maternal phenylketonuria: magnetic resonance imaging of the brain in offspring

Maternal phenylketonuria: resonance imaging of the offspring

Magnetic brain in

Harvey L. Levy, MD, Deborah Lobbregt, BS, Patrick D. Barnes, MD, and Tina Young Poussaint, MD

From the Genetic and Radiology Services, Children's Hospital, and the Departments of Pe- diatrics and Radiology, Harvard Medical School, Boston, Massachusetts

Phenylketonuria (PKU) produces white matter changes identifiable by magnetic resonance imaging. These changes occur postnatally. Offspring of untreated mothers with PKU also have a brain effect, expressed as microcephaly and mental retardation. This effect occurs prenatally. To determine whether the white matter changes seen in PKU are also present in maternal PKU offspring, despite the different developmental stages of exposure to PKU, we performed brain magnetic resonance imaging studies in seven maternal PKU offspring, five from essentially untreated pregnancies and two from treated pregnancies. None had white matter changes, although the one offspring with PKU had delayed myelination. How- ever, hypoplasia of the corpus callosum was present in three of the four offspring from untreated pregnancies and in the offspring from a maternal PKU pregnancy not treated until the third trimester. Unlike PKU, white matter changes are not a feature of the brain effect in maternal PKU. However, hypoplasia of the corpus callosum is a feature of maternal PKU and is probably a result of inhibition of corpus callosum development at 8 to 20 weeks of gestation. The hypoplastic corpus callosum could be a marker for brain effect in maternal PKU and may have implications for the cognitive deficits in these offspring. (J Pediatr 1996;128:770-5)

The most striking and consistent neuropathologic finding in phenylketolmria has been myelin deficiency in the brain. This was originally reported from direct neuropathologic examinations.i, 2 More recently, magnetic resonance imaging of the brain in PKU has revealed bright T2 signal abnormalities in the cerebral white matter interpreted as evidence of hypomyelination. 3-6 These MRI abnormalities have been present even in early-treated individuals with PKU and seem not to correlate with age at onset of treatment or with IQ. 3, 5, 7

Even severe changes have been present as early as 8 years

Supported by contract No. N01-HD-2-3149 from the National In- stitute of Child Health and Human Development. Submitted for publication Oct. 26, 1995; accepted Jan. 29, 1996. The conclusions in this article do not necessarily represent the conclusions of the Maternal PKU Collaborative Study. Reprint requests: Harvey L. Levy, MD, I. C. Smith 106, Children's Hospital, 300 Longwood Ave., Boston, MA 02115. Copyright © 1996 by Mosby-Year Book, Inc. 0022-3476/96/$5.00 + 0 9/20/72373

of age. 3 Whether the cognitive loss in PKU can be explained by these changes is unclear.

Offspring of mothers with PKU untreated during pregnancy (maternal PKU) also have cognitive loss, often expressed as frank mental retardation. 8 Unlike PKU, in which the abnormalities result from postnatal exposure to increased phenylalanine levels, the damage in maternal PKU occurs from prenatal exposure to maternal hyperphenylala-

I MRI Magnetic resonance imaging PKU Phenylketonuria

ninemia. 9 Thus, although the toxic element in PKU and maternal PKU may be the same, the periods of brain development when exposure occurs are different. This raises the question of whether the neuropathologic features are the same regardless of whether the toxic effect is due to PKU or maternal PKU. If so, the response may not dependent on the stage 'of development during exposure. Different neuro-

7 7 0

The Journal of Pediatrics Levy et al. 7 7 1 Volume 128, Number 6

Fig. |. Nine-year-old boy from a treated maternal PKU pregnancy (offspring 7); MRI of the brain demonstrates normal corpus callosum. The parts of the corpus callosum are (a) rostrum, (b) genu, (c) body, and (d) splenium.

pathologic features, however, may indicate that the stage of development is critical in determining the neuropathologic

response. In addition, maternal PKU offspring, unlike those with PKU, have microcephaly at birth, s indicating that prenatal brain growth is retarded. These offspring also have high frequencies of anomalies such as congenital heart disease and esophageal atresia,S which raises the question of whether

anomalies of the brain may also be present. To study these questions, we performed cerebral MRI

studies on five offspring from untreated or very late treated maternal PKU pregnancies and, as a comparison, two offspring from early-treated maternal PKU pregnancies that were in excellent metabolic control.

M E T H O D S

Birth head circumference percentiles were detemlined according to the growth chart compiled by the National

Center for Health Statistics, Centers for Disease Control and Prevention, Atlanta, Ga. Plasma amino acids were measured by an amino acid analyzer.

We performed MRI of the brain on seven children, aged 6 to 17 years (four girls). Four patients had MRI performed on a Signa GE 1.5 T system (GE Medical Systems, Milwau-

kee, Wis.), one patient on a Magnetom 1.0 T system (Sie- mens Medical Systems Inc., Iselin, N.J.), and another patient on a Diasonics system (Diasonics Inc., Milpitas, Calif.); for a seventh patient the imaging system was unknown. Sagittal Tl-weighted, axial proton density, and axial T2-weighted

images were analyzed for myelination and the presence of morphologic abnormalities.

The midline sagittal Tl-weighted image was used to as- sess the size of the corpus callosum and its component parts.

The splenium should be approximately equal in size to the genu by age 12 months 1° (Fig. 1). The definition of a par-

tially formed corpus callosum was one in which the genu is

present, the body less commonly so, and the rostrum and splenium are absent or small. 11

Informed consent was obtained for the study of maternal PKU in all cases. 12, 13

CASE REPORTS

The pertinent clinical data on these offspring and their mothers are summarized in Table I.

Offspring 1. Offspring 1 was a 17-year-old girl from an untreated PKU pregnancy. Her mother has classic PKU with a plasma phenylalanine level of 1260 pmol/L (20.8 mg/dl). The mother has low normal intelligence and was not identified as having PKU until she bore this child and the umbilical cord blood was found, on the basis of routine screening, to have an elevated phenylalanine level.12 This offspring was small for gestational age and had microcephaly at birth. Her newborn PKU test result was normal. Her development was slow, and she manifested marked hyperactivity through child- hood. Her full-scale IQ at age 8 years was 98. At age 17 she had severe emotional difficulties.

Offspring 2. Offspring 2, an 8-year-old boy, was the product of an untreated maternal PKU pregnancy. His mother has classic PKU with a plasma phenylalanine level of 1860 Nnol/L (30.7 mg/dl). The

7 7 2 Levy et al. The Journal of Pediatrics June 1996

Table I. Maternal phenylalanine levels and characteristics of offspring from maternal PKU pregnancies

Maternal Offspring characteristic

plasma Head circumference at birth Offspring phenylalanine Gestation

No. (IJmol/L) (wk) Measurement (cm) Percentile IQ Other

1 1260 40 31.0 <5th 2 1860 40 29.5 <5th 3 1510 37 29.5 <5th 4 1630 39 29.0 <5th 5 1370 40 33.0 10th 6 1740 40 36.0 75th 7 1740 40 34.5 40th

98 Hyperactive; emotional difficulties 50 Hyperactive 54 Hyperactive

<50 Hyperactive 79 Emotional difficulties

108 Hyperactivity 94 Hyperactivity

Table II. MR[ findings in the brain of offspring of

maternal PKU pregnancies

MRI findings Offspring

No. Myelination Corpus callosum

1 Normal Small splenium 2 Normal Absent rostrum, hypoplastic

splenitun 3 Normal Thinned splenium, possibly

absent rostrum 4 Delayed Hypoplastic splenium 5 Normal Normal 6, 7 Normal Normal

offspring had marked microcephaly at birth and continues to have microcephaly with mental retardation (IQ 50) and hyperactivity. His newborn PKU test result was normal.

Offspring 3. Offspring 3, a 6-year-old girl, was from a late- treated maternal PKU pregnancy. The mother has classic PKU with a plasma phenylalanine level of 1510 N-nol/L (24.9 mg/dl). The pregnancy was not identified until late in the second trimester. Di- etary treatment was initiated at 24 weeks' gestation, with good metabolic control until delivery. The offspring had marked microcephaly at birth and continues to have microcephaly with mental retardation (IQ 67) and hyperactivity. Her newborn PKU test result was normal.

Offspring 4. Offspring 4, a girl, has been reported previously. TM

She was an 8-month-old girl with PKU who was the product of an essentially untreated maternal PKU pregnancy. Her mother has classic PKU. At 8 weeks' gestation the mother had a blood phenylalanine level of 1630 omol/L (26.9 mg/dl). A phenylalanine-restricted diet was attempted, but the blood phenylalarline levels re- mained at greater than 1200 ~mol/L (20 mg/dl). At birth the offspring was small for gestational age and had marked microcephaly. Shortly after birth she was fotmd to have esophageal atresia and PKU. Her plasma phenylalanine level was 1365 gmol/L (22.5 mg/ dl) before the start of a phenylalanine-restricted diet at 14 days of age. The esophageal atresia was surgically repaired. The gift is se- verely retarded.

Offspring 5. Offspring 5, a 17-year-old girl, was the product of

an untreated maternal PKU pregnancy. Her mother has classic PKU, with a blood phenylalanine level of 1370 ~mol/L (22.6 mg/dl). The offspring did not have microcephaly at birth but was found to have microcephaly with borderline intelligence (IQ 79) when subse- quently examined in a maternal PKU study. 12 Her newborn PKU test result was normal.

Offspring 6 and 7. Offspring 6 and 7, male siblings aged 11 and 9 years, were products of treated maternal PKU pregnancies in a mother with classic PKU whose untreated blood phenylalanine level is 1740 pmol/L (28.7 mg/dl). Strict dietary control for maternal PKU was maintained during both pregnancies, and her blood phenylalanine levels were generally less than 360 ~unol/L (6 mg/dl). Both offspring had a normal head circumference at birth and have had normal to low normal IQ scores (108 and 94). Their newborn PKU test results were normal. However, both are receiving medi- cation for hyperactivity and are having problems at school.

M R I R E S U L T S

Table II lists the results of cerebral MR[ studies in the

seven maternal PKU offspring. In none of the offspring

without phenylketonuria were there changes in the white

matter or any other changes in brain substance. There was

delayed myelination in the offspring with PKU, as evidenced

by lack of mature myelin in the posterior limb of the inter-

nal capsule at age 8 months. As noted in Table lI, four of the

five offspring from untreated or very late treated maternal

PKU pregnancies had hypoplasia of the corpus callosum

(Figs. 2 and 3). The most severe changes in the corpus cal-

losum were present in the three offspring (offspring 2 to 4)

who were mentally retarded (Table I). No anatomic changes

were observed in the MR[ scan of the fifth offspring. None

of the offspring had any other associated brain anomalies.

The two siblings from early-treated maternal PKU pregnan-

cies had normal MRI scans (Fig. 1).

D I S C U S S I O N

This study of brain MRIs in offspring of untreated or late-

treated maternal PKU pregnancies indicates that the white

matter changes noted in individuals with PKU may not be


Fig. 2. Six-year-old girl from a late (third-trimester) treated matemal PKU pregnancy (offspring 3); MRI of the brain demonstrates a possibly absent rostrum (anterior arrow) and a thin splenium (posterior arrow) of the corpus callosum.

present in offspring of mothers with PKU. On this basis the changes in the brain produced by increased phenylalanine levels prenatally in maternal PKU may differ from those in response to the postnatal phenylalanine exposure of PKU. This difference might be expected because, in human beings, cerebral myelinafion is largely a postnatal event, 15 though oligodendrocytes, the cells that elaborate myeli n, could be irreversibly damaged in the maternal PKU fetus. A more direct explanation for normal myelination postnatally in maternal PKU offspring might be that the white matter changes are simply a reflection of the phenylalanine level at the time of the MR[. White matter changes seen on an MR[ scan of persons with PKU seem to correlate most strongly with the blood phenylalanine level when the MR[ is performed 7 or within a year of MRI examination. 4 All the offspring whom we reported, except the one with PKU, had normal phenylalanine levels at the time of MRI examination, which suggests that normal phenylalanine levels might have been the determining factor in their lack of white matter changes. In the 8-month-old offspring with PKU, the young age could also have been a factor in the lack of white matter changes, because in PKU these changes have not been reported during the first year of life.

If the white matter changes seen by MR[ in PKU reflect aberrant myelin formation, recent experimental work by Dyer et al)6 explains the basis of this abnormality relative to the phenylalanine level and why myelin changes are absent in maternal PKU offspring. The brain of the mouse model for PKU that they are studying is hypomyelinated and

gliotic, just as is the brain of human beings with PKU. 17 Their data indicate that as a result of exposure to high levels of phenylalanine, oligodendrocytes switch to a nonmyelinating phenotype that expresses an astrocyte marker, glial fibrillary acidic protein. Lowering blood phenylalanine levels results in the loss of cells expressing this protein and the production of myelin. Thus in PKU the oligodendrocyte is likely to be nonmyelinating, whereas in the offspring of mothers with PKU, normal phenylalanine levels postnatally would favor myelin production by oligodendrocytes.

The other interesting fending in our study is the presence of an abnormal corpus callosum in four of the five offspring born after untreated or very late treated pregnancies, but not in either of the siblings born after early-treated pregnancies. We are aware of an additional offspring from an untreated maternal PKU pregnancy who had hypoplasia of the corpus callosum identified by computed tomography at age 2 years.14 Abnormalities of the corpus callosum have been reported in a number of metabolic disorders that seem to have a prenatal effect. 18, 19 They have also been seen in offspring

from pregnancies complicated by teratogenic factors such as maternal rubella, maternal diabetes, and the fetal alcohol syndrome 2° and have been reported in many other condi- tions. 21 Conversely, in a disorder such as PKU, where the effect is postnatal, abnormalities of the corpus callosum have not been observed. 3, 5, 7 The abnormalities of the corpus callosum that we found in the three offspring from untreated maternal PKU pregnancies and the offspring from a very late treated pregnancy would be consistent with the prenatal ef-

7 7 4 Levy et al. The Journal of Pediatrics June 1996

Fig. 3. Eight-year-old boy from an untreated maternal PKU pregnancy (offspring 2); MRI of the brain demonstrates absent rostrum (anterior arrow) and hypoplastic splenium (posterior arrow) of the corpus callosum.

fect of maternal PKU. The corpus callosum was normal, however, in the siblings from early-treated pregnancies, whereas one offspring with an abnormal corpus callosum was from a pregnancy treated in the third trimester. This observation might suggest that inhibition of corpus callosum growth in maternal PKU occurs rather early in pregnancy and can be prevented by treatment that begins during the first trimester but may not be affected by treatment late in pregnancy, particularly when treatment is delayed beyond the twentieth gestational week, at which time the corpus callosum is fully formed. Further MRI studies of the brain in offspring from maternal PKU pregnancies would be required to examine this question more closely.

The development of the corpus callosum and its precursors occurs between 8 and 20 weeks of gestation. 1°' 11 This development involves a series of events whereby the genu, body, splenium, and rosmma form in sequence. Disordered development will result in a completely or partially absent corpus callosum. Hypoplasia of the corpus callosum may be secondary to arrested growth or delayed continued development. zl Because fetal brain growth seems to be impaired in maternal PKU, 9 it is likely that arrested growth is the major reason for tile hypoplasia of the corpus callosum in these offspring. Anomalies of the corpus callosum have been found to be a marker of additional brain anomalies and likely occur because of insults to the brain during the formation of

callosal precursors. 22 However, we did not detect other brain anomalies in our study. In addition to agenesis and partial agenesis of the corpus callosum, hypoplasia of the corpus callosum has been suggested as a marker of cerebral dysgenesis and has been correlated with mental retardation.Z3, 24 With regard to this correlation, the offspring in our study with frank mental retardation (offspring 2 to 4) had the most severe changes in the corpus callosum.

The development of the corpus caUosum is dependent on the formation of a "glial sting" that serves as a developmental matrix to guide axons to their targets during development. 25 Perhaps the elevation of phenylalanine levels during development disrupts the formation of this glial bridge, preventing the normal migration of axons and lead- ing to an abnormal corpus caUosum. Such a hypothesis would be consistent with the observation of Dyer et a1.16 that gtial cells are a primary target of elevated phenylalanine levels in the postnatal brain.

Whether maldevelopment of the corpus callosum in offspring of mothers with PKU has functional implications re- mains to be determined. In particular, disruption of informa- tion transfer between the two cerebral hemispheres could be an effect of this abnormality, as demonstrated in callosal agenesis. 26 It is of interest that children with early-treated PKU have also been found to have slowed hemisphere transfer, believed to be a result of hypomyetination in the


corpus callosum. 27 Offspring of mothers with PKU should

be studied by neuropsychologic indicators of integrity of the corpus callosum. A finding of slowed interhemispheric transfer could explain the learning disabilities and perhaps other cognitive deficits commonly present in these children.

Hypoplasia of the corpus callosum could be a marker for

adverse brain development in maternal PKU. Whereas the three offspring in our study with the most marked callosal changes are mentally retarded, the girl with a normal IQ had

only mild changes in the splenium (offspring 1) and another girl with borderline intelligence had a normal corpus callosum. Thus the development of the corpus callosum might be

somewhat predictive of intellectual outcome in maternal PKU and, if so, could be useful in this assessment for indi- vidual families and for studies such as the current Maternal PKU Collaborative Study. 13 However, further study of the

corpus Callosum in offspring from treated maternal PKU pregnancies would be required to make this determination and is urgently needed.

We thank the families who so generously cooperated in the per- formance of these studies.

R E F E R E N C E S

1. Malamud N. Neuropathology of phenylketonuria. J Neuro- pathol Exp Neurol 1966;25:254-68.

2. Poser CM, Van Bogaert L. Neuropathologic observations in phenylketonufia. Brain 1959;82:1-9.

3. Pearsen KD, Gean-Marton AD, Levy HL, Davis KR. Phe- nylketonuria: MR imaging of the brain with clinical correlation. Radiology 1990;177:437-40.

4. Thompson AJ, Smith I, Brenton D, et al. Neurological deteri- oration in young adults with phenylketonuria. Lancet 1990; 336:602-5.

5. Bick U, Fahrendorf G, Ludolph AC, Vassallo P, Weglage J, Ullrich K. Disturbed myelination in patients with treated hyperphenylalaninaemia: evaluation with magnetic resonance imaging. Eur J Pediatr 1991;150:185-9.

6. Lou HC, Toft PB, Andresen J, et al. An occipto-temporal syndrome in adolescents with optimally controlled hyperphenylalaninaemia. J Inher Metab Dis 1992;15:687-95.

7. Cleary MA, Walter JH, Wraith JE, Jenkins JPR, Alanis M, Tyler K, et al. Magnetic resonance imaging of the brain in phenylketonuria. Lancet 1994;344:87-90.

8. Lenke RR, Levy HL. Maternal phenylketonuria and hyperphenylalaninemia: an international survey of the outcome of untreated and treated pregnancies. N Engl J Med 1980;303: 1202-8.

9. Levy I-IL. Maternal phenylketonuria: review with emphasis on pathogenesis. Enzyme 1987;38:312-20.

10. Barkovich AJ, Kjos BO. Normal postnatal development of the

corpus callosum as demonstrated by MR imaging. AJNR Am J Neuroradiol 1988;9:487-91.

11. Barkovich AJ. Congenital malformations. In: Barkovich AJ, editor. Pediatric neuroimaging. 2nd ed. New York: Raven Press, 1995.

12. Levy HL, Waisbren SE. Effects of untreated phenylketonuria and hyperphenylalaninemia on the fetus. N Engl J Med 1983; 309:1269-74.

13. Koch R, Levy HL, Matalon R, et al. The international collaborative study of maternal phenylketonuria: status report 1994. Acta Paediatr Suppl 1994;407:111-9.

14. Levy HL, Lobbregt D, Sansaricq C, Snyderman SE. Compar- ison of phenylketonuric and nonphenylketonuric sibs from untreated pregnancies in a mother with phenylketonuria. Am J Med Genet 1992;44:439-42.

15. Gilles FH, Leviton A, Dooling EC. The developing human brain: growth and epidemiologic neuropathology. Boston: John Wright, 1983.

16. Dyer CA, Philipotte T, Levy H, Cruz J. Mechanisms of hypomyelination and gliosis in a genetic mouse model for phenylketonuria. J Neurochem 1995;64(suppl):S 111.

17. Alford EC, Stevenson LD, Vogel FS, Engle RL. Neuropatho- logical findings in phenyl-pymvic oligophrenia (phenylketonuria). J Neuropathol Exp Neurol 1950;9:298-310.

18. Bamforth F, Bamforth S, Poskitt K, Applegarth D, Hall J. Ab- normalities of corpus callosum in patients with inherited metabolic diseases [letter]. Lancet 1988;2:451.

19. Kolodny EH. Agenesis of the corpus callosum: a marker for inherited metabofic disease? Neurology 1989;39:847-8.

20. Riley EP, Mattson SN, Sowell ER, Jemigan TL, Sobel DF, Jones KL. Abnormalities of the corpus callosum in children prenatally exposed to alcohol. Alcohol Clin Exp Res 1995;I9: 1198-202.

21. Rnbinstein D, Youngman V, Hise JH, et al. Partial development of the corpus callosum. AJNR Am J Neuroradio11994; 15:869- 75.

22. Barkovich AJ, Norman D. Anomalies of the corpus callosum: correlation with further anomalies of the brain. AJNR Am J Neuroradiol 1988;9:493-501.

23. Bodensteiner J, Schaefer GB, Breeding L, et al. Hypoplasia of the corpus callosum: a study of 445 consecutive MRI scans. J Child Neurol 1994;9:47-9.

24. Fujii Y, Konish Y, Kuriyama M, et al. Corpus callosum in de- velopmentally retarded infants. Pediatr Neurol 1994;11:219- 23.

25. Lefkowitz M, Durand D, Smith G, Silver J. Electrical proper- ties of axons within Probst bundles of acallosal mice and cal- losi that have reformed upon gfial-coated polymer implants. Exp Neurol 1991;113:306-14.

26. Mihier AD, Jeeves MA, Silver JP, Lines CR, Wilson J. Reac- tion times to lateralized visual stimuli in calllosal agenesis: stimulus and response factors. Neuropsychologia 1985;23:323- 31.

27. Gourovitch ML, Craft S, Dowton SB, Ambrose P, Sparta S. Interhemispheric transfer in children with early-treated phenylketonuria. J Clin Exp Neuropsychol 1994;16:393-404.

maternal phenylketonuria: magnetic resonance imaging of the brain in offspring

Documents