Transverse maxillary deficiency is a skeletal deformity characterized by a constricted maxilla, showing unilateral or bilateral cross-bite and crowded teeth [19]. SARME is the recognized treatment approach for that situation. Its primary goal is to achieve skeletal expansion, rather than dental expansion, and to minimize dental tipping by separating the midpalatal and lateral maxillary sutures [20]. This surgical technique does not only determine a palatal expansion, but also causes craniofacial structural changes, such as the increase in nasal cavity width and nasal volume [21,22,23,24].

In fact, even if SARME is an effective and relatively safe technique [25, 26], many soft and hard tissue changes, such as nasal soft tissue changes, may occur after the surgical procedure and distraction phase [27]. It is important to analyze the nasal region in particular, being one of the most important aesthetic parts of the face [28, 29]. Several studies have reported that SARME affects the nasal area, in particular causing alar base widening, which is aesthetically undesirable in patients with a wide nose [14,15,16,17,18]. The increase in nasal volume after expansion generally determines a functional improvement in respiration. Patients with maxillary transverse deficiency, who underwent SARME, experienced a subjective improvement or did not have worsening of nasal obstruction after follow-up for 6 months [30]. However, to our knowledge, there are no evidences in the literature that correlate NSD to SARME as a possible postoperative outcome.

For measurements, after standardization and orientation of the tomography, we chose the level of the inferior turbinate, because this is above the Le Fort 1 osteotomy and at the same time it is a relatively caudal portion of the nasal fossa, where there would be lower resistance to the deviating forces, as compared to the higher, more cranial portions of the septum, which are anchored to the skull base. Furthermore, the head of the inferior turbinate, its tail and midpoint are easily identifiable landmarks, so the method is reproducible, independently of the operator.

There were statistically significant variations of the bony septum position. Even if there was not statistical difference for the absolute wall-septum distance values, for all the three studied landmarks the Delta index was significant (Table 1). In particular, in two patients out of 29, we noticed an important deviation of the nasal septum. The first is a 30-year-old woman with a left wall-septum distance of 8.8 mm at the inferior turbinate head, which became 5.2 mm after expansion. The second is a 40-year-old woman with a right wall-septum distance at the inferior turbinate midpoint of 11.1 mm which became 12.6 mm, while the left side went from 9.9 to 8.2 mm. In both cases, the absolute difference in the wall-septum distance after expansion does not show a large variation, but the Delta index was statistically different (5.5 mm in the first case at the turbinate head point and 3.2 mm in the second case at the turbinate midpoint).

Another patient presented the opposite situation. The wall-septum distance increased by almost 3 mm both to the right and to the left sides. It does not mean that the septum has changed, but that the nasal cavity has undergone an expansion process. In fact, despite the big variation in wall-septum distance, the Delta index in this case was not significant.

The examples of those three patients demonstrate that the direct differences in the wall-septum distances, before and after maxillary expansion, are not indicative of NSD, because the nasal walls may undergo alterations after SARME. The Delta between the pre- and postoperative measurements on the one side, compared to the obtained for the other side, is efficient to represent the changes in the bony septal position. The Delta value can be negative or positive, depending on how the septum has moved. All the numbers were considered with a positive sign, to calculate the mean amount of NSD, regardless of the side of the septal deviation.

The average NSD value was 0.79 mm at the point of the lower turbinate head, 0.72 mm at the turbinate midpoint and 0.65 mm at the tail. It is interesting to note that the deviation is greater anteriorly. In most patients, small changes in the bony septum position were not symptomatic. Despite the absence of a clinically detected NSD, small oscillations of the position of bone septum have been found in this sample, without difference between sides. Considering a confidence interval of 95% (IC95), it is possible to assume that SARME, in this study sample, produced an NSD between 0.4 and 1.2 mm, more pronounced at the anterior part of bony septum (Fig. 3).

Standardized septal deviation and confidence interval of 95% (IC95)

In this sample, there was an increase in the mean transverse dimension of the nasal fossa at the level of the inferior turbinate, where measurements were taken, after SARME. In fact, despite this level is above the Le Fort 1 osteotomy, the expansion of the bone segments below the osteotomy line creates a remodeling of the nasal walls that extends higher. The influence of SARME on the nasal cavity is due to the separation of the nasal lateral walls, although only at their inferior portion. It is interesting to notice that the remodeling also affects the bone above the osteotomy line. The increase in the distance between the nasal lateral walls enlarges the cross-sectional area, increasing the nasal volume and facilitating breathing [20]. Warren et al. [31] reported that the nasal volume increased 55% after SARME.

There were no side differences in the deviation of the nasal septum. In fact, at the head of inferior turbinate there were 15 deviations to the left and 14 to the right, at the midpoint 16 to the left and 13 to the right, and at the tail of inferior turbinate 8 to the left and 21 to the right. It is probably a coincidence that most of the posterior deviations were to the right side. In most cases, rather than a real deviation to a side, small subclinical oscillations of the position of the bone septum occurred.

In summary, 27 patients (93.1%) presented minor changes in bony septum position, which were not symptomatic. Despite the absence of a clinically detected NSD, small oscillations in the position of bony septum were found throughout the sample, without side differences. In 2 cases (6.8%), a significant NSD was found. The deviation was clinically manifested and perceived by the patients, who noticed reduction of the air flow at the affected side.

Our method proved to be effective for the study of the positional changes of the bony nasal septum. Deviations of the cartilaginous septum were neither measured nor clinically identified: this is a limitation of our study. Moreover, we have not carried on functional studies of the nose after SARME. Our future purpose is to submit a NOSE questionnaire to those patients and evaluate their nasal septum position by nasal fibroendoscopy, to obtain a subjective evaluation of the nasal symptoms reported by the patients, coupled with an objective study of the nasal fossa.