Interceptive orthodontic treatment should be predictable, short, and efficient aiming to correct malocclusion traits that would affect function, future dental occlusion and a favorable growth of the jaws [14, 24].

Orthopedic maxillary expansion treatments have been used for more than a century to correct transverse maxillary deficiency by exerting forces at the mid-palatal and intermaxillary sutures in growing patients [17]. On the other hand, dentoalveolar expansion is performed applying the force directly to the teeth and producing a lateral displacement of the upper dento-alveolar structures. Several removable or fixed appliances, including aligners, can be used [14,15,16,17].

The purpose of the present retrospective study was to evaluate the effects on the transverse plane of CA in early mixed dentition in a sample of subjects that require dentoalveloar expansion, and to assess the predictability of expansion at the end of the first set of aligners and at the end of treatment for a better understanding of expansion movements during treatment. CA had no effect on the mid-palatal suture and represented an alternative to conventional approaches to obtain maxillary expansion exclusively at the dentoalveloar level. Ideally, all treatment goals should be achieved in the first set of aligners [25], but in clinical practice this condition happens almost never.

Indeed, in our study sample, each patient required at least two sets of additional aligners, in agreement with Pinho et al. [14], who reported that only 69% of malocclusions traits can be solved within the first set of aligners.

In particular, dentoalveolar expansion correction of non-skeletal constricted arches with initial negative molars torque was reported to reach 80% of the planned movements with the initial series of aligners [14]. According to literature, several Authors [14, 16, 20] reported that additional aligners are needed to overcome challenges such as the breakage of aligners or the need to improve the fit of aligners due to tooth loss and eruption. Although Invisalign First® clear aligners are specifically designed to treat in mixed dentition, the management of short clinical crowns and of tooth replacement was complex, and consequently the efficiency of the device may also be affected. Therefore, the use of additional aligners can represent a treatment strategy suitable to achieve all the objectives of therapy and to manage the development of dentition.

Invisalign® First is designed to achieve up to 8 mm expansion. However, as described by Pinho et al. [14], a maxillary first permanent molar transversal distances evaluation suggested that dentoalveolar expansion between 3 and 4 mm was a predictable movement, instead an expansion movement of > 4 to 6 mm with negative molar torque is considered to be an intermediate correction. An expansion greater than 6 mm was considered a treatment having a skeletal component and therefore to be treated with orthopaedic devices. Furthermore, expansion values between 4 and 6 mm are the most requested by clinicians in interceptive treatment performed by CA [14] and applicable for this sample.

To our best knowledge, only few articles evaluated the expansion predictability in growing patients presenting with mixed dentition and constricted maxillary arch [14, 16, 19, 20].

Gonçalves et al. [19] compared the results of the expansion obtained after the first set of aligners in both arches, pointing out an efficiency of maxillary transversal expansion of 55,2% on deciduous canines, 60.7% on first deciduous molars and 63.3% on second deciduous molars at the end of the first round of aligners. The predictability of expansion at the level of first upper permanent molars was 61,1%. More recently, Kim et al. [20] evaluated the predictability at the end of the expansion treatment on a sample of 90 patients in mixed dentition. The authors found a mean predictability of maxillary tooth expansion of 71.1% for deciduous canines, 67.5% for first deciduous molars, 65.2% for second deciduous molars and 53.4% for first permanent molars at the end of the treatment.

In the present study, the amount of movement achieved compared to the amount of movement programmed resulted in less than 47% for all measurements at the end of the first set of aligners, with a greater improve at the end of interceptive treatment, in the upper and lower arch.

At T2 observation period, the highest predictability was observed at the level of second deciduous molars (70.76%) followed by intercanine width (67.31%). The predictability of expansion movements was detected to be of about 56% at the level of first permanent molars (54.86% on first intermolar mesial width and 58.92% on first intermolar distal width).

The lower predictability at the level of the first upper permanent molars can be explained by several factors, including differences in root surface area, aligner material limitations in exerting appropriate force magnitudes for different teeth, differences in cortical bone thickness, occlusal load and soft tissue pressures [26, 27]. Furthermore, the terminal part of the aligners is the most flexible part and, therefore, a lower expansion force can be exerted [18].

Regarding the lower arch, our results are in agreement with Kim et al. [20], who described the maxillary arch expansion less predictable than mandibular expansion in mixed dentition.

Generally, the amount of planned transversal expansion in the upper arch is greater than in the lower. When the maxilla is narrow, the lower curve of Wilson is accentuated as result of dental compensations [28, 29].

Therefore, the sequential expansion protocol involves only a vestibular tipping movement of the crowns in the mandibular arch.

In our study, the predictability of expansion at the end of first set of aligners was 45.71% on first deciduous molars, 48.70% on second deciduous molars and 41.52% on deciduous canines, and of about 45.44% (46.67% on first intermolar mesial width and 43.21% on first intermolar distal width) at the level of lower permanent molars.

These results are in agreement with Gonçalves et al. [19], who found a predictability of expansion on deciduous canines of 52.2%, 46.2% on first deciduous molars, 59.9% on second deciduous molars and 66.8% on first permanent molars.

As for the maxillary arch, the predictability highly increased at the end of treatment, with the greatest percentage at the level of the second interdeciduous molar width (75.32%), followed by the first interdeciduous molar width (72.75%) and by the interdeciduous canine width (70.61%). About the percentage of the transversal expansion of the lower first permanent molars, the greatest amount of predictability was detected at the level of the intermolar mesial width of about 68.30% (intermolar distal width 62.05%; intermolar transpalatal width 50.47%).

Kim et al. [20] detected higher percentages of predictability of expansion on deciduous teeth at the end of treatment than the results of this study, reporting 81.1% for deciduous canines, 81.2% for first deciduous molars, 77.8% for second deciduous molars. The predictability on first permanent molars (69.4%) was similar to the value found in this research.

Despite the discrepancy between the percentages, all studies collected amount of expansion significantly lower in the first permanent molars than in the other deciduous teeth [19, 20].

One limitation of the present investigation was the retrospective nature of the paper and the absence of a control group. Despite these limitations, the present study allowed to explain the need of additional set of aligners to achieve a good predictability.