We thank Dr Antosz for his interest in our article, “Three-dimensional upper-airway changes with maxillomandibular advancement for obstructive sleep apnea treatment.” We hope we can clarify the issues for him.
Analysis of the airway by cone-beam computed tomography (CBCT) is of great value in assessing the functional airway, based on our considerable experience in sleep medicine, surgery, and the literature. The CBCT study is indeed static; however, it provides important information in evaluating the airway in patients with obstructive sleep apnea (OSA). The total volume of the airway is usually smaller in patients with OSA, although the most important measurement is the smallest cross-sectional area because this is where the obstruction occurs. This may be at just 1 point or multiple points depending on the overall airway shape. Small cross-sectional areas have been shown to be correlated to OSA. Thus, to be clear, CBCT analyses are of value in assessing the functional airway, but the volume is only one measurement taken into consideration. Of course, CBCT is not the only method by which we evaluate the functional airway. OSA must be confirmed by a polysomnogram and other modalities, such as direct nasendoscopy and, occasionally, sleep endoscopy. In surgical correction of the airway, it is imperative to reveal the extent and significance of the obstruction, for which purpose CBCT is invaluable.
The study by Alsufyani et al that Dr Antosz refers to actually goes on to say that it is difficult to generate a strong conclusion regarding the validity and reliability of CBCT-generated 3-dimensional models. That is to say, few articles met the authors’ search criteria in a review of the literature regarding studies that used automatic or semiautomatic modeling of the airway. This is somewhat different than the implication Dr Antosz leaves us with. The gold standard in this study was manual segmentation of the airway, which Alsufyani et al thought was accurate but that has inherent flaws. The study did not address the validity of the CBCT scan itself. The mathematics involved in the production, the segmentation of computed tomography data, and the internal protocols of each cone-beam manufacturer in this regard are beyond the scope of this response. Alsufyani et al also quoted from the article by Schendel and Hatcher that validated an automatic airway technique that was used in our study. Their only criticism of this article was that the validation was done on a phantom and not the human airway. We have used this system clinically since 2010 and found it to be valid and accurate in evaluating the human airway. A standard protocol when using CBCT must be used to eliminate variables that can alter the result in discrepancies, as Dr Antosz mentions. Recently, Uesugi et al showed that differences in the airway volume range from −0.4 to 0.1 cm 3 if a standard protocol is followed. The position of the patient during scanning can affect the measurements; airway dimensions can vary up to 30% in those with OSA depending on whether the patient is seated or lying down. So, not only must the protocol be standardized but also the same machine should be used; this is what our study’s experimental design was based on.
It is equally important to understand that our study measured not just the volume of the airway but the cross-sectional area and linear measurements, all recommended by Alsufyani et al. Changes in all of these variables were correlated to improvement in sleep apnea symptoms and the polysomnogram, thus justifying the conclusions.
In a previous study of 1300 subjects from ages 6 to 60 years, we found each airway to be consistent from year to year, increasing with growth and then decreasing in later life. What is quite variable is the size of different airways in subjects of the same age. In spite of this, statistically significant data could be obtained. An anecdotal case, as mentioned by Dr Antosz, has no place in evidence-based treatment planning, and it is impossible to comment on this without knowing the specifics.
The premise of evidence-based medicine is often confusing to many. Simply, it means “integrating individual clinical experience with the best available clinical evidence from systematic research.” In addition, the patient’s desires must be considered. If anyone is unclear on this, I suggest they read the article by Swanson et al entitled “How to practice evidence based medicine.” Evidence-based medicine is not restricted to randomized trials and meta-analyses, but finding the best evidence to answer a clinical question. Scientifically sound research occurs at different levels of significance and should be evaluated as such. The American Journal of Plastic Surgery has instituted an evidence-based rating system for articles in this regard that helps the clinician rate the importance of each article. The idea that if the best-available evidence is flawed then the whole premise goes out the window is akin to throwing the baby out with the bath water.
That this study seems to indirectly validate abusing CBCT imaging is ludicrous. We take a CBCT scan only when the clinical examination warrants it, in preparation for surgery, and as a postsurgery follow-up. This certainly doesn’t amount to abuse, nor do we advocate the routine use of the CBCT as a screening tool. We are pleased that the author has treated patients with jaw surgery for OSA and believes in the treatment. Other issues the author mentions don’t really apply to this article but to orthodontic practice in general and were not part of our study.
Lastly, nowhere in our article do we advocate surgery based solely on a CBCT scan, but we clearly indicated that it is one factor that we consider in treatment planning.