinical findings were document-ed, including onset of clinical symptoms and disease course,presence of mental and motor problems, signs of bulbar dys-function, epilepsy, head circumference, height, and weight.

Results

Patients with a Histologically ConfirmedDiagnosis of Alexander Disease

Of the three patients with an autopsy-confirmeddiagnosis, two had the severe form of infantile Al-exander disease and one had the more protractedcourse of juvenile Alexander disease (Table 1). Inthe two patients with the infantile variant, the clin-ical symptomatology was dominated by failure ofnormal development, seizures, serious feedingproblems, macrocephaly, and rapid neurologic de-terioration. The patient with juvenile Alexanderdisease showed signs of mild developmental delay

in infancy. Onset of neurologic deterioration wasdelayed and gradual. Signs of bulbar dysfunctionbecame prominent, including speech problems,bouts of vomiting, and progressive swallowing dif-ficulties, eventually necessitating tube feeding.Spasticity and cerebellar ataxia arose late in thecourse of the disease, and the patient remainednormocephalic.

Two MR studies were obtained in the two pa-tients with infantile Alexander disease, in one pa-tient at an early stage and in the other patient at alate stage of the disease (Table 2). The early MRstudy (Fig 1) showed that the frontal white matterhad a slightly higher signal intensity than normalunmyelinated white matter on T2-weighted imagesand slightly lower signal intensity on T1-weightedimages. Other findings included signal abnormalityand some swelling of the basal ganglia, a periven-tricular rim of low signal intensity on T2-weightedimages and high signal intensity on T1-weightedimages, and areas of signal abnormality in the brainstem, including the medulla and the entire area ofthe midbrain, except for the red nuclei (Fig 1). Af-ter contrast administration, enhancement was foundin the ventricular lining, periventricular rim, partsof the frontal white matter, caudate nucleus, thala-mus, dentate nucleus, parts of the midbrain, fornix,and optic chiasm (Fig 1). The most striking find-ings on the late, as compared with the early, MRstudies included tissue loss, producing a thinnerperiventricular rim, cystic degeneration of the fron-tal white matter, and atrophy of the basal ganglia,

AJNR: 22, March 2001 ALEXANDER DISEASE 543

FIG 1. Early MR imaging study at the ageof 4 months in a patient with autopsy-proved infantile Alexander disease.

A–D, T2-weighted images show abnor-mally high signal in the medulla (A), thehilus of the dentate nucleus (arrows, A),the entire midbrain except for the red nu-clei (B), the basal ganglia, and the thala-mus (C). The frontal white matter has aslightly higher signal intensity than the oc-cipital white matter (C). The head of thecaudate nucleus is swollen (arrowheads,C). Around the ventricles, there is a rim oflow signal intensity (arrows, B–D).

E–G, T1-weighted images show highsignal intensity of the periventricular rim(arrows, E). After contrast administration,the T1-weighted images show enhance-ment of areas in the midbrain (F), ventric-ular lining (arrows, F), and periventricularrim (arrows, G).

thalamus, pons, and cerebellum. Areas of signal ab-normality were again seen in the brain stem, in-cluding the posterior part of the midbrain and thecentral part of the medulla. Distinctive findings on

this late MR study were large, apparently arach-noid, cysts in the sylvian fissure, in combinationwith a cystic enlargement of a cavum septi pellu-cidi and cavum vergae.


TABLE 2: MR findings in patients with a histologically confirmed diagnosis

Infantile Variant

Early MR Late MR

Juvenile Variant

Early MR Late MR

No. of patientsAge at MR imagingFrontal predominance wma

Extent of wmaSwelling wmaDegree of cystic degenerationDegree of signal changeWidth of periventricular rimContrast enhancement

Relative sparing of occipital wmRelative sparing of temporal wm

1 (1 MR study)4 mo1010101/1n.e.n.e.

1 (1 MR study)7 mo100100n.c.0n.e.

1 (1 MR study)20 mo1100100/111

1 (2 MR studies)6 and 9 y, respectively111110n.c.11

Periventricular rim of increased signal onT1W images, decreased signal on T2Wimages 1 1 1 1

Aspect of abnormal wmSwellingAtrophyCystic degeneration

Hydrocephalus

0001 (shunted)

1011

0000

1011

Involvement of deep nucleiCaudate nucleusPutamenGlobus pallidusThalamus

1111

1111

1111

1111

Aspect of basal ganglia abnormalitiesSwellingAtrophyIncreased signal on T2W images

101

011

101

010

Involvement of cerebellumCerebellar wmHilus dentate nucleusCerebellar atrophy

010

111

010

110

Brain stem lesionsMidbrainPonsMedulla

101

10 (atrophy)1

001

101

Thickened optic chiasmThickened fornixContrast enhancement

111

00n.c.

001

00n.c.

Periventricular rimVentricular lining onlyFrontal white matterCaudate nucleusPutamenGlobus pallidusThalamusDentate nucleusMidbrainMedullaOptic chiasmFornix

101100111011

011000000000

Note.—wm indicates white matter; wma, white matter abnormalities; n.e., not evaluable; n.c., no contrast material given; T1W, T1-weighted;T2W, T2-weighted. Numbers indicate the number of patients in whom a particular feature was observed.

In the patient with juvenile Alexander disease,three MR studies were available, one obtained earlyin the course of the disease and two at a late stage(Table 2). All MR images showed extensive cere-bral white matter changes with a frontal predomi-nance and relative sparing of occipital and temporalwhite matter (Fig 2). The early MR study (Fig 2A)

showed a periventricular rim with low signal inten-sity on T2-weighted images and high signal inten-sity on T1-weighted images. The basal ganglia andthalami were mildly abnormal in signal intensityand mildly swollen. Areas of signal abnormalitieswere seen in the posterior part of the medulla. Onthe two late MR studies, the distribution and extent


FIG 2. A–D, Early (A) and late (B–D) MRstudies of a patient with autopsy-confirmedjuvenile Alexander disease, obtained atages 20 months (A) and 9 years (B–D).The early T2-weighted image (A) showsextensive cerebral white matter abnormal-ities with partial sparing of the occipital re-gion. There is a thin periventricular rim oflow signal intensity (arrows, A). The basalganglia and thalamus have an increasedsignal intensity. The putamen and caudatenucleus are mildly swollen (A). On follow-up, the extent of the cerebral white matterabnormalities is more or less the same; theoccipital white matter is still partiallyspared (D). The basal nuclei are dark andatrophic on the T2-weighted images (D). Athin periventricular rim of low signal inten-sity is visible (arrows, D). The proton den-sity–weighted image (C) shows enormouscysts in the frontoparietal white matter, alarge cavum vergae, and enlarged lateralventricles. A lesion is seen in the posteriorpart of the medulla (B).

of the white matter abnormalities were essentiallythe same (Fig 2B–D). Progression of disease wascharacterized mainly by tissue loss with cystic de-generation of the frontoparietal white matter andenlargement of the lateral ventricles (Fig 2). Themildly elevated signal intensity on T2-weightedimages and swelling of the basal ganglia and thal-ami were replaced by low signal intensity and at-rophy (Fig 2).

On the basis of these MR findings we definedfive MR imaging criteria: 1) extensive cerebralwhite matter abnormalities with a frontal prepon-derance, either in the extent of the white matterabnormalities, the degree of swelling, the degree ofsignal change, or the degree of tissue loss (whitematter atrophy or cystic degeneration); 2) presenceof a periventricular rim of decreased signal inten-sity on T2-weighted images and elevated signal in-tensity on T1-weighted images; 3) abnormalities ofthe basal ganglia and thalami, either in the form ofelevated signal intensity and some swelling or ofatrophy and elevated or decreased signal intensityon T2-weighted images; 4) brain stem abnormali-ties, in particular involving the midbrain and me-dulla; and 5) contrast enhancement involving one

or more of the following structures: ventricular lin-ing, periventricular rim of tissue, white matter ofthe frontal lobes, optic chiasm, fornix, basal gan-glia, thalamus, dentate nucleus, and brain stemstructures.

We required that four of the five criteria be ful-filled for an MR imaging-based diagnosis to allowfor the facts that it may be difficult or impossibleto assess the presence and extent of white matterchanges in very young infants and that not all pa-tients may have had contrast-enhanced studies. It isimportant to realize that in young infants whitematter abnormalities may not yet be evident. Bothnormal, unmyelinated, and abnormal white matterhave high signal intensity on T2-weighted imagesand low signal intensity on T1-weighted images,making differentiation between normal and abnor-mal white matter difficult. Also, assessment of rel-ative sparing of occipital and temporal white matteris difficult or impossible in young infants, as oc-cipital U fibers only become myelinated in the sec-ond half of the first year of life and the temporalwhite matter becomes myelinated in the course ofthe second year. It may therefore not be possible toassess the presence of frontal predominance in ear-


TABLE 3: Clinical signs and symptoms in patients with an MRimaging-based diagnosis

Infantile VariantJuvenileVariant

No. of patients 5 14Age at first problems Birth to 4 mo ,2 yDelayed motor development Severe (5) Severe (2),

moderate(11), none(1)

Highest motor milestone None (3), grasping(2)

Sitting (2),walking(12)

Delayed mental development Severe (5) Mild-moder-ate (13)

Behavioral problemsEpilepsy

2Severe (5)

9Mild (11)

Feeding problemsVomitingDifficulty swallowing/chokingInsufficient gain in weight

5545

126115

Speech problems 2, not applicable in3

14

Macrocephaly 4, 98th percentilein 1

9, 98th per-centile in2, none in3

Poor eye contact 5 0Hypotonia Generalized (2),

axial (3)Generalized

(3)Hypertonia of the limbsHyperreflexia

35

814

Cerebellar ataxia Not testable in 3,no in 2

12

Extrapyramidal signs 3 3Deterioration Rapid (3), moder-

ate (2)Moderate (1),

slow (11),not to date(2)

Present age 2½ y in 1 3–20 y in 10Age at death 4 mo to 2½ y 7–18 y in 4Histologic confirmation 3 1

Note.—Numbers indicate the number of patients in whom a partic-ular feature was observed.

ly infancy. The criteria would allow an MR imag-ing-based diagnosis in young infants. It is not com-mon practice to administer contrast material topatients with leukoencephalopathy, and the criteriawould also allow an MR imaging-based diagnosisin the absence of contrast-enhanced studies.

Patients with an MR-Based Diagnosis ofAlexander Disease

Among the 217 patients with leukoencephalop-athy of unknown origin, 19 fulfilled the MR im-aging criteria. Five of them had a severe diseasecourse and 14 had a milder and more protracteddisease course (Table 3).

In the five patients with severe disease, thesymptomatology was dominated by failure of nor-

mal development, seizures, feeding problems, mac-rocephaly, and rapid neurologic deterioration. Abrain biopsy was performed in one patient and dis-closed Rosenthal fibers. Autopsy confirmation ofthe diagnosis of Alexander disease was obtained intwo other patients.

All 14 patients with mild disease showed somesigns of mild neurologic dysfunction in infancy,mainly developmental delay and epileptic seizures.Onset of clinical neurologic deterioration was de-layed and gradual in all patients. Signs of bulbardysfunction became prominent. Speech problemswere always present, starting with delayed speechdevelopment, progressing to dysarthria, and endingin loss of speech. Increasing problems with swal-lowing were frequent, often leading to insufficientgain or loss of weight, and culminating with tubefeeding. Bouts of vomiting, particularly duringmorning hours, occurred in 40% of the patients.Macrocephaly was not invariably present. Threechildren have remained normocephalic. Autopsyconfirmation of the diagnosis of Alexander diseasewas obtained in one of the 14 patients.

Thirty-seven MR imaging studies were availablein these 19 patients, eight in the five patients witha severe disease course and 29 in the 14 patientswith a mild disease course (Table 4). In the patientswith severe disease, five early MR studies and threelate MR studies were available. Early MR findings(Fig 3A and B) included a periventricular rim withlow signal intensity on T2-weighted images andhigh signal on T1-weighted images; signal changesand mild swelling of the basal ganglia and thalami,most pronounced in the head of the caudate nucleusand putamen; and areas of signal abnormality inthe brain stem, most frequently involving the mid-brain and medulla. One patient received contrastmaterial, which produced enhancement of the peri-ventricular rim, parts of the frontal white matter,parts of the basal ganglia, a midbrain lesion, theoptic chiasm, and fornix. In three patients, the ce-rebral white matter did not have an abnormal signalintensity on early MR studies considering its un-myelinated state for being unmyelinated. In one ofthese patients, the periventricular rim was moreprominent in the frontal area with greater contrastenhancement than in the parietooccipital area, butin the other two patients there was no frontal pre-dominance in any aspect. Both the latter patientsalso had late MR studies, which showed moreprominent signal abnormality of the cerebral whitematter, with an evident frontal predominance in thedegree of signal change, and swelling; in one, therewas also the beginning of cystic degeneration. Thelate MR studies showed signs of atrophy with en-largement of the lateral ventricles and atrophy ofthe basal ganglia and thalamus (Fig 3C and D).Contrast enhancement on the late MR studies in-volved the periventricular rim, the frontal whitematter, parts of the frontal cortex, the head of thecaudate nucleus, the putamen, and a midbrainlesion.


TABLE 4: MR findings in patients with an MR imaging-based diagnosis

Infantile Variant

Early MR Late MR

Juvenile Variant

Early MR Late MR

No. of patientsAge at MR imagingFrontal predominance wma

Extent of wmaSwelling wmaDegree of cystic degenerationDegree of signal changeWidth of periventricular rimContrast enhancement

Relative sparing of occipital wmRelative sparing of temporal wm

5 (5 MR studies)1–11 mo3110221/11, n.e. in 4n.e. in 5

3 (3 MR studies)3–20 mo3032301/20, n.e. in 20, n.e. in 2

10 (20 MR studies)1½–12 y9981600/298

7 (9 MR studies)7–15 y6545200/154

Periventricular rim of increased signal onT1W images, decreased signal on T2Wimages 5 3 10 7

Aspect of abnormal wmSwellingAtrophyCystic degeneration

Hydrocephalus

1001

3021

8010

6154

Involvement of deep nucleiCaudate nucleusPutamenGlobus pallidusThalamus

5554

3333

101095

7774

Aspect of basal ganglia abnormalitiesSwellingAtrophyIncreased signal on T2W images

505

023

80

10

074

Involvement of cerebellumCerebellar wmHilus dentate nucleusCerebellar atrophy

3230

3231

8781

6565

Brain stem lesionsMidbrainPonsMedulla

5513

3313

8647

743 (1atrophy in 2)7

Thickened optic chiasmThickened fornixContrast enhancement

221/1

112/2

002/2

001/1

Periventricular rimVentricular lining onlyFrontal white matterFrontal cortexCaudate nucleusPutamenGlobus pallidusDentate nucleusCerebellar cortexMidbrainMedullaOptic chiasmFornixTrigeminal nerve (intraparenchmal)

1/10/11/10/11/11/10/10/10/11/10/11/11/10/1

2/20/21/21/22/22/20/20/20/21/20/20/20/20/2

0/20/22/20/21/21/21/20/20/20/21/20/20/21/2

0/11/10/10/10/10/10/11/11/11/11/10/10/10/1

Note.—wm indicates white matter; wma, white matter abnormalities; n.e., not evaluable; n.c., no contrast given; T1W, T1-weighted; T2W, T2-weighted. Numbers indicate the number of patients in whom a particular feature was observed.

In the 14 patients with mild disease, 20 early MRstudies and nine late MR studies were available.Three patients had a prolonged follow-up of 4 to 8years with 12 MR examinations documenting ab-normalities from late infancy, before onset of neu-

rologic deterioration, up to an advanced stage of thedisease. The 20 early MR studies showed extensivecerebral white matter changes with frontal predom-inance and relative sparing of the occipital and tem-poral white matter in almost all patients (Fig 4). The


FIG 3. MR imaging of a patient with bi-opsy-confirmed infantile Alexander dis-ease.

A and B, At the age of 1½ months, thefrontal white matter has a slightly highersignal intensity on T2-weighted imagesand slightly lower signal intensity onunenhanced T1-weighted images thandoes the remainder of the cerebral whitematter, which has normal signal intensityfor unmyelinated white matter. There is aperiventricular rim of low signal intensity onT2-weighted images (arrows, A) and highsignal intensity on T1-weighted images(arrows, B), with some extensions into thefrontal white matter (arrowheads, A and B).The caudate nucleus and putamen havehigh signal intensity on T2-weighted im-ages and are mildly swollen.

C and D, At the age of 3 months, a ma-jor increase in ventricular size is seen withextreme thinning of the posterior cerebralmantle. The frontal white matter has amore abnormal signal intensity than theoccipital white matter, appears markedlyswollen, and shows early cystic degener-ation (arrows, D). There is a thin periven-tricular rim of low signal intensity on T2-weighted images (arrows, C). The basalganglia are now markedly atrophic. Aftercontrast administration, enhancement oc-curs in the ventricular lining, caudate nu-cleus, putamen, frontal white matter, andparts of the frontal cortex (D).

FIG 4. Early MR imaging studies in a patient with presumed juvenile Alexander disease, obtained at the age of 4 years.A and B, Extensive cerebral white matter abnormalities are seen on these T2-weighted images (B), with sparing of the occipital U

fibers (arrows, B). The signal abnormality is more pronounced in the frontal than in the occipital white matter. There is an irregularperiventricular rim of low signal intensity (arrowheads, B). The basal ganglia and thalamus have a mildly increased signal intensity.Within the posterior fossa, signal abnormalities are seen in the central part of the medulla, the hilus of the dentate nucleus, and thecerebellar hemispheric white matter, characteristically with the normal dentate nucleus prominently visible in between (A).

FIG 5. Contrast-enhanced MR image in a patient with presumed juvenile Alexander disease, obtained at the age of 12 years. Noteenhancement of the intraparenchymal trajectory of the fifth cranial nerve on both sides.


basal ganglia were involved in all cases, invariablyaffecting the head of the caudate nucleus and puta-men, often also affecting to a lesser extent the glo-bus pallidus and thalamus. This was always char-acterized by a mild signal change and in most casesalso by mild swelling. Brain stem lesions were seenin eight of the 10 patients. Most frequently affectedwere the midbrain (in the anterior part, the periaq-ueductal region, or the entire area except for the rednuclei and colliculi) and the medulla (either in theposterior or central part). The pontine tegmentumwas affected in some patients. Two patients receivedcontrast material, which produced subtle enhance-ment in one of them and more prominent enhance-ment in the other. The enhancing areas included avariable combination of foci in the frontal whitematter, caudate nucleus, putamen, globus pallidus,brain stem lesions, and intraparenchymal trajectoriesof the trigeminal nerve (Fig 5). Note that the com-plete picture was already present on early MR stud-ies obtained during the stage of minimal neurologicdysfunction, sometimes years before the onset ofneurologic deterioration. On the late MR studies, theextent of the white matter abnormalities was moreor less the same as on the early studies (Fig 6). Themain changes in comparison to the early MR ex-aminations consisted of signs of tissue loss with cys-tic degeneration of the frontal white matter, enlarge-ment of the lateral ventricles, cerebellar atrophy, andbrain stem atrophy. The cysts in the frontal whitematter became very large in one patient, and wereless impressive in four patients. The slightly elevat-ed signal intensity on T2-weighted images and theswelling of the basal ganglia and thalami were grad-ually replaced by low signal intensity and atrophy(Fig 6). In end-stage disease, the basal ganglia andthalami were reduced to thin rims. One patient re-ceived contrast material, which produced enhance-ment in the ventricular lining, multiple areas of thebrain stem, the dentate nucleus, and the cerebellarcortex (Fig 6).

Discussion

MR Imaging PatternWe started our investigation with MR imaging

studies of three patients in whom the diagnosis ofAlexander disease had already been proved histo-logically. On the basis of these findings, five MRimaging criteria were defined that allowed the iden-tification of another 19 patients from a cohort of217 children with leukoencephalopathy of un-known origin. The MR imaging pattern observedin this group of 19 patients was very homogeneousand identical to the pattern observed in the firstthree patients. After a preliminary diagnosis of Al-exander disease was made on the basis of typicalMR findings, the diagnosis was subsequently con-firmed histologically in four of the 19 patients.

The MR imaging pattern is quite specific, dis-similar from patterns observed in other white mat-

ter disorders, either with known (25, 26) or un-known (27) origins. In a study of 92 patients withleukoencephalopathy of unknown origin (27), 11had the above MR imaging picture. Statistical anal-ysis showed that the pattern was significantly dif-ferent from the MR imaging patterns observed inthe rest of the patients (27).

Several white matter disorders with known path-ogeneses share some of these MR imaging char-acteristics, but none shares all of them. A predom-inant involvement of the frontal lobes together withinvolvement of diencephalic nuclei and brain stemtracts, as well as contrast enhancement may be ob-served in X-linked adrenoleukodystrophy (26).However, in this disorder, only or mainly the lateralgeniculate bodies are involved among the dience-phalic nuclei. The brain stem lesions primarily in-volve the corticospinal, corticobulbar, visual, andauditory tracts (28, 29). Contrast enhancement oc-curs within the outer border of the white matterlesion (26). Similarly, some patients with meta-chromatic leukodystrophy have predominantlyfrontal white matter abnormalities together with in-volvement of the brain stem (26). The brain stemlesions involve the long tracts. Contrast enhance-ment is not a feature of this disease. Canavan dis-ease is characterized by a combination of macro-encephaly, extensive cerebral white matter changes(without frontal preponderance), and basal gangliaabnormalities (26). However, the thalamus and glo-bus pallidus are involved with typical sparing ofthe putamen and caudate nucleus (26). Contrast en-hancement does not occur. In merosin-deficientcongenital muscular dystrophy, extensive cerebralwhite matter changes are present with relative spar-ing of the occipital white matter (30). However, thebasal ganglia and brain stem are spared (30). Invacuolating leukoencephalopathy with subcorticalcysts, extensive cerebral white matter changes areobserved with slight swelling (31). However, thereare invariably anterotemporal cysts and often sub-cortical cysts in the frontoparietal border zone (31).In contrast, the cysts in Alexander disease affectprimarily the deep frontal white matter. Thus, allthe above disorders can be distinguished from Al-exander disease on the basis of MR imagingcriteria.

MR findings have been published in several cas-es of infantile Alexander disease (7–9, 11, 14, 32,33) and in a few cases of juvenile Alexander dis-ease (21, 34) with histologic confirmation. Apartfrom a study by Gingold et al (32), none of thereports describes in detail abnormalities other thanthose involving the white matter, but the publishedimages show abnormalities that are similar to thosefound in our patients. Images showing brain stemabnormalities and contrast-enhanced studies havebeen limited to the report of Gingold et al (32).These authors described signal abnormalities in theperiaqueductal gray matter and contrast enhance-ment in the periventricular white matter, deep fron-


FIG 6. Late MR imaging study of a patientwith presumed juvenile Alexander disease,obtained at the age of 10 years.

A and B, There is extensive white matterinvolvement with frontal preponderance(A). The basal ganglia are dark and atro-phic on T2-weighted images (A). A thinperiventricular rim of low signal intensity isjust visible (arrows, A). After contrast ad-ministration, enhancement of the entirecerebellar surface and dentate nucleus isseen (B).

tal white matter, basal ganglia, and brain stemareas.

Characteristic MR Findings as Related toHistopathologic Findings

None of our patients had MR examinationsshortly before death. Therefore, we are not able toprovide a direct comparison between MR and his-topathologic findings in any of our patients. How-ever, autopsy findings have been reported in manycases of Alexander disease. In the infantile variant,the brain is typically too large and too heavy. Ex-tensive paucity of myelin is observed in the cere-bral hemispheres, with the frontal lobes affectedmore severely than the occipital and temporal lobes(3, 13). In the noncystic areas, axons are intact (3,10). Myelin paucity can be observed to a variableextent in the cerebellar white matter, the hilus ofthe dentate nucleus, and areas of the brain stem.There are few signs of active myelin breakdown:phagocytes are not increased in number and, inmost cases, no sudanophilic breakdown products orinflammatory cells are seen. This is more consistentwith a failure of myelin formation than with myelinbreakdown (3, 13). A striking increase in astrocytesfilled with Rosenthal fibers is seen throughout thebrain, with a preferential subpial, perivascular, andsubependymal pattern. The greatest density of Ro-senthal fibers is seen in the outer, subpial layers ofthe cerebral cortex (3, 10), frontal white matter (3,6), periventricular region (3, 35), basal ganglia andthalami (2, 3, 13), and brain stem (3, 6, 13). As-trocytes containing Rosenthal fibers may surroundand disrupt the ependyma of the aqueduct, resultingin narrowing of its lumen and obstructive hydro-cephalus (6). Deposition of Rosenthal fibers in thecerebellum is variable, being sparse in some cases(2) and, in others, present to a great extent in thecerebellar white matter (3), dentate nucleus (3), or,rarely, the subpial layers of the cerebellar cortex(13). The fornix (3) and optic nerves, chiasm, and

tracts (5) may contain many Rosenthal fibers. Theperipheral or schwannian parts of the other cranialnerves are always free of Rosenthal fibers, whereasthe intraparenchymal root bundles may containheavy depositions (3, 5). Similar autopsy findingsin patients with extensive white matter disease andRosenthal fiber deposition have been described inseveral juvenile (5, 19–21) and adult (36) cases.Incidentally, a greatly expanded cavum septi pel-lucidi and cavum vergae are seen bulging into thelateral ventricles (1, 2). The autopsy findings in ourpatients were in agreement with the abovedescription.

These autopsy reports correspond to the MR im-aging findings in all our patients, including those withhistologically confirmed and MR imaging-based di-agnoses. Considering the histologic observations, afailure of normal myelination is most likely to beresponsible for at least part of the abnormal whitematter signal on MR images. This also explains thepresence of extensive leukoencephalopathy on earlyMR imaging studies in juvenile disease. Another ob-servation is that in the juvenile patients, clinical de-terioration was not accompanied by an increase inthe extent of the white matter changes on MR im-ages, as would have been expected in progressivemyelin loss. However, the affected white matter hashigher signal on T2-weighted images and lower sig-nal on T1-weighted images than does normal, un-myelinated, white matter. Therefore, the signal ab-normality of the white matter cannot be explained byhypomyelination alone. Hyperplasia and hypertrophyof astrocytes and Rosenthal fiber deposition are prob-ably contributing factors. The periventricular rim oflow signal intensity on T2-weighted images and highsignal intensity on T1-weighted images is probablydue to an extreme density of Rosenthal fibers in theperiventricular white matter. Numerous Rosenthal fi-bers are probably responsible for the signal abnor-mality and swelling of the basal ganglia and thala-mus. It remains unclear why the periventricular rimhas a different signal behavior. Brain stem abnor-


malities are a consistent finding on both MR imagesand histopathologic sections. The areas that showcontrast enhancement are those with the highest den-sity of Rosenthal fibers. The same observation wasmade by Farrell et al (6), who attributed the contrastenhancement to insufficiency of the blood-brain bar-rier due to impaired function of astrocytic footplates,the main location of Rosenthal fibers.

The Clinical Picture as Related to MR andAutopsy Findings

The clinical symptomatology was the same inpatients with and without histologic confirmationof Alexander disease. The symptoms were alsosimilar to those described previously in patientswith a histologically confirmed diagnosis (1–3, 5,6, 13, 16, 35).

Among our patients, the clinical severity of dis-ease was highly variable. The distinction betweensevere and mild cases was not always clear. Someof the severely affected patients had a slower pro-gression of disease than others, and some of themildly affected patients had a more rapid diseasecourse than others. The clinical symptoms werequalitatively the same in both severe and mild cas-es. Macrocephaly was present in all severely af-fected patients and in most mildly affected patients;it can be ascribed to the macroencephaly andgeneralized white matter swelling as depicted byMR imaging, with a component of obstructive hy-drocephalus in some patients. The white matterswelling has usually been attributed to the accu-mulation of Rosenthal fibers, but astrocytic hyper-plasia is a contributing factor (35). A few mildlyaffected patients remained normocephalic; they hadless severe white matter swelling and no ventricularenlargement on MR images. Early in the course ofthe disease, our patients, in particular the ones withlate-onset disease, displayed only a few signs ofwhite matter abnormalities. Spasticity and ataxiawere usually late findings, occurring in the stage ofcystic degeneration and atrophy of the white matter.Signs of brain stem dysfunction were often rela-tively early findings. They are in agreement withthe signal abnormalities in that area, shown by MRimaging from an early stage onward. In severalchildren, signs of extrapyramidal dysfunction werenoted, although they were not prominent. Thesemay be ascribed to abnormalities of the basal nucleias evidenced by MR imaging and autopsy findings.Epileptic seizures were also consistent findings ofthe disease, pointing to cortical dysfunction. Thecerebral cortex is known to contain numerous Ro-senthal fibers in its outer layers, and in one casewe found cortical enhancement on postcontrast MRimages. Mental retardation and behavioral prob-lems may be related to the frontal white matter dis-ease, cortical abnormalities, or a combination ofthese.

Conclusion

We have defined MR imaging criteria on the ba-sis of findings in patients with a histologically con-firmed diagnosis of Alexander disease. We haveshown that application of these criteria to patientswith leukoencephalopathy of unknown origin leadsto the identification of similar cases. We have dem-onstrated the accuracy of the criteria by subse-quently providing histologic confirmation in sev-eral of the patients with an MR imaging-baseddiagnosis. The specificity of these MR imagingcharacteristics had already been demonstrated in aprevious study of a large number of patients withleukoencephalopathy of unknown origin (27). Wehave shown that the clinical symptoms of the pa-tients with a histologically based diagnosis andwith an MR imaging-based diagnosis are the sameand in agreement with MR imaging abnormalitiesand histopathologic findings. The only new findingamong the 19 patients identified with the previous-ly defined criteria concerned the areas of contrastenhancement. Apart from the structures mentionedin the inclusion criteria, contrast enhancement wasobserved in the cerebral and cerebellar cortex andintraparenchymal parts of cranial nerves.

We concentrated on patients with typical MRfindings of Alexander disease. It is possible or evenprobable that we did not recognize atypical casesof Alexander disease, because they did not fulfillthe inclusion criteria. Both juvenile (17–19) andadult (19, 22–24) patients have been seen in whomRosenthal fibers were evident at autopsy despiteminimal or absent white matter changes. MR im-aging also has shown minimal or no white matterabnormalities in such cases (37, 38). In addition,possible variants of Alexander disease have beendescribed with predominantly cerebellar involve-ment (39). It is difficult to decide whether theseatypical cases are in fact Alexander disease or not(40). The problem is that Rosenthal fibers are notrestricted to Alexander disease but may occur inseveral other chronic conditions, including multiplesclerosis (19), tumors (41), and drug toxicity (42).At present, there is no biochemical marker for Al-exander disease. It is known that major componentsof Rosenthal fibers are stress proteins (ie, aB-crys-tallin and HSP27) (43, 44). HSP27 and aB-crys-tallin have been reported to be elevated in the CSF(45); however, the sensitivity and specificity of thistest is not known and as such its diagnostic contri-bution has not been assessed. With the present stateof knowledge, it is not possible to determinewhether patients with extensive, preponderantlyfrontal cerebral leukoencephalopathy and Rosen-thal fibers on histopathologic examination and pa-tients with Rosenthal fibers but without extensivecerebral leukoencephalopathy have different dis-eases or variants of the same disease.

Our final conclusion is that if a patient fulfillsthe MR imaging criteria set forth in this study, itis justified to presume a diagnosis of Alexander dis-


ease without histologic confirmation, and that onlyin atypical cases of the disease is a brain biopsystill necessary for a definitive diagnosis.

AcknowledgmentsWe thank all the physicians who referred patients and all

the neuropathologists who provided histologic confirmation fortheir collaboration.

References1. Alexander WS. Progressive fibrinoid degeneration of fibrillary

astrocytes associated with mental retardation in a hydroce-phalic child. Brain 1949;72:373–381

2. Crome L. Megalencephaly associated with hyaline neuropathy.Brain 1953;76:215–228

3. Wohlwill FJ, Bernstein J, Yakovlev PI. Dysmyelogenic leuko-dystrophy. J Neuropathol Exp Neurol 1959;18:359–383

4. Holland IM, Kendall BE. Computed tomography in Alexan-der’s disease. Neuroradiology 1980;20:103–106

5. Towfighi J, Young R, Sassani J, Ramer J, Horoupian DS. Alex-ander’s disease: further light- and electron-microscopic obser-vations. Acta Neuropathol 1983;61:36–42

6. Farrell K, Chuang S, Becker LE. Computed tomography in Al-exander’s disease. Ann Neurol 1984;15:605–607

7. Hess DC, Fischer AQ, Yaghmai F, Figuera R, Akamatsu Y. Com-parative neuroimaging with pathological correlates in Alex-ander’s disease. J Child Neurol 1990;5:248–252

8. Bobele GB, Garnica A, Schaefer GB, et al. Neuroimaging find-ings in Alexander’s disease. J Child Neurol 1990;5:253–258

9. Schuster V, Horwitz AE, Kreth HW. Alexander’s disease: cranialMRI and ultrasound findings. Pediatr Radiol 1991;21:133–134

10. Neal JW, Cave EM, Singhrao SK, Cole G, Wallace SJ. Alexan-der’s disease in infancy and childhood: a report of two cases.Acta Neuropathol 1992;84:322–327

11. Garcia L, Gascon G, Ozand P, Yaish H. Increased intracranialpressure in Alexander disease: a rare presentation of whitematter disease. J Child Neurol 1992;7:168–171

12. Pridmore CL, Baraitser M, Harding B, Boyd SG, Kendall BE,Brett EM. Alexander’s disease: clues to diagnosis. J ChildNeurol 1993;8:134–144

13. Klein EA, Anzil AP. Prominent white matter cavitation in aninfant with Alexander’s disease. Clin Neuropathol 1994;13:31–38

14. Staba MJ, Goldman S, Johnson FL, Huttenlocher PR. Allogeneicbone marrow transplantation for Alexander’s disease. BoneMarrow Transplant 1997;20:247–249

15. Springer S, Kraegeloh-Mann I, Naegele T, et al. Alexander dis-ease: classification revisited and isolation of a neonatal form.Neuropediatrics 2000;31:86–92

16. Vogel FS, Hallervorden J. Leukodystrophy with diffuse Rosen-thal fiber formation. Acta Neuropathol 1962;2:126–143

17. Russo LS, Aron A, Anderson PJ. Alexander’s disease: a reportand reappraisal. Neurology 1976;26:607–614

18. Goebel HH, Bode G, Caesar R, Kohlschutter A. Bulbar palsywith Rosenthal fiber formation in the medulla of a 15-year-old girl. Neuropediatrics 1981;12:382–391

19. Herndon RM, Rubinstein LJ, Freeman JM, Mathieson G. Lightand electron microscopic observations on Rosenthal fibers inAlexander disease and in multiple sclerosis. J Neuropathol ExpNeurol 1970;29:524–551

20. Cole G, De Villiers F, Proctor NSF, Freiman I, Bill B. Alexander’sdisease: case report including histopathological and electronmicroscopic features. J Neurol Neurosurg Psychiatry 1979;42:619–624

21. Reichard EAP, Ball WS, Bove KE. Alexander disease: a casereport and a review of the literature. Pediatr Pathol Lab Med1996;16:327–343

22. Mastri AR, Sung JH. Diffuse Rosenthal fiber formation in theadult: report of four cases. J Neuropathol Exp Neurol 1973;32:424–436

23. Spalke G, Mennel HD. Alexander’s disease in an adult: clini-copathologic study of a case and review of the literature. ClinNeuropathol 1982;1:106–112

24. Schwankhaus JD, Parisi JE, Gulledge WR, Chin L, Currier RD.Hereditary adult-onset Alexander’s disease with palatal my-oclonus, spastic paraparesis, and cerebellar ataxia. Neurology1995;45:2266–2271

25. Van der Knaap MS, Valk J, de Neeling N, Nauta JJP. Patternrecognition in magnetic resonance imaging of white matterdisorders in children and young adults. Neuroradiology 1991;33:478–493

26. Van der Knaap MS, Valk J. Magnetic Resonance of Myelin, My-elination and Myelin Disorders. Heidelberg: Springer;1995

27. Van der Knaap MS, Breiter SN, Naidu S, Hart AAM, Valk J.Defining and categorizing leukoencephalopathies of unknownorigin: MR imaging approach. Radiology 1999;213:121–133

28. Kumar AJ, Rosenbaum AE, Naidu S, et al. Adrenoleukodystro-phy: correlating MR imaging with CT. Radiology 1987;165:497–504

29. Barkovich AJ, Ferriero DM, Bass N, Boyer R. Involvement ofthe pontomedullary corticospinal tracts: a useful finding in thediagnosis of X-linked adrenoleukodystrophy. AJNR Am J Neu-roradiol 1997;18:95–100

30. Van der Knaap MS, Smit LME, Barth PG, et al. Magnetic reso-nance imaging in classification of congenital muscular dystro-phies with brain abnormalities. Ann Neurol 1997;42:50–59

31. Van der Knaap MS, Barth PG, Stroink H, et al. Leukoencepha-lopathy with swelling and a discrepantly mild clinical coursein eight children. Ann Neurol 1995;37:324–334

32. Gingold MK, Bodensteiner JB, Schochet SS, Jaynes M. Alexander’sdisease: unique presentation. J Child Neurol 1999;14:325–329

33. Madsen JR, Partington MD, Hay TC, Tyson RW. A 2-month-oldfemale infant with progressive macrocephaly and irritability.Pediatr Neurosurg 1999;30:157–163

34. Deprez M, D’Hooge M, Misson JP, et al. Infantile and juvenilepresentations of Alexander’s disease: a report of two cases.Acta Neurol Scand 1999;99:158–165

35. Borrett D, Becker LE. Alexander’s disease: a disease of astro-cytes. Brain 1985;108:367–385

36. Walls, TJ, Jones RA, Cartlidge NEF, Saunders M. Alexander’sdisease with Rosenthal fiber formation in an adult. J NeurolNeurosurg Psychiatry 1984;47:399–403

37. Duckett S, Schwartzman RJ, Osterholm J, Rorke LB, FriedmanD, McLellan TL. Biopsy diagnosis of familial Alexander’s dis-ease. Pediatr Neurosurg 1992;18:134–138

38. Martidis A, Yee RD, Azzarelli B, Biller J. Neuro-ophthalmic,radiologic, and pathologic manifestations of adult-onset Alex-ander disease. Arch Ophthalmol 1999;117:265–267

39. Torreman M, Smit LME, van der Valk P, Valk J, Scheltens P. Acase of macrocephaly, hydrocephalus, megacerebellum, whitematter abnormalities and Rosenthal fibers. Dev Med ChildNeurol 1993;35:732–736

40. Herndon RM. Is Alexander’s disease a nosological entity or acommon pathologic pattern of diverse etiology? J Child Neurol1999;14:275–276

41. Gluszez A. Disseminated cerebral gliomatosis with fibrillarydegeneration of the glia with Rosenthal fibers. Acta Neuropath-ol 1964;4:212–217

42. Kress Y, Gaskin F, Horoupian DS, Brosnan C. Nickel inductionof Rosenthal fibers in rat brain. Brain Res 1981;210:419–425

43. Iwaki T, Iwaki A, Tateishi J, Sakaki Y, Goldman JE. aB-crystallinand 27-kd heat shock protein are regulated by stress condi-tions in the central nervous system and accumulate in Rosen-thal fibers. Am J Pathol 1993;143:487–495

44. Head MW, Corbin E, Goldman JE. Overexpression and abnor-mal modification of the stress proteins aB-crystallin andHSP27 in Alexander disease. Am J Pathol 1993;143:1743–1753

45. Takanashi J, Sugita K, Tanabe Y, Niimi H. Adolescent case ofAlexander disease: MR imaging and MR spectroscopy. PediatrNeurol 1998;18:67–70

alexander disease: diagnosis with mr imagingings of alexander disease. conclusion: the deﬁned...

Documents