Customize your JAMA Network experience by selecting one or more topics from the list below.
Hypothesis: The proper correction of pediatric upper airway obstruction may prevent dentofacial abnormalities.
The concept that obstruction from tonsillar and/or adenoidal hypertrophy adversely effects dentofacial development was put forth as early as the 1840s. More than 150 years have passed, and the issue continues to be controversial. A review of the current literature reveals both passionate support of and opposition to the hypothesis.
The dentofacial abnormalities that have been associated with chronic nasorespiratory obstruction are many.1 Those most commonly cited are an elongated face, open-mouth posture, protrusive maxillary incisors, narrow maxillary arch, high-arched palate, and angle class II malocclusion. The term adenoid face encompasses most of these features. One way to restate the hypothesis would be that the term adenoid face describes an abnormality that is accurately named and appropriately correctable.
William D. Clark, MD, DDS
Contention of any proposed individual step in the sequence of events leading to the final hypothesis results in rejection of the hypothesis altogether. First, it must be accepted that upper airway obstruction from tonsillar and/or adenoid hypertrophy can and does produce abnormal facial posture. Second, one has to accept that abnormal facial posture can and does induce abnormal dentofacial growth and development. Finally, one must accept that relieving airway obstruction by adenoidectomy and/or tonsillectomy results in normalized breathing and normalized facial posture and avoids or stimulates correction of abnormal dentofacial growth and development. The last of the 3 theories seems to be the most difficult to prove.
A major confounding factor in addressing this hypothesis is poor availability of the early articles that were quoted by later authors. When one cannot critically review an article, it is dangerous to fully accept another’s interpretation of that work. In some cases, multiple generations of authors may have repeated the interpretations of earlier authors. Even directly contacting authors of recent articles addressing these issues failed to produce copies of original articles. Therefore, with reluctance, some weight has been given to the interpretation of others.
Critical review of the literature in this area also reveals problems with definitions of terms as basic as mouth breathing and nasal obstruction. Some physicians declare children with open-mouth facial postures to be nasally obstructed mouth breathers without further evidence, whereas others point out the fallacy of such assumptions.
Having the advantage of both dental and otolaryngologic training, I can see the strength of arguments that adenoid obstruction can affect dentofacial structure. It makes sense that adenoid hypertrophy can cause nasal obstruction leading to mouth breathing and that the open-mouth posture pressures the lateral maxilla, impeding expansive maxillary growth. This, in turn, can result in a narrow maxillary dental arch and a high-arched palate. After all, it is easy to prove that mild forces over an extended period move teeth through bone. Orthodontists regularly expand the maxillae of patients with cleft palates by using such forces. The forces produced by an open-mouth posture have only to restrict expansive growth, seemingly an easier task. Further support comes from experience in seeing and treating children who fit the theoretic mold with apparent precision. It is more difficult to accept that relief of nasal obstruction will produce normal facial posture and prevent abnormalities from developing or reverse already-developed abnormalities. Nevertheless, I have long held the opinion that my hypothesis is true, at least as it relates to airway obstruction secondary to adenoidal hypertrophy. Whether tonsillar hypertrophy produces similar effects on dentofacial development seems quite another question and does not have a similar status of acceptance with me. Thus, I had some biases going into the literature research for this article.
There is a striking lack of scientific studies (as opposed to anecdotal reports), and the scientific studies are for the most part poorly designed. This makes critical evaluation of this hypothesis difficult.
At least 2 studies of identical twins tend to support nasal obstruction as the cause of dentofacial deformities and the relief of such obstruction as leading toward normalization.2,3 In Krause’s study,3 1 monozygotic twin had isolated nasal trauma at age 9 years. Bilateral nasal obstruction developed, and changes in dentofacial development became apparent with time. The photographs of the twins at age 12 years show a marked contrast in their facial configuration. The child with nasal obstruction showed the classic features of adenoid face, whereas her sister did not. Improvement was reported to have occurred after correction of the nasal obstruction, but cephalometric data and specific details of improvement were not provided.
The body of scientific work that best supports my hypothesis is reported in a series of articles from the Craniofacial Anomalies Clinic, School of Dentistry, University of California, San Francisco, with Harvold, Miller, and Vargervik4,5 as authors. These studies were conducted on primates and were summarized by Vargervik and Harvold4 in 1987 in the otolaryngology literature. Critics of these studies point out the danger of applying findings of studies conducted on animals to humans and the lack of uniform responses of the study subjects to the induced changes.
One pertinent primate study involved young rhesus monkeys. The study group had near-total nasal obstruction (total obstruction to inhalation, partial to expiration) produced by temporary soft silicone implants. The study animals and paired controls were monitored for facial growth changes with cephalograms both during the period of nasal obstruction and after relief thereof. The study animals developed the classic changes associated with long-face syndrome (increased vertical height of the face and increased angles of the palatal plane, occlusal plane, mandibular plane, and ramus) while their nasal cavities were obstructed, and some showed a tendency toward partial normalization postobstruction. Facial height was an important measurement that did not improve after relief from obstruction.
In a related experiment from the same group of authors,5 electromyographic recordings from dorsal tongue fibers, suprahyoid muscles, and upper lip elevators were studied. Tonicity as measured by electromyography was increased in these muscles during the period of nasal obstruction. Tonicity normalized after relief of nasal obstruction in all studied muscles except for those of the genioglossus and geniohyoid.
Linder-Aronson6 did a 5-year study on patients who underwent adenoidectomies for nasal obstruction. Among those in this group who were relieved of nasal obstruction (mean age, 7.9 years), he found improvements in upper and lower incisor inclination, upper arch width, sagittal depth of the nasopharynx, and anterior facial height. A matched group of subjects who had not undergone adenoidectomies (mean age, 7.5 years) served as controls. Most of the improvement was seen in the first year after surgery and remained stable for up to 5 years postoperatively.
One anecdotal study7 describes 3 children with untreated bilateral choanal atresia who are said to have had normal dentofacial development; this report is difficult to ignore. This is one of several older articles often quoted but not readily available for critical review.
Among the nonanecdotal articles is the oft-quoted study by Humphreys and Leighton,8 who are said to have evaluated 1033 children between the ages of 2 and 5½ years without seeing a significant difference in the incidence of malocclusion among mouth breathers. The authors are said to have noted that half of those with an open-mouth posture were breathing at least partially through their nasal cavities. A review of the article revealed that the authors looked at many factors possibly related to the development of class II and III malocclusions. Their study comprised 500 “postnormal” subjects (angle, class II), 500 “normal” subjects (angle, class I), and 33 “prenormal” subjects (angle, class III) culled from 3380 children seen for routine medical inspection and advice. They found no correlation between enlarged tonsils and malocclusion but noted that more children with class II malocclusions “leave their mouths open during either the day or the night” than children with normal molar relationships.8 They go on to say that, even so, those “children do not mouth-breathe appreciably more” than children without a class II malocclusion. They found only 6 (3 each in the class I and class II groups) of the 1033 children to be “breathing solely through the mouth.” Interestingly, only 1 paragraph was devoted to the issue of mouth breathing in this article of more than 12 pages. In fact, they devoted equal space to the defense of their hypothesis that women with class II malocclusions have “an advantage in the quest for a mate” in that the associated abnormal maxillary incisor inclination “gives the appearance of a smile.”
I found no convincing evidence in the literature to support tonsillar hypertrophy as a major etiologic factor in producing the dentofacial abnormalities in question. The study that came closest to making a case for this part of the hypothesis was by Harvold et al.9 Again, juvenile rhesus monkeys were studied. An acrylic wedge was placed in the posterior hard palate to simulate tonsillar hypertrophy, and the effects were monitored. All expected changes occurred (increase in anterior face height and anterior open bite and narrowing of the maxillary dental arch). Despite these results, the validity of the model for accurately simulating tonsillar hypertrophy seems suspect in that the acrylic wedges differed in consistency and location to palatine tonsils.
Hultcrantz et al10 studied children who underwent tonsillectomy without adenoidectomy for obstructive sleep apnea syndrome. They found a greater-than-expected incidence of malocclusion in this group compared with a random sample. The 22 children were followed up for 2 years, and 50% to 77% of the 3 types of malocclusion (anterior open bite, buccal crossbite, and anterior crossbite) resolved. A major weakness of the study was lack of nonsurgical controls.
Taken literally, my hypothesis cannot be completely supported. Certainly, the case for promoting tonsillar hypertrophy as a major etiologic factor seems weak. However, the preponderance of the evidence suggests that the hypothesis as it relates to adenoidal hypertrophy has merit. The amended first subhypothesis seems noncontroversial: airway obstruction from adenoid hypertrophy often leads to altered facial posturing. The second subhypothesis seems reasonable, although challenged by some in the literature: the resulting facial posturing produces forces that result in altered dentofacial development. The amended third subhypothesis seems to have partial scientific support based on the work by Vargervik and Harvold4 and Linder-Aronson6: relief of airway obstruction by adenoidectomy, when performed early enough, prevents further dentofacial deformity and may result in changes toward normalization.
Probably, both environmental (eg, nasorespiratory obstruction) and genetic factors are important in the development of the dentofacial abnormalities discussed herein. As Cooper,1 comparing the causative effects of environmental influences with pure genetic etiology, stated in his exhaustive review of this subject, “The truth probably lies somewhere between these two extreme positions.”
Correspondence: William D. Clark, MD, DDS, 1735 W Hibiscus Blvd, Suite 100, Melbourne, FL 32901 (email@example.com or firstname.lastname@example.org).
Submitted for Publication: April 29, 2005; accepted April 29, 2005.
Financial Disclosure: None.
Clark WD. Preventing Dentofacial Abnormalities With the Proper Correction of Pediatric Upper Airway Obstruction. Arch Otolaryngol Head Neck Surg. 2005;131(10):916–918. doi:10.1001/archotol.131.10.916
Browse and subscribe to JAMA Network podcasts!
Create a personal account or sign in to: