Systematic review for orthodontic and orthopedic treatments for anterior open bite in the mixed dentition

Quality of the studies

Randomization increases the reliability of a study and allows final differences to be ascribable to the treatment and not to random or systematic errors [49].

Only three RCTs about the early treatment of open bite were available [23, 28, 40]. However, randomization process was not described.

Sample size was judged as adequate in six studies [21, 27–29, 33, 34]. In the others, it was partly sufficient or insufficient implying low power and high risk to achieve insignificant outcomes.

Previous estimation of sample size was done by two authors [28, 34], but only one [34] described how it was calculated.

The selection description was adequate or fair in all studies except one [21].

The number of dropouts was declared in four studies [17–19, 27], and it was low.

All the trials provided a clear description of the type and duration of the intervention.

The methods used to detect the treatment effects were valid in 12 studies [16, 18, 19, 22, 30–35, 38, 41] and partly valid in ten [17, 20, 23, 24, 27–29, 36, 37, 40].

Some studies [21, 27, 29, 39] lack of an adequate untreated control group probably due to the difficulty in gathering many patients with open bite or the lack of ethical rationale to leave these patients untreated.

Some trials used patients who refused orthodontic therapy [16, 20, 34] or longitudinal data of untreated individuals enrolled in published growth studies [21, 29, 30, 35, 37, 41] as control group.

Weinbach [39] compared the treatment with published cephalometric standards [50].

In one study [21], the method was considered not valid since participants of the groups were not matched according to their dento-skeletal characteristics but just according to age and sex. In another one [39], the appliance was not used exclusively in patients with anterior open bite and there was not a valid control group.

Groups examined by Kiliaridis [17] and Doshi [23] had a too wide age range with subjects treated in the permanent dentition; Pedrin [27] considered too wide ranges of open bite and MPA angle. In two studies [20, 29], treated and control subjects did not have the same age at the beginning and were not observed for the same amount of time which could have influenced cephalometric evaluation of changes.

In six studies, two interventions were tested at the same time, e.g., MBB or RMI and LSE [24], crib and VCC [27, 28], FR and LSE [36, 37, 40], so the results can be attributed either to one or to the other.

SLBB was reported to break frequently in two studies [18, 23]. Doshi did not specify if the treatment was stopped or the appliances were replaced [23]. Kuster and Ingervall replaced the appliances to the patients [18].

Kiliaridis [17] interrupted the treatment earlier than planned and changed the experimental design because of side effects. This did not allow to perform statistical evaluations of the results.

The analysis of the stability of treatment results can not be considered adequate in most studies, in fact follow-up periods were too short [23, 30, 31, 38] and some patients wore contentions while others did not [30, 31], besides some authors applied multibracket appliances during the follow-up [21, 35].

Nineteen studies [17–24, 27–35, 38, 41] included a method error analysis, and only one had blind outcome assessment [17].

Furthermore, five studies declared a power analysis [24, 30, 31, 34, 35].

Fourteen studies used proper statistical methods [18, 20, 22, 29–37, 40, 41]. Among the remaining studies, one did not report any statistics [17], whereas in the others, statistics was inadequate, e.g., parametric tests used in insufficient sample size [16, 19, 23, 24, 39], paired t test used improperly to compare changes between groups [21, 27, 28], and inadequate level of significance [38].