4. Discussion
To the best of our knowledge, this is the first paper to examine NIV adherence and efficacy in children with DS with a control population. Our study demonstrates that satisfactory NIV adherence and efficacy rates are achievable in children with DS compared with a non-DS cohort. Overall, acceptable leakage levels and NIV efficacy were demonstrated across different age groups and cohorts in our study. Children with DS with known congenital cardiac disease who underwent cardiac surgery had decreased NIV adherence when compared with children who had not undergone cardiac surgery.
There are conflicting reports regarding NIV adherence in children with DS throughout the literature. Previous studies have shown that poor compliance and refusal are significantly more common in children with DS due to behavioural and cognitive impairment 31. We collected adherence data for twelve months in both DS and non-DS cohorts and conversely to previous studies, we have shown that children with DS showed increased usage relative to their non-DS counterparts, with the DS cohort using NIV on more than 75% of nights, with > 4hrs use on more than 50% of nights. These data are in keeping with that observed by Dudoignon et al, where good compliance was observed in the majority of children with DS 20. Similarly, Trucco et al showed satisfactory NIV adherence within a paediatric DS population, with particular emphasis being placed on the initiation period as a key factor in overall and enduring compliance23. The importance of early optimisation and compliance is a sentiment echoed in the literature, Mansell et al have highlighted the important role of early domiciliary ventilator downloads to identify modifiable risk factors, such as adherence, as part of a treatment protocol to optimise and personalise NIV treatment32.
Having established satisfactory compliance within our DS paediatric cohort, we next looked at another key factor of therapeutic success, effective NIV delivery and system leakage. To date, there have been no studies investigating NIV leakage in a paediatric population; furthermore the level of clinical ”tolerabilty” of leaks is not well established in home NIV 33. Despite this, it is widely known that leakage will significantly effect NIV performance34. In this study we identified that there was significantly greater leak in the DS cohort relative to their non-DS counterparts and this is particularly evident in those aged >5yrs.
Upper airway surgery is often the first-line treatment of OSA in children with DS 18. Although significant clinical improvements are seen post-adenotonsillectomy, the prevalence of residual OSA is reported to be between 47% and 70%22,35. In this study there was no difference in NIV adherence in those with residual OSA relative to those with no history of upper airway surgery. Furthermore, leak and time at excess leak was more pronounced in those with DS that displayed residual OSA post ENT surgery. This finding may have been due to craniofacial abnormalities which are well described in children with DS. Reducing leak at the patient-ventilator interface is important, not only to further promote comfort and compliance but to enhance NIV efficiency. However, a degree of mask leak is acceptable and the levels of leakage recorded across both cohorts was low overall therefore was not likely to be clinical significant.
Children with DS are more susceptible to developing the adverse sequelae of OSA largely due to the myriad of co-morbidities 27, and as such The American Academy of Pediatrics (AAP) recommend a screening PSG for OSA be carried out before the age of 4 in all children with DS, a sentiment echoed by the Royal College of Paediatrics and Child Health 36-38. Up to 60% of children with DS have a history of congenital and structural heart disease increasing their predisposition to persistent, development and recurrence of pulmonary hypertension if OSA goes untreated or undertreated14,15,26. Congenital heart disease has previously been suggested to undermine the efficacy of upper airway surgical outcome22. Those with a history of congenital heart disease and particularly those previously requiring cardiac surgery within the DS population have an increased risk of the cardiac sequelae of untreated OSA.
In this study, seventy-seven percent of this DS cohort has a past medical history that includes cardiac disease. This higher than normal percentage is likely due to the fact that our centre is the national referral centre for congenital heart disease. We have shown that those with a prior history of cardiac surgery have significantly poorer NIV adherence compared with their non-surgical counterparts.
To date, this is the first study of its kind that has identified poor NIV adherence within a DS cohort with prior cardiac surgery. Similarly, previous studies have noted poor therapeutic compliance within an adolescent and young adult cohort following heart or heart and lung transplant 39. In this study by Wray et al, unintentional and intentional non-adherence was assessed, factors surrounding intentional non-compliance included perceived difficulty of the therapeutic intervention, as well as, perceived benefit39. Treatment compliance in long-term paediatric conditions is a complex issue, it must be looked at in the context of the issue at hand, past medical and surgical history of the child as well as the interaction of the family with healthcare and practitioners40. Parental non-compliance commonly arises from the emotional stresses associated with the recommended treatment; NIV initiation can be difficult and emotionally taxing on caregivers. Parents make continued attempts to balance concerns about the treatment, such as perceived effectiveness and the perceived long-term threat of the condition 41. Impaired quality of life, as well as psychosocial issues including, adolescence, racial/ethnic minority status, and presence of mental health issues have been identified as factors associated with non-compliance 42. Familial predictors associated with non-adherence problems included single-parent households, lower socioeconomic status, lower family cohesion, presence of family conflict, and poor family communication42. It could be postulated that the perceived threat of OSA is not as great as congenital heart disease and the therapeutic intervention is viewed as labour intensive particularly in the context of milder threat. These data highlight that increased surveillance and follow-up may be required in children with DS on NIV who have undergone prior cardiac surgery.
The major strength of this study is the inclusion of a control group. Previous similar studies have reported only on DS cohorts and the addition of a non-DS cohort adds significant weight to our findings. The authors are aware of the limitations of this study. Firstly, this is a single-centre retrospective study. The foremost scientific limitation of this study was the restricted availability of follow-up PSG data once NIV had been initiated, PSG data was not available to support and compliment the at home recordings at the investigated time-points due to limited resources. However the use of downloadable data from these NIV machines is emerging across the literature demonstrating its importance within clinical decision-making 32,43. Home NIV data is becoming a key source of reliable information; it is not subject to the recall bias of previous usage questionnaires and can inform the clinician while helping to optimise patient care in a personalised fashion 32,43.
In conclusion, this study confirms satisfactory NIV adherence in children with DS compared with a non-DS population. Particular attention should be given to children with DS and co-existing congenital heart disease in respect to NIV adherence. As we enter an age of personalised medicine, future studies should examine ways of improving NIV adherence and efficacy further in children with personalised NIV masks that incorporate 3D solutions to reduce system leak and discomfort44. Our study demonstrates that NIV delivery in complex children in a tertiary paediatric setting can be successful and that good adherence and efficacy levels can be achieved.