Pediatric Radiology

, 38:1054

Methylmalonic acidemia: brain imaging findings in 52 children and a review of the literature

Authors

    • Department of RadiologyBrigham and Women’s Hospital, Harvard Medical School
  • Talieh Zaman
    • Department of Pediatric Metabolic DisordersTehran University of Medical Sciences
  • Hossein Ghanaati
    • Department of RadiologyTehran University of Medical Sciences
  • Sanaz Molaei
    • Department of RadiologyShahid Beheshti University of Medical Sciences
  • Richard L. Robertson
    • Department of RadiologyChildren’s Hospital Boston
  • Amir A. Zamani
    • Department of Radiology, Brigham and Women’s HospitalHarvard Medical School
Original Article

DOI: 10.1007/s00247-008-0940-8

Cite this article as:
Radmanesh, A., Zaman, T., Ghanaati, H. et al. Pediatr Radiol (2008) 38: 1054. doi:10.1007/s00247-008-0940-8

Abstract

Background

Methylmalonic acidemia (MMA) is an autosomal-recessive inborn error of metabolism.

Objective

To recognize the CT and MR brain sectional imaging findings in children with MMA.

Materials and methods

Brain imaging studies (47 MR and 5 CT studies) from 52 children were reviewed and reported by a neuroradiologist. The clinical data were collected for each patient.

Results

The most common findings were ventricular dilation (17 studies), cortical atrophy (15), periventricular white matter abnormality (12), thinning of the corpus callosum (8), subcortical white matter abnormality (6), cerebellar atrophy (4), basal ganglionic calcification (3), and myelination delay (3). The brain images in 14 patients were normal.

Conclusion

Radiological findings of MMA are nonspecific. A constellation of common clinical and radiological findings should raise the suspicion of MMA.

Keywords

Methylmalonic acidInborn error of metabolismCTMRIChildren

Introduction

Methylmalonic acid is generated during metabolism of certain amino acids (isoleucine, methionine, threonine, and valine) and odd-chain fatty acids. Methylmalonyl coenzyme-A (CoA) is changed into succinyl-CoA by the action of an adenosylcobalamin-dependent enzyme, namely methylmalonyl-CoA mutase. Defects in methylmalonyl-CoA mutase or its coenzyme, cobalamin (vitamin B12), lead to the accumulation of methylmalonic acid and a clinical picture of methylmalonic acidemia (MMA), also called methylmalonic aciduria [1].

A neonatal screening program in Massachusetts estimates the incidence of MMA to be 1 in 48,000 [2]; if less-severe cases are included, the incidence might be as high as 1 in 25,000 [3]. In a study describing 326 patients with different inherited metabolic diseases, MMA was diagnosed in 31 patients [4]. In light of the fact that MMA inheritance is autosomal recessive, it is not surprising to see a higher prevalence in the Middle East, where consanguineous marriage is more common. Because of lack of a neonatal screening program in this region, the prevalence of the disorder in this region is unknown.

Because MMA usually presents clinically with nonspecific symptoms such as seizure, psychomotor retardation, poor feeding, respiratory distress, loss of consciousness, and muscle tone abnormality, patients often undergo extensive work-up before the correct diagnosis is made. Because neurological symptoms are common in patients with MMA, brain imaging is often employed to rule out congenital or acquired central nervous system abnormalities. Recognition of the imaging features of the disorder might help to establish an appropriate diagnosis earlier in the course of the disease when treatment is more effective [5, 6].

Materials and methods

We enrolled 70 children with an established diagnosis of MMA at the Children’s Medical Center, Tehran, Iran, from September 2004 to September 2005. In some children the MMA was newly diagnosed; others had received treatment for variable durations. The standard protocol for the diagnosis of MMA was increased plasma methylmalonyl carnitine or increased urine methylmalonic acid detected by gas chromatography-mass spectrometry and the propionate incorporation test. Standard drug therapy for MMA in this center is oral administration of 100 mg of l-carnitine per kilogram of body weight with 10 mg of vitamin B12 per day and a protein-restricted diet of 1–1.5 g/kg per day. Informed consent was obtained from parents, and all stages of the study were approved by the medical ethics committee of the hospital. The presence of at least one other concurrent metabolic disorder led to the exclusion of 18 children with disorders such as propionic acidemia, phenylketonuria, or galactosemia.

As shown in Table 1, 52 imaging examinations from 52 children were studied comprising 37 (71.2%) boys and 15 (28.8%) girls. Their ages at the time of presentation ranged from 0 to 42 months (mean 8.8 months, SD 12.4 months). The children’s ages at the time of imaging ranged from 1 month to 11 years (mean 42.8 months, SD 32.1 months). Of the 52 children, 19 had not received any treatment before the imaging, and 33 had received treatment for a mean duration of 18.9 months (range 3 to 48 months, SD 10.6 months).
Table 1

Demographic data and imaging findings in 52 children with MMA

Patient

Sex

Age at presentation (months)

First presenting symptom

Age at treatment onset (months)

Age at the time of imaging (months)

Myelination abnormality

Cortical atrophy (sulcal widening)

Periventricular white matter change

Internal capsule change

Ventricular dilation

Basal ganglionic calcification

Cerebellar atrophy

Corpus callosum thinning

Subcortical white matter change

Signal change consistent with focal infarct

Bright signal in centrum semiovale

Brainstem thinning

Other findings

1

M

1

Seizure

48

3

+

Moderate

+

 

2

M

24

Developmental regression

30

18

Mild

+

Mild

Prominent basal cistern; Dandy-Walker variant

3

M

0

Hypotonia

24

42

 

4

F

42

Seizure

120

132

+

Partial collapse of lateral ventricle

5

M

6

Seizure; microcephalus

60

60

Generalized pachygyria; left maxillary sinusitis

6

M

0

Seizure; psychomotor delay

84

108

+

+

 

7

M

1

Recurrent vomiting; respiratory distress

24

48

+

Mild

+

Diffuse

Cavum septum pellucidum

8

M

36

Seizure

48

72

Severe

Mild

 

9

F

0

Seizure

102

96

 

10

M

1

Agitation

6

24

 

11

M

6

Psychomotor delay

12

30

Moderate

+

 

12

M

1

Screening

1

24

 

13

F

30

Seizure; psychomotor delay

180

132

 

14

F

0

Psychomotor delay

0

24

Mild

Moderate

Cavum septum pellucidum

15

M

1

Screening

2

36

Mild

Sinusitis

16

M

12

Seizure; hypotonia

48

12

Mild

Moderate

Occipital interior table erosion

17

F

0

Psychomotor delay

24

36

+

 

18

M

6

Coma

66

84

 

19

F

1

Screening

1

48

Maxillary and posterior ethmoid sinusitis

20

M

0

Respiratory distress

18

24

Moderate

Arnold-Chiari I malformation; obliteration of the cisterna magna; arachnoid cyst in the left posterior fossa

21

M

0

Seizure; respiratory distress

18

6

Moderate

+

Moderate

+

Arachnoid cyst

22

M

0

Lethargy

36

60

 

23

F

12

Skeletal abnormality

24

48

Mild

Arnold-Chiari I malformation

24

M

6

Seizure

114

72

Moderate

+

Frontal

 

25

M

36

Recurrent vomiting

42

60

 

26

M

6

Psychomotor delay

24

12

+

Mild

Moderate

+

Abnormal left medial temporal lobe

27

M

24

Psychomotor delay

66

36

Severe

Dilated cisterns

28

M

0

Seizure

0

48

+

Widening of extracerebral spaces in frontal regions

29

M

24

Communication defect

72

72

Mild

 

30

M

2

Growth retardation

30

42

+

Frontal and parietal

Sinusitis

31

F

3

Psychomotor delay

30

36

+

 

32

M

0

Poor feeding

48

12

Moderate

Moderate

Occipital and parietal

 

33

F

3

Seizure

6

12

Mild

+

+

+

 

34

M

6

Psychomotor delay; dystonia

12

36

 

35

M

1

Poor feeding

48

72

Mild

Mild

 

36

M

30

Seizure

54

48

+

+

+

Occipital

Right mastoiditis; sinusitis

37

M

0

Screening

3

24

 

38

M

6

Psychomotor delay

6

6

Sinusitis

39

F

3

Psychomotor delay; growth retardation

3

6

+

+

Benign infantile subdural effusion on the frontal lobe; arachnoid cyst in left temporal fossa; dilated cisterna magna

40

M

1

Poor feeding

18

24

Mild

+

Occipital and left parietal

occipital

Parietal encephalomalacia; arachnoid cyst in middle fossa; dilation of the occipital horn and both atria

41

F

9

Growth retardation; lethargy

78

96

 

42

M

1

Poor feeding

24

18

 

43

M

6

Psychomotor delay

36

1

Moderate

+

 

44

M

0

Respiratory distress

30

48

+

+

Left parietal

 

45

F

6

Seizure; psychomotor delay

18

12

 

46

M

12

Psychomotor delay

18

12

Moderate

Prominent cisterna magna

47

M

42

Psychomotor delay

48

18

+

 

48

F

6

Seizure

30

42

+

+

+

Small lipoma in the territory of the splenium

49

M

1

Psychomotor delay; swallowing problem

42

72

Moderate

Mild

 

50

F

9

Seizure

9

12

Moderate

Prominent cisterna magna

51

M

36

Communication defect

60

78

+

 

52

F

0

Seizure

12

30

 

For each patient, we collected clinical data including age at presentation of the disease, age at the time of imaging and age at the beginning of drug therapy. These data were obtained from the children’s hospital records, and complementary history was obtained from the parents if needed. Brain images of the patients were retrieved, and in those who had no previous brain imaging MRI was performed. We obtained 52 studies (5 CT and 47 MRI) in 52 patients.

All CT scans consisted of 10-mm-thick contiguous axial sections without contrast agent. All MR imaging was performed using a 1.5-T MR scanner (GE Medical Systems, Milwaukee, WI) with a dual echo T2-W sequence (TR 2,000 ms; TE 40, 80 ms; excitations 2), T1-W (TR 600 ms, TE 20 ms, excitations 2) 7-mm-thick axial sections with a gap of 0–2.5 mm (depending on head circumference), and T1-weighted sagittal sections.

Each study was evaluated by a neuroradiologist with more than 20 years experience who was unaware of the established diagnosis. There are reported scoring schemes for the severity of the cortical atrophy and ventricular dilation, such as the semiquantitative ten-point scale devised by Yue et al. [7]. However, our neuroradiologist used a simpler scheme and scored the findings as mild, moderate or severe based on gross visual assessment. We studied the findings in two subgroups of treated and nontreated children.

Results

In 14 children (26.9%) the studies were reported as normal or within the normal range for age (mean age 54 months, SD 36 months). Of these 14 children, 10 had already received treatment for a mean of 18.9 months (SD 3.5 months). The remaining 38 children (73.1%) showed at least one abnormal radiological finding (mean age 42 months, SD 30 months).

The most common radiological findings overall were ventricular dilation, cortical atrophy, and periventricular white matter abnormalities. The frequency of findings in treated and nontreated subgroups are shown (Fig. 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0940-8/MediaObjects/247_2008_940_Fig1_HTML.gif
Fig. 1

Findings of brain imaging in 52 children with MMA in the nontreated (gray columns) and treated (white columns) subgroups. The numbers above the columns show the percentage of each finding in the represented subgroup. The numbers within parentheses are the absolute numbers of each finding in non-treated and treated subgroups, respectively

Cortical atrophy was seen in 37 boys (100%) but in only 2 girls (13.3%). In addition, the distribution of severity of cortical atrophy was different between boys and girls (Fig. 2).
https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0940-8/MediaObjects/247_2008_940_Fig2_HTML.gif
Fig. 2

Frequency of different grades of cortical atrophy in boys and girls with MMA

All three children with basal ganglia calcification were boys and they presented clinically during the neonatal period (before 1 month of age). The calcifications were seen on CT scans in all three boys. Two of these three boys had globus pallidus calcification. The third showed bilateral calcification of the putamen.

Thinning of the corpus callosum was seen in eight children (15.4%) (Fig. 3), five boys (62.5%) and three girls (37.5%). Three of these eight children had not received any treatment before imaging, and three had been treated for 1 year or less. In all of these children, MMA had presented before 6 months of age and in four of them, before 3 months of age. Four of these eight children showed periventricular white matter abnormality, while none of them had basal ganglia calcification or cerebellar atrophy.
https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0940-8/MediaObjects/247_2008_940_Fig3_HTML.gif
Fig. 3

T2-W images show thinning of the corpus callosum (a) along with significant bilateral T2 prolongation, especially in the occipital regions and the frontal white matter (b)

Six patients (11.5%, all boys) showed subcortical white matter lesions as hyperintensity on T2-weighted images (Fig. 3). In four of these six boys, MMA had presented before 3 months of age. Three of the six had not been treated before imaging, and two had received treatment for 1 year or less. Periventricular white matter hyperintense lesions (T2-weighted images) were seen in 12 patients.

Discussion

Neuropathological changes in MMA include cerebral and cerebellar atrophy, reactive gliosis, hypomyelination, multifocal cerebellar hemorrhage, and depletion or incomplete development of the cerebellar external granule cells in children more than 10 months of age. Recently, Kanaumi et al. [8] have reported the results of an autopsy in a child who died of MMA at the age of 3 years. Recent lesions consisted of multiple small hemorrhagic and necrotic foci in the caudate nucleus, cerebellum and brainstem, and old lesions consisted of hypomyelination and spongy changes scattered in the cerebral cortex, white matter, brain stem nuclei and cerebellar cortex. Previously reported findings of brain imaging in MMA include prominence of the ventricles [913], myelination abnormality [14], cortical atrophy [1113, 15] and sulcal widening [9, 11, 13], periventricular white matter abnormality [10, 1416], internal capsule changes [17], basal ganglia changes (calcification, atrophy and necrosis) [9, 15, 1724], cerebellar atrophy [14], subcortical white matter changes [9], and thinning of the corpus callosum [12].

In four children in our study the MMA was diagnosed through neonatal screening and these children received early treatment. It is interesting to note that three of these had normal images and in one, mild ventricular dilation was the only finding. On the other hand, all children who presented with psychomotor delay had abnormal examinations showing at least one abnormal radiological finding.

In our study, there were 17 children with ventricular dilation (Fig. 3). Six of these children were younger than 2 years at the time of the radiological studies and we could not be sure whether the perceived ventricular dilation was caused by MMA or reflected other causes of ventriculomegaly in infants and young children such as benign external hydrocephalus [25]. Of note, five of these patients were boys and their ages at the time of imaging were 6, 12, 12, 12, and 18 months. There was also a 12-month-old girl in this group. It is tempting to say that the severity of the ventricular dilation is affected by treatment. Figure 4 presents evidence in favor of the ameliorating effect of drug treatment on the severity of ventricular dilation. However, a longitudinal study with more than one MR study in each patient is necessary to confirm this hypothesis.
https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0940-8/MediaObjects/247_2008_940_Fig4_HTML.gif
Fig. 4

Frequency of ventricular dilation with different durations of treatment before imaging

Cortical atrophy diagnosed by observation of sulcal widening in our study is a common finding in MMA. Diffuse supratentorial white matter edema, which was reported in a study from Italy [26], was not seen among our patients. Despite reports of cerebellar hemorrhage in metabolic disorders [27, 28], there was no radiological evidence of this complication in these 52 children. However, gradient-echo imaging or other susceptibility-weighted sequences that are more sensitive in detecting hemorrhage were not performed in this study. In addition, there were four children with cerebellar atrophy that could have conceivably resulted from cerebellar lesions. Although Rutherford [14] reports lack of brainstem and internal capsular changes on MRI in neonates with MMA, recent reports speak of hemorrhagic and necrotic foci in the brainstem as well as hypomyelination and spongy changes in the brainstem nuclei [8]. In our study, two children (patients 4 and 33) showed brainstem atrophy and one child (patient 36) showed internal capsule change.

One of our patients (Fig. 5) showed a pattern suspicious of hypoglycemic brain injury with lesions in parietooccipital regions as well as periventricular white matter changes [29]. A search of the literature showed that hypoglycemia does occur in children with MMA. Indeed Lyon et al. [30] reported a 40% incidence of hypoglycemia in MMA. However, based on our patient population, the prevalence of hypoglycemic brain injury, to the extent that can be seen on sectional images, did not seem to be so high.
https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0940-8/MediaObjects/247_2008_940_Fig5_HTML.jpg
Fig. 5

A T2-W axial image of a 6-month-old child shows significant loss of brain tissue in the parietooccipital regions, with dilation of the posterior aspect of the lateral ventricles. These changes can be related to MMA or concomitant neonatal hypoglycemia

Thinning of the corpus callosum, which was reported by Enns et al. [12] in patients with MMA, was seen in eight children in our study (Fig. 3). Thinning of the corpus callosum is common in patients with white matter disease, either acquired or hereditary. Hypomyelination and delayed myelination have been reported in patients with MMA [14]. It is tempting to assume that callosal atrophy is a reflection of prevalent white matter changes in MMA, as the patient illustrated in Fig. 3 demonstrates; however, this needs further clarification. In our series, periventricular and subcortical white matter changes were present in five of eight patients with callosal thinning and the area of thinning was consistent with the areas of white matter T2 hyperintensity in these five patients (two patients with diffuse hyperintensity, two with frontal hyperintensity, and one with frontal and occipital hyperintensity). Other factors such as ventricular enlargement might be operational in thinning of the corpus callosum; three out of eight patients with thinning of the corpus callosum had ventricular dilation (one mild, two moderate).

Mild delay in myelination can be difficult to identify on brain imaging [14]. However, “suspected myelination abnormality” was suggested by the radiologist in three of our patients younger than 1 year of age. No child older than 1 year showed this finding, which is consistent with the results of a study by Brismar and Ozand [9] in 1994 showing that MMA leads to some delays in myelination rather than an “absent development.” Of three patients with myelination abnormality, two had not received any treatment until the time of imaging and one had received treatment for 3 months.

According to previous reports, basal ganglia, particularly the globus pallidus, are affected in MMA [9, 15, 1724]. Both necrosis and hemorrhage have been reported. In the radiology literature calcification of basal ganglia has been repeatedly mentioned. Calcification could be the end result of either hemorrhage or necrosis. Among the 52 children in our study, two showed focal basal ganglia calcification, and one had bilateral putaminal calcification. None of the children had atrophy or necrosis in the globus pallidus.

Although there have been a few case series in the literature dealing with radiological findings in MMA [9], most articles on this subject are case reports [1521, 23, 24, 3133] and cannot pinpoint the real frequency of these findings in MMA patients. Despite this, we were puzzled by the low incidence of basal ganglionic calcification in our series compared with the incidence reported in the literature. This low incidence might be a reflection of the fact that most of our patients had MR examinations rather than CT, and MRI is not very sensitive in detecting calcification. Our three children with detected basal ganglionic calcification had had CT examinations. It is also possible that globus pallidus changes do not appear until advanced stages of the disease [12]. Furthermore, considering the slice thickness and the gap between slices used in our imaging protocol, missing tiny abnormalities in the basal ganglia was possible. Also, because our patients had received different durations of treatment, the radiological findings might have been modified by treatment. There are reported patients with MMA in whom treatment caused regression of basal ganglionic changes [32]. A longitudinal study would be very helpful in this regard.

Although this investigation did not specifically address diffusion imaging and spectroscopy in MMA, several reports discuss these techniques. Trinh et al. [21], Gropman [34] and others have reported diffusion-weighted imaging (DWI) and MR spectroscopy (MRS) findings in patients with MMA. The globi pallidi lesions are markedly hyperintense on DWI consistent with cytotoxic edema [20, 21]. Spectroscopy, including multislice proton MRS, shows decreased N-acetyl aspartate (NAA) and sometimes increased lactate in basal ganglia and cerebrospinal fluid (ventricles) [21, 34]. Decreased NAA is a reflection of neuronal damage. Increased lactate is consistent with a conversion of energy production from aerobic respiration to anaerobic glycolysis.

Conclusion

Clinical and radiological presentations of MMA are nonspecific. However, the presence of a constellation of relevant clinical and radiological findings should raise a high index of suspicion in neurologists, radiologists and pediatricians, thus leading to appropriate referral and testing of these patients.

In addition to the previously reported findings, corpus callosum thinning, subcortical white matter changes, and brainstem thinning are findings of brain imaging in MMA. We hypothesize that treatment has an ameliorating effect on the severity of ventricular dilation in these patients.

Copyright information

© Springer-Verlag 2008