The concept that nonsurgical maxillary expansion can be successful in adults has raised questions in the literature. Overall, the consensus is that, once patients are out of their teens, that type of expansion is no longer feasible, and instead, surgically assisted rapid maxillary expansion is necessary. The purpose of this article is to challenge this commonly accepted orthodontic paradigm.

Let us assume that the first consult at your office is a 30-year-old woman with bilateral posterior and anterior crossbites with crowding of the maxillary left lateral incisor and edge-to-edge occlusion of the right lateral incisor. You estimate transarch deficiencies of 9 mm at the first premolars and 7.5 mm at the first molars. You suggest surgically assisted rapid maxillary expansion to correct the posterior occlusion and to gain arch length to correct the crossbite of the maxillary left lateral incisor. Surprised by the suggestion of surgery, the patient asks whether you can just correct the displaced lateral incisor. Obviously, she would much prefer that you treat her malocclusion without surgery.

In 1997, I presented a series of 5 cases (including the one just described) on nonsurgical maxillary alveolar expansion in adults using the Haas expander. In a commentary on these 5 cases and in a letter to the editor, it was suggested that these 5 patients might have been selected for the excellence of the results and the lack of true skeletal deficiency.

To move beyond the anecdotal case series, I collected the records of every adult patient in my office who had nonsurgical expansion with the Haas expander from 1978 to 1995. To this group of 29 subjects, I added 18 patients from the office of Dr Andrew Haas. This combined group of 47 adults (mean age, 30 years) became the adult nonsurgical expansion group in our study. We also looked at 2 additional groups: 47 children (mean age, 9.5 years) who underwent nonsurgical rapid maxillary expansion and 52 adult orthodontic patients (mean age, 33 years) who did not require expansion and served as the controls. Lagravère et al, in their systematic review of long-term dental arch changes after rapid maxillary expansion, found only 4 articles that met their stringent evidence-based criteria, and ours was one of them.

The Haas expansion screws were activated for both expansion groups once daily until the palatal cusps were almost in buccal crossbite. The expander was maintained for 12 weeks and then replaced with a removable retainer. (In adults, I now activate no more frequently than every other day and, often, every third to fifth day. Thus, for adult patients, the technique is better described as slow maxillary expansion rather than rapid maxillary expansion.) We studied the following phenomena: efficacy, long-term stability, and complications. The efficacy of adult nonsurgical maxillary expansion was excellent: averages of 4.6 mm at the first molars and 4.7 mm at the first premolars, with no statistical difference between the adult and child expansion groups. The adult nonsurgical expansion also compared favorably with the results of child and adolescent groups reported in the literature.

However, the nature of the expansion in adults is different. Trimming and then photocopying the backs of study models at the level of the first molar buccal groove allowed us to superimpose pretreatment and posttreatment contour tracings of the models. This analysis clearly showed that adult expansion was the result of displacement of the alveolar process, which carried the teeth buccally. The displacement generally started at the apical third to the midlevel of the palatal vault. In children, about 50% of the expansion occurred at the midpalatal suture and the remaining 50% by displacement of the dental alveolar complex. This finding challenges the assumption of many orthodontists that all or most of the expansion in children occurs at the midpalatal suture.

Interestingly, this assumption had previously been challenged in at least 2 other studies. Krebs, in the late 1950s and early 1960s, using metallic bone markers, estimated that only 50% of the expansion after rapid maxillary expansion was skeletal. More recently, Iseri and Özsoy also used metallic bone markers and confirmed these findings and noted that only 40% of maxillary expansion in adolescents (average age, 14.5 years) was in the maxillary bone. I quote these authors: “The above findings also indicated that the widening of the maxilla was mainly achieved with the expansion of the maxillary dentoalveolar structures.” In our study, the adults were mostly in their 30s; therefore, almost all the correction was due to expansion of the dentoalveolar complex.

Garrett et al used cone-beam computer tomography to the skeletal effects to the maxilla after rapid maxillary expansion on adolescents (average age, 13.8 years). They concluded that, at the level of the first premolars, 55% of the expansion was skeletal; and, at the level of the first molars, only 38% was skeletal; the remainder was dentoalveolar. Recently, Kartalian et al used cone-beam computerized tomography to evaluate the dentoskeletal complex before and after rapid maxillary expansion in growing subjects. They concluded that approximately 5 mm of the increase at the dental level was associated with the 2-mm increase at the skeletal level—ie, a 40% skeletal contribution.

For those who still doubt the possibility of significant nonsurgical maxillary expansion in adult patients, I would argue that the evidence that 50% to 60% of expansion in children and adolescents occurs in the alveolus and not at the midline suture is the basis for the success of nonsurgical adult maxillary expansion. If alveolar displacement did not occur, cases such as those I presented in my article who had 8 to 10 mm of expansion would have had perforation of the thin buccal plate of the posterior alveolus. The palatal acrylic bodies of the Haas expander are critical to the orthopedic displacement of the alveolus.

Data on long-term stability are difficult to obtain because patients must be recalled for records many years after the completion of their treatment. In our study, 21 subjects who had discontinued maxillary retention for a mean of 5.9 years after 5 years with nighttime retainer wear were reevaluated. Not 1 molar or premolar relapsed into crossbite. The posttreatment decreases in molar and premolar transarch widths were 0.5 to 0.6 mm, and some of these decreases can be accounted for by the overcorrection retained at the end of treatment. The literature does report cases of relapse after expansion in children and adolescents. Possible explanations for such relapses are the use of dentally based expanders (hyrax) vs our use of the Haas expander with its palatal acrylic bodies, use of removable instead of fixed expanders, failure to significantly overexpand, and too short a retention period after expansion.