Neurodevelopmental disorders in children with cleft lip and palate: a systematic review

The findings of this review indicate a significant increase in prevalence of ASD among individuals with an OFC when compared globally, or to control cohorts. ADHD painted a less clear picture, with rates similar to global prevalence figures, although matched control populations did show an increase among the OFCs cohort. The increase was observed across both non-syndromic and syndromic populations (to a much greater extent in the syndromic OFC studies), however variations in methodology warrants mention. Larger national or multi-national cohort studies are needed with strict diagnostic processes to ascertain the true prevalence in this population. This is the first systematic review to assess for an association between both syndromic and non-syndromic clefts and both ADHD and ASD, thus there is no prior consensus to compare with.

Variations in the criteria used for ASD diagnosis can make analysis difficult (Table 2). For example, the DSM-IV assesses patients across three categories: (1) impaired social interaction, (2) impaired social communication, (3) restricted behaviour pattern, whereas the DSM-5 assesses across only two: (1) Impairments in social communication and social reciprocity, (2) the presence of restricted interests and repetitive behaviours [21, 22]. There has also been concerns raised about the DSM-5 criteria in numerous aspects including diagnosis in infancy and adulthood. While both these criteria were the most recent at the time of publication, it is possible that differing proportions of patients would have met the threshold for diagnosis, should the ASD diagnostic criteria be standardised across all studies. A similar discrepancy may arise when comparing ICD-9 & 10 criteria to DSM-IV & 5 criteria. Again, assuming standardised criteria across studies, it is possible that prevalence may change in certain populations. The same issue arises when comparing ADHD diagnosis; DSM-5, for example, distinguishes different diagnostic criteria depending on the patients age, which DSM-IV does not. Similarly, DSM-5 requires some symptoms to have been present before age 12 years, compared to before age 7 years in DSM-IV [23].

Heterogeneity of diagnostic methods also complicates analysis of the rates. The ADOS and ADI-R differ significantly in the fact that ADOS involves a direct interactive interview with the child, compared to the retrospective interview with the child’s parent or carer used in the ADI-R. It has been suggested that the ADI-R plays a less significant role in ASD diagnosis than behavioural observation (especially in adolescents and adults); instead, although time consuming, a combination of both should be used to guide complete diagnosis [24]. Unlike with ASD, ADHD has a less clear cut ‘gold standard’ for clinical diagnosis, and instead often is made up of a mixture of clinical assessments, interviews and diagnostic scoring criteria [25]. It is therefore difficult to be critical of the diagnostic methods used in the included studies, although again, methodological differences are clear and may play some role in explaining the discrepancies between rates. Even with strict diagnostic criteria, constricted ranges of facial affect and speech difficulties leading to communicative impairment in ASD may be difficult to distinguish from those caused by clefts or other orofacial defects. Feeding difficulties due to clefts have previously been misdiagnosed as eating disorders [12] - it therefore stands that a similar misdiagnosis may occur for ASD or ADHD.

Across all studies, age range varied dramatically. Although there is relative diagnostic stability for ASD from as young as 14 months, there is also a high rate of false negatives at younger ages (23.8% receiving a diagnosis at 3–4 years of age had a missed diagnosis at their first assessment) [26]. Thus, it may be difficult to draw confident conclusions from studies with larger age ranges and particularly younger participants, such as Fine et al.. who reported a range from 2 to 12 years [14] and Niklasson et al. with a range from 1 to 35 years [18]. Of note, those reporting larger age ranges noted similar ASD and ADHD rates in the older proportion of their populations compared to the younger ones.

Various potential reasons for an increased risk of ASD/ADHD have been identified in the literature. The use of general anaesthesia in cleft patients has been suggested to play some role, with research suggesting early exposure was associated with significantly increased risk of developing ASD (1.65% vs. 0.98% in the control, P < 0.01) [27]. Another potential reason for the association is the early use of antibiotics accompanying the surgeries required by patients with OFCs. A recent cohort study found slightly elevated risk among those receiving antibiotics before 2 months of age (Adjusted HR:1.23, 95% CI 1.17 to 1.28) and with antibiotic use over 15 days [28]. There is some evidence to suggest that risk factors such as the use of alcohol and smoking during pregnancy increase the risk of both ADHD and OFCs in children, although data around this remains controversial [29, 30]. Antshel et al. also found neuroanatomic variations (including an enlarged amygdala) reported in idiopathic ASD that were present in children with both VCFS and ASD [17]. Whilst more recent research is suggestive of variations across broader, interconnected neuroanatomic systems, structural imaging of the brain may still aid future assessment and management of ASD [31]. Each of these factors could provide further avenues for research in determining aetiology, as well as enhancing earlier identification of neurodevelopmental disease and subsequent management to improve patient quality of life. It is possible that a combination of several of these risks with other biopsychosocial influences contribute to the development of ASD and ADHD, thus future clinical interventions may include preventative measures and an emphasis on patient education and awareness for early recognition.