Peter J. Koltai MD, FACS, FAAP

Current Research Interests

It has been well-recognized that tonsillectomy and adenoidectomy is the primary treatment for pediatric obstructive sleep disordered breathing. However, it is also recognized that approximately 15-20% of the children will continue to have problems with obstructive sleep disordered breathing, despite having their tonsils and adenoids out. The primary problem of sleep apnea in children who have had their tonsils and adenoids out was identifying the site of obstruction. Since fiberoptic laryngoscopy is a routine part of our office exam in our evaluation of children with sleep apnea, it seemed like a natural evolutionary step to perform a similar type of examination while the children are under anesthesia. Clearly an anesthetic induced sleep is not real sleep; on the other hand, it is about the closest model to real sleep that we have that still allows us to perform fiberoptic examination. Based on this insight, we began to offer sleep endoscopy to the parents of children who we were seeing who had failed tonsillectomy and adenoidectomy and had persistent sleep apnea. What we found out during sleep endoscopy was that there can be multiple levels of obstruction. However, the two most consistent sites of obstruction were due to enlarged lingual tonsils, where the lingual tonsils caused a prolapse of the epiglottis up against the posterior pharyngeal wall during recumbent sleep and from an occult form of laryngomalacia, where the soft tissues of the posterior glottis prolapsed into the laryngeal introitus on inspiratory effort during sleep.

Lingual tonsillar hypertrophy is recognized as a cause of obstructive sleep apnea in children, however, the form that was typically seen prior to our current work was in children who had grossly enlarged lingual tonsils, easily seen on an office exam. What we were seeing on our sleep endoscopies was a more subtle form of lingual tonsillar hypertrophy which was obvious only on the sleep endoscopy but was not readily discernible on fiberoptic laryngoscopy in the office.

Similarly, while laryngomalacia is an airway problem that is well recognized in new born infants, it has not been previously demonstrated to be a cause of sleep apnea in older children, especially without any daytime manifestation of the obstruction. We now have many video recordings demonstrating the phenomenon in older children. Our experience with infant laryngomalacia provided a means of treating this form of obstruction.

We also observed other types of obstruction on sleep endoscopy not related to lingual tonsillar hypertrophy or to occult laryngomalacia,. These obstructions were from hypotonia, due to excessive relaxation of the pharyngeal musculature during sleep, obesity with a marked narrowing of the entire oropharyngeal space probably as a consequence of fatty deposition in the surrounding musculature.

We have been studying the utility of high pressure balloons for radial dilatation of the stenotic larynx and trachea, avoiding the shearing trauma to the airway tissues caused by conventional dilation techniques. Our clinical experience of the last 3 years with the off label use of balloons in the airway has confirmed the validity of of this technology. Acclarent asked me to join their Scientific Advisory Board and begin development with them of a line of airway balloons (5mm, 7mm, 10mm, and 14mm) which were FDA approved in 2009.