Only a quarter of children with moderate-to-severe OSA exhibited hypertrophic tonsils in a new study, raising the possibility that other factors might play a more significant role than previously believed.

Interview by Alyx Arnett

While diagnosing and treating obstructive sleep apnea (OSA) in children has primarily centered around subjective tools like the Pediatric Sleep Questionnaire (PSQ) and surgical methods such as adenotonsillectomy, a cross-sectional study conducted in Akureyri, Iceland, has brought to light gaps in these practices.

Involving 373 children aged 4-8 years, the study used the SleepImage home sleep test (HST) to evaluate the accuracy of the PSQ and examine the otorhinolaryngologic status of children diagnosed with moderate-to-severe OSA by grading tonsil and adenoid size. 

Magnús I. Birkisson, MD

The research, co-authored by Magnús I. Birkisson, MD, was designed to investigate the underlying causes of OSA and the efficacy of treatment options, as well as offer treatment based on the results of both the sleep recordings and clinical evaluation. “With adenotonsillar hypertrophy being known as one of the major risk factors of OSA, there was an emphasis on the otorhinolaryngologic evaluation,” says Birkisson.

The results indicated a notable discrepancy between PSQ responses and HST findings. Furthermore, the study observed the continued presence of OSA symptoms in a significant proportion of children who had previously undergone adenotonsillectomy.

Birkisson discussed the study, presented at the 2023 World Sleep Congress in Rio de Janeiro, with Sleep Review over email. Birkisson was the recipient of the Young Investigator’s Award.

What surprised you in this research? 

When comparing the size grading of tonsils and the adenoid size grading between moderate and severe OSA, the average grade in both categories is practically the same, with moderate OSA even having a slightly higher grade on average. This underlines that clinical evaluation is not sufficient when evaluating children with OSA and emphasizes the importance of implementation of therapy tracing when evaluating and treating OSA in young children.

How do the results contribute to our knowledge of diagnosing OSA in children?

Our findings suggest that subjective evaluation using pediatric sleep questionnaires tends to overestimate children without OSA when confirmed with HST and underestimate children with mild, moderate, and severe OSA. It can be deduced that a more objective way of determining the likelihood of OSA is needed to go alongside an examination by a specialist, especially when deciding if a child should undergo surgery.

How might the findings influence treatment protocols for pediatric sleep apnea?

Our results suggest that adenotonsillectomy surgeries are not as effective, at least when it comes to treating OSA, as was previously assumed. The answers from the general health questionnaires show that a quarter of the children measured with OSA and treated with adenotonsillectomy are still experiencing OSA. Furthermore, we evaluated only a quarter of the children with moderate/severe OSA to have hypertrophic tonsils and/or adenoids. This begs the question of other causes of OSA being a greater factor than was previously suspected and, therefore, a potential need for treatment protocols to be modified. 

What recommendations would you make for clinicians relying on subjective assessments like the PSQ to diagnose OSA in children?

The study is still ongoing as there are still children being re-evaluated with a new sleep measurement after treatment/intervention based on clinical findings during examination, so I will reserve my recommendations until the study has been concluded and the results have been published. For now, I will recommend clinicians acquire a way to objectively evaluate children if there is a suspicion of OSA, especially before subjecting them to surgical interventions. 

What are the next steps in your research?

The next steps are to finalize the study with the remaining sleep measurements and complete the dataset. When the dataset is complete, we plan to publish our results with a research paper. There are plans for several future research projects based on the original study and the data collected, with the potential for increased theoretical knowledge and innovation in the field.

Photo 176026505 © Alona Siniehina |