As part of an ongoing natural history study, we present the common phenotypic features observed in our initial evaluation of 92 children with AS up to 5 years old. Although molecular diagnosis of AS is not required for participation if the clinical diagnostic criteria () are met, all participants in this report had molecularly proven AS. The distribution of the molecular etiology among our participants is consistent with previous studies [Clayton-Smith and Laan, 2003
; Williams et al., 1993–2009
]. Therefore, we believe that our data are representative of the clinical manifestations of AS in children up to 5 years old. Although this is one of the largest cohorts of AS individuals in the medical literature, we have relatively few participants with UPD/imprinting defects and UBE3A
mutations, and hence the results in these two subclasses should be interpreted with caution.
Nearly half of the children with UPD/imprinting defects in our study were obese and all but one were overweight (BMI >85th centile) despite a high prevalence of feeding difficulties in infancy. Obesity has previously been reported in AS, particularly among the “non-deletion” AS individuals, but it is generally thought that obesity occurs mainly in older individuals [Williams et al., 2006
]. Several children in our study with UPD/imprinting defects were reported to have food-seeking behavior reminiscent of children with Prader–Willi syndrome, which raises the question of why having two presumably epigenetically identical copies of chromosome 15 results in food-seeking behavior. Having two copies of chromosome 15 with a paternal imprint might lead to over-expression of the paternally expressed genes on that chromosome. However, in Prader–Willi syndrome, there is a lack of expression of these paternally expressed genes. Therefore, it is difficult to reconcile how any of the paternally expressed genes on chromosome 15 could be directly responsible for the hyperphagia and obesity seen in AS individuals with UPD/imprinting defects.
In contrast to the previously reported studies, only 71% of our participants with deletions and fewer than 60% of our participants with either UPD/imprinting defects or UBE3A
mutations had clinical seizures. However, our study may have underestimated the prevalence of seizures, especially in those with UPD/imprinting defects in whom the age of seizure onset is usually between 5 and 6 years old [Moncla et al., 1999b
; Lossie et al., 2001
; Varela et al., 2004
]. On the other hand, 95% of the participants with deletions who had experienced seizures did so by the age of 36 months, which suggests that our estimated prevalence of seizures in this subgroup of AS children may be reasonably valid. Nonetheless, abnormal EEG findings similar to those previously reported in this population such as rhythmic 4–6 Hz activity that persists with eye closure and high voltage slow 2–3 Hz delta activity with spikes and sharp waves were observed in all 84 participants who had this investigation [Laan and Vein, 2005
]. This suggests that specific EEG findings may provide clues to the diagnosis of AS even in the very young children who have not had any clinical seizures, and it may be an effective screening tool with what appears to be 100% sensitivity, even if not highly specific for AS.
Frequent, inappropriate, or easily provoked laughter is another defining feature of AS [Angelman, 1965
]. This has previously been observed in 96–97% of individuals with AS due to a deletion [Saitoh et al., 1994
; Varela et al., 2004
], 78% of those with UPD [Varela et al., 2004
], and 100% of those with UBE3A
mutations [Moncla et al., 1999a
], although the number of participants in the two latter groups was relatively small. In contrast, fewer than 70% of our participants had easily provoked laughter. Of note, in a survey of 72 individuals between the ages of 5 and 33 years with AS due to a deletion, “episodes of inappropriate laughter” were reported in only 57% of them [Clarke and Marston, 2000
], suggesting that such behavior may become less common with age [Pelc et al., 2008
]. Although these individuals might laugh more readily than individuals without AS, some caregivers consider the laughter “appropriate” as long as a triggering stimulus can be identified. The true prevalence of “frequent” or “inappropriate” laughter recorded in surveys may therefore depend on the congruence between the investigators’ and the parents’ definitions of these terms. Moreover, it has been shown that children with AS initiate contact with adults, and smile at adults before being smiled at, more often than children with other intellectual disabilities of similar severity [Oliver et al., 2007
]. This has led to the hypothesis that children with AS find interactions with adults unusually rewarding and use smiling and laughter to maintain their “social resources,” which further suggests that the laughter and smiles are not “inappropriate” [Oliver et al., 2007
]. We have also observed that some of these children also tend to laugh when they are upset or fatigued, suggesting that laughter might be an “emotional” response to stress as well as a form of expression.
Fascination with water, another characteristic well known to the parents of children with AS, was observed in approximately 80% of our participants with deletions and 60–65% of those with either UPD/imprinting defects or UBE3A
mutations, similar to reports from previous case series in which 68–79% of participants with deletions exhibited this trait [Clarke and Marston, 2000
; Kara et al., 2008
Mouthing of objects, although not often reported in case series, is said to occur in fewer than 80% of AS individuals [Williams et al., 2006
]. However, it was reported in 95% of our participants, including 99% of those with a deletion. This difference may be due to the fact that the overall age of our subjects is younger than that of previous studies.
To assess the specificity of the various behavioral characteristics in AS, some authors have compared the prevalence of specific behavioral traits in individuals with AS to those with other intellectual disability syndromes. AS individuals were found to exhibit more mouthing behavior, hand-flapping, excitability, and had shorter attention span, and were more cheerful and less anxious, than those with Down syndrome and Prader–Willi syndrome [Walz and Benson, 2002
]. Comparing AS individuals with deletions and UPD/imprinting defects to individuals with moderate or profound “intellectual disability,” Barry et al., found that significantly more AS individuals had mouthing behavior (76% vs. 43%) and sleep disturbances than the control group (68% vs. 26%), but significantly fewer AS individuals had short attention span compared to the control group (46% vs. 74%); there was no significant difference in the prevalence of “laughs or giggles for no obvious reasons” between AS individuals and the control group (49.2% vs. 49.6%) [Barry et al., 2005
]. However, individuals with UPD/imprinting defects were over-represented (29%) in their cohort, and their control group was heterogeneous by design. A more recent study showed that individuals with AS had a stronger preference for, and attraction to, water-related activities compared to individuals with Down syndrome and those with “non-specific intellectual disability” [Didden et al., 2008
We propose that the clinical suspicion for AS should be based on the neurobehavioral phenotype rather than specific dysmorphic features. The mid-face hypoplasia and prognathism depicted in standard textbooks were observed in fewer than 30% of our participants [Jones, 2006
], and may be more characteristic of older individuals. Although behavioral attributes are often subjective and lack standardized operational definitions, identifying a “behavioral phenotype” can be useful for the diagnosis of some genetic syndromes such as Williams syndrome and Prader–Willi syndrome [Dykens, 1995
; Finegan, 1998
; Cassidy and Morris, 2002
]. Our data, together with the findings of the comparative studies described above, suggest that the constellation of mouthing behavior, sleep difficulties, and fascination with water in a child with developmental delay, absent or minimal speech development, abnormal EEG, and an ataxic or broad-based gait should raise the suspicion for AS. More importantly, the absence of seizures or the lack of easily provoked or “inappropriate” laughter should not discourage consideration of this diagnosis. A few studies have suggested that the phenotypic features of AS change with age such that facial features become more prominent (though wide-spaced teeth become less common), and the attention span and sleep difficulties improve, but the fascination with water and EEG abnormalities persist [Buntinx et al., 1995
; Laan et al., 1996
; Smith, 2001
]. In our cohort, participants with deletions were significantly younger than those in the non-deletion subclasses; however, other than having a higher prevalence of mouthing behaviors and lighter skin or hair color, and being less heavy and more microcephalic, there were no significant differences in the physical and behavioral traits between the deletion and the non-deletion subclasses.
The clinical phenotype of AS in the older individuals remains poorly defined, and it is hoped that as an increasing number of adults with AS enroll in natural history studies, we will learn more about these individuals and provide them with the most appropriate medical care. For example, although the majority of the participants in our cohort who were ambulatory had an ataxic or broad-based gait, we could assess this trait in only 41/92 (45%) of the participants because the rest had yet to develop the ability to walk. There is also a need to study greater numbers of individuals with the non-deletion subclasses to better understand the genotype–phenotype correlations in AS, as illustrated by the limited precision of our analyses.
In summary, we hope that this broad overview of the clinical features of AS will assist general pediatricians and pediatric sub-specialists in identifying young children with AS, leading to earlier diagnoses and interventions. Having a definitive diagnosis could help these children obtain more intensive services through Early Intervention and other programs, thereby maximizing their developmental potential. It would also enable their parents and other family members to receive appropriate genetic counseling for future pregnancies. Healthcare providers should also be aware that children with AS due to UPD/imprinting defects are at risk of obesity and its complications.