Maternal Phenylketonuria: Long-term Outcomes in Offspring and Post-pregnancy Maternal Characteristics

Abstract

Maternal phenylketonuria (MPKU) is a well-recognized complication of PKU and one of the most potent teratogenic syndromes of pregnancy. Virtually all offspring from untreated pregnancies in women with classic PKU have intellectual disabilities and microcephaly. Congenital heart disease and intrauterine growth retardation occur many times more often than expected in the general population. Control of maternal blood phenylalanine during pregnancy prevents most if not all of these complications. Previous studies demonstrated the benefits of treatment in terms of birth parameters and early development. In this study, physical examinations, a medical history, and neuropsychological evaluation were obtained in 47 children from 24 mothers with PKU who received treatment during pregnancy. Mothers were interviewed and administered an abbreviated IQ test. Associations between maternal factors and offspring outcomes were also analyzed.

The 21 male and 26 female offspring ranged in age from 1 month to 26 years with 21 (62%) over 6 years. Results indicated mean intercanthal distances above the 70th percentile. Microcephaly was present in 19% of offspring, with head circumference below the third percentile. None of the offspring had cardiac anomalies. Mean offspring IQ was 94 ± 19, with 12% performing in the range of intellectual disability (IQ < 70). Among children >5 years of age, 25% had learning disabilities, 31% had attention deficit hyperactivity disorder (ADHD), 22% were on ADHD medication, and 34% had a diagnosis of anxiety and/or depression. Among the 24 mothers, 12 reported following the diet for PKU. Only one woman on diet had a blood phenylalanine concentration <360 μmol/L (recommended range) and the majority had indications of poor nutritional status. Mean maternal Full Scale IQ was 94 ± 16 (range = 61–117), with 25% performing in the borderline intellectual range (IQ < 85). Verbal IQ was significantly lower than Performance IQ (p = 0.01, CI 2.7, 16.1). On the self-report Beck Depression Inventory, Second Edition, 25% received scores indicating mild to moderate depression, and on the Beck Anxiety Inventory, 46% reported mild to moderate anxiety. Offspring IQ correlated with maternal metabolic control during pregnancy (r = 0.51), maternal IQ (r = −0.62), and socioeconomic position (r = −0.48). Offspring with ADHD, learning disabilities, or emotional disturbances were more likely to have mothers with anxiety and/or depression. To ensure optimal offspring outcomes, healthcare providers need to assess maternal nutrition, blood phenylalanine concentrations, cognitive abilities, and socioeconomic position. Interventions can then be initiated that reduce psychosocial stressors and enhance adherence to diet and positive parenting, which in turn can lead to better cognitive functioning, behavior, and emotional well-being in their children.

Introduction

Maternal phenylketonuria (MPKU) is a well-recognized complication of PKU and one of the most potent teratogenic syndromes of pregnancy. Virtually all offspring from untreated pregnancies in women with severe PKU have intellectual disabilities and microcephaly. Congenital heart disease and intrauterine growth retardation occur many times more often than expected in the general population (Lenke and Levy 1980). The frequencies of these abnormalities in offspring are lower when the woman has a milder form of PKU but are still much greater than normally expected (Lenke and Levy 1980; Levy et al. 2003; Güttler et al. 2003).

Control of the maternal blood phenylalanine during pregnancy prevents most if not all of these complications (Lenke and Levy 1980; Rohr et al. 1987; Koch et al. 2003). The International Maternal PKU Collaborative Study (MPKUCS), a prospective, longitudinal study, showed that this was especially true if dietary therapy with control of maternal phenylalanine began before pregnancy or within the first 6 weeks of gestation (Koch et al. 2003.) In the MPKUCS, 228 children who were born to mothers with treated PKU or untreated mild hyperphenylalaninemia were compared to 70 control subjects at 7 years of age. Offspring cognitive outcome negatively correlated with the number of gestational weeks that elapsed until maternal metabolic control was achieved (r = −0.61). Behavioral outcome was similarly affected. Postnatal measurement of stimulation in the home was also related to offspring IQ (Waisbren and Azen 2003). However, the MPKUCS followed offspring only to age 7 years, thus was not able to evaluate cognitive performance into the more challenging school years.

The ability of the mother with PKU to provide a secure and intellectually stimulating environment for her offspring needs further examination. Poor metabolic control in adults with PKU is associated with deficits in executive functioning, including planning, organization, and behavioral inhibition as well as fatigue, health problems, depression, and anxiety (Koch et al. 2010; Brumm et al. 2010), all of which may limit parenting ability and result in a suboptimal environment for the offspring.

In this study, we examined offspring, ages 1 month to 26 years, from treated maternal PKU pregnancies and analyzed associations between maternal factors, including current adherence to recommendations for treatment, and offspring outcomes, such as cognitive abilities and emotional/behavioral characteristics.

Methods

Results

Discussion

The purpose of this study was to obtain a longer-term assessment of offspring from treated MPKU pregnancies and to identify maternal and environmental characteristics associated with offspring outcome. Published results of offspring outcome from treated maternal PKU have been limited to developmental studies of the neonatal and early infancy periods or to cognitive assessments in the first few years of childhood. The oldest offspring from treated maternal PKU pregnancies in our previous New England study was 4 years old (Rohr et al. 1987), and the evaluations of offspring in the MPKUCS ended at age 7 years (Koch et al. 2003; Waisbren and Azen 2003). In this study, 62% of the offspring were older than age 5 years with the oldest age 26 years. This study extended the length of follow-up into adolescence and early adulthood and also examined emotional and behavioral characteristics of these offspring. Particular attention was given to those factors of PKU in the mothers that might affect their ability to provide a secure and stimulating environment for their offspring.

None of the maternal PKU offspring in our study had evidence of heart disease or a history of having had congenital heart disease. The mean head circumference of the 48th percentile was within the expected range for the general population although in several instances, MPKU might have had a lingering adverse effect on head growth since 19% had a head circumference below the 3rd percentile, lower than the 31% for microcephaly among offspring in the MPKUCS. Facial dysmorphology included evidence of hypertelorism, as previously reported (Rouse et al. 1997, 2000), but other facial features reported as due to MPKU were absent. In summary, we found growth and somatic development in the offspring similar to the general population.

Despite the variability in current adherence to the diet for PKU, the majority of women who participated in this study perceived themselves as functioning well in their daily lives. Nonetheless, 25% performed in the borderline intellectual range, much higher than the 12.3% reported in epidemiological studies (e.g., Hassiotis et al. 2008) and higher than the 15.9% expected based on the normal population curve for IQ. Their mean IQ was 94, lower than that of adults with PKU recently studied in Europe (Weglage et al. 2013). In the study of 57 adults with PKU, the mean IQ was 100.6 compared to an IQ of 110.4 in a matched control group. However, the group of mothers in our study taking their medical food (formula) was 100, the same as that in the European study. Nearly 40% of mothers in our study were in the two lowest categories of socioeconomic position. Closer examination of their scores on tests of verbal expression, verbal reasoning, depression, and anxiety revealed vulnerabilities that may have an important impact on their parenting skills and hence on their children’s development and well-being.

Within the general population, individuals with IQ in the borderline range are significantly more likely to be at social disadvantage, experience neurotic disorders (such as anxiety and depression), and suffer from substance misuse. They take psychotropic medications and seek emergency services at a higher rate but are not more likely to seek psychotherapy (Hassiotis et al. 2008). Other studies report that adults with borderline IQ are at risk for poor occupational attainment and depression (Seltzer et al. 2009).

Children of parents with borderline IQ, anxiety, or depression also seem to be negatively affected, not so much during the preschool period but after age 7 years, with increased rates of conduct disorders and emotional and attention problems (Whitely et al. 2011). Likewise in our study of maternal PKU offspring, children under age 3 years exhibited fewer cognitive or behavioral deficits than children older than 3 years. By school age, the children were much more likely to have ADHD, learning disabilities, and anxiety or depression than children in the general population. This suggests that the deficits noted later in childhood may be related not only to prenatal effects but also to environmental circumstances, including the home environment, maternal depression, anxiety, and educational opportunities. We found that the majority of school-aged maternal PKU offspring with behavioral disturbances, emotional difficulties, or low IQ had mothers who were depressed or anxious. The percentage of children with psychiatric symptoms or low IQ may be even higher than we reported, given that one mother appeared to be an "outlier." Her 5 offspring had a range of problems including anxiety and depression, ADHD, low IQ, and learning disabilities, but she rated herself as having no anxiety or depression. Based on clinical observation and prior medical reports, however, she experiences both anxiety and depression. If she had rated herself as such, all offspring with ADHD, all with low IQ, all with learning disabilities, and 82% of those with anxiety and depression would have mothers who were experiencing anxiety or depression.

While there was overlap between current maternal blood phenylalanine results for the on- and off-diet groups, median blood phenylalanine concentrations were lower in the on-diet group. Notably, only one woman on diet had blood phenylalanine concentration <360 μmol/L, which is considered to be safe for an MPKU pregnancy (Vockley et al. 2014). The typical "on-diet" approach for this population consisted of taking about 75% of the prescribed phenylalanine-free or low-phenylalanine medical food (formula) for PKU and avoiding high-protein foods such as meat, eggs, nuts, and dairy products. Few women strictly reduced their phenylalanine intake, as evidenced by elevated plasma phenylalanine concentrations.

The women off-diet were at significantly higher risk of having low blood tyrosine, which may be counterintuitive since the off-diet group consumes more natural protein. However, medical food is a source of tyrosine and may account for the significantly higher blood tyrosine in the on-diet group. Other discrepancies between amino acid values in the two groups are notable, especially higher blood concentrations of the large neutral amino acids—valine, leucine, and tryptophan—in the on-diet group. Valine and leucine are large neutral amino acids (LNAAs) which compete for transport of phenylalanine into the brain. High amounts of LNAAs in the blood block phenylalanine uptake as well as promote neurotransmitter synthesis (Pietz et al. 1999).

Half of the women had 25-OH vitamin D results below the lower limit of 30 ng/mL, but only one had severe vitamin D deficiency (<20 ng/mL) (Christesen et al. 2012). A recent Cochrane review indicates that vitamin D supplementation and higher levels of 25-OH vitamin D in pregnant women have been associated with increased birth weight. Moreover, observational studies show a positive effect of vitamin D status on other health outcomes in children. Indices of iron nurtiture (hemoglobin, hematocrit, ferritin) were normal in both groups, as were zinc and folate. Plasma vitamin B12 was significantly lower in the off-diet group compared to the on-diet group, again indicating that the medical food is a significant source of vitamin B12 in the diets of mothers with PKU. Vitamin B12 deficiency in adults with PKU is a well-established phenomenon arising from diets that do not contain either medical food or animal products (Hvas et al. 2006).

The prevalence of overweight women with PKU (45%) and obesity (27%) was somewhat above national norms that indicate 64% of the population is either overweight or obese Fryar and Ogden 2014; Fryar et al. 2014. Female children with PKU have been reported to have a higher incidence of being overweight or obese in a single study (Burrage et al. 2012). In a previous study of pregnant women, a bimodal distribution of prepregnancy weight was observed with nearly equal numbers of overweight and underweight women (Rohr et al. 2004).

The suboptimal nutritional status of these women with PKU may have implications for their emotional well-being. For example, recent studies demonstrate associations between low vitamin D or low vitamin B12 and depression (Anglin et al. 2013; Kalita et al. 2013), and low tyrosine has been implicated in depressive symptoms in PKU (Sharman et al. 2012). Moreover, anxiety and agoraphobia have long been recognized as symptoms of poor metabolic control in PKU patients (Waisbren and Levy 1991; Brumm et al. 2004).

Very few women were being treated with tetrahydrobiopterin supplementation and none with large neutral amino acids. The women who were enrolled in the study were older, and many were off-diet and not seen frequently in clinic. Thus, they may be part of the "lost generation" of adults with PKU who do not have the opportunity to learn about updated therapies for PKU (Burton and Leviton 2010). However, this MPKU follow-up study offered an opportunity to update women about treatment options, and yet none chose to change their current treatment. Responses to the nutrition questionnaire may provide insight into this behavior. About half of the women who were off-diet reported that they had tried to return to diet at some point in the postpartum period. When asked about obstacles to returning to diet, the most common responses were not the expected ones (difficulty with formula, protein restriction, or insurance) but rather that they did not perceive a need for any treatment. Only one mother reported lack of access to treatment as an obstacle.

One limitation of this study was the use of self-report and parental report instruments. Rates of offspring difficulties may be underestimated in our study, since the mothers tended not to report themselves or their children as functioning below the average range while cognitive test scores tended to suggest otherwise. Subgroup comparisons of on-diet versus off-diet women or children with versus without cognitive deficits or mood disorders involved small sample sizes.

Conclusion

The primary conclusion of this study is that maternal diet influences offspring outcome—before, during, and after pregnancy. Mothers who are well treated from the very beginning of their lives have a higher IQ and therefore often a higher SES, both of which were found to correlate with offspring outcome. They are also less likely to be anxious or depressed which was a determining factor in offspring with ADHD, anxiety, depression, and low IQ. While maternal metabolic control during pregnancy explains much about offspring outcomes, other parental characteristics may also contribute to the increased rates of low IQ and attention problems in these children, and these appear to be directly related to diet and medical food intake. Moreover, it takes time for the maternal PKU offspring’s problems to emerge. Most social-emotional problems were not evident until the children were school-aged. For this reason, it is important that MPKU offspring have psychological evaluations throughout childhood and adolescence and that mothers continue to receive therapy for PKU throughout the lifespan. Evidence for positive maternal PKU pregnancy outcomes is now well established. This study illustrates that risks associated with maternal PKU do not end with the birth of the infant, but continue throughout the child’s life. To ensure optimal offspring outcomes, healthcare providers need to assess maternal nutrition, blood phenylalanine concentrations, cognitive abilities, mood, and socioeconomic position. Interventions can then be initiated that reduce psychosocial stressors and enhance adherence to medical recommendations and positive parenting, which in turn can lead to better cognitive functioning, behavior, and emotional well-being in the children.

The authors gratefully acknowledge Wen-Hann Tan, M.D., who assisted with the study as well as Annie Gardner who helped edit this manuscript, the families who participated in this research, and the staff at the Clinical and Translational Study Unit (CTSU) at Boston Children’s Hospital. This study was supported by a research grant from the National PKU Alliance and the Milton Foundation, Harvard Medical School.

Footnotes

Contributor Information

S. E. Waisbren, Email: susan.waisbren@childrens.harvard.edu

Collaborators: Johannes Zschocke

References