The dentigerous cyst is an interdisciplinary problem. Its development is usually asymptomatic for a long time. Delay in tooth eruption or its absence in the dental arch, or its incorrect position arouses anxiety. DC's consequences are a) bone destruction, b) root resorption, and sometimes when untreated, c) transformation into an odontogenic tumor (e.g., ameloblastoma) or a malignant tumor (e.g., squamous cell carcinoma) [1, 2]. The need to evacuate the cyst is obvious. However, preservation of the DC-associated teeth requires a reliable, not only intuitive protocol. Regardless of the fact that 81% of premolars may erupt spontaneously after surgical removal of the cyst or marsupialization/decompression [24], it does not exempt from continuous monitoring the factors conducive/inconducive to the eruption (both in terms of quantity and quality), which fully justifies the subject of our study.
Some authors claim that spontaneous tooth eruption should occur no later than three months after marsupialization; otherwise, orthodontic traction should be applied [5, 11, 17]. It is in accordance with the results of studies by Hyomoto et al. [21], where the teeth erupted within 109 days on average; after 110 days the number of erupting teeth evidently decreased. Duration of the spontaneous eruption was comparable or longer (5.8 ± 2.8 months) in the study by Yahara et al. [23]. Here, the authors determined that if the tooth did not erupt at least halfway within 3-months post-surgery, it would most likely not reach the occlusal plane even within 12 months. Fujii et al. [20] and Hyomoto et al. [21] observed a longer duration of the spontaneous eruption, compared to the forced one (Table 1). This difference, however, was either statistically insignificant [20] or not analyzed [21]. In the patients studied by Qian et al. [22], all teeth erupted spontaneously after marsupialization, in a significantly shorter time than the teeth without DC (p = .001). It allowed the conclusion that marsupialization induces osteogenesis and the pressure reduction in the alveolar bone, which may stimulate faster eruption. Nonetheless, such interpretation must be treated with caution; there are no reviews on a physiological pace of permanent tooth eruption, which one can compare with a DC-associated tooth eruption duration time.
The spontaneous eruption occurred mostly in patients within the first decade and the beginning of the second decade of their life. The average age of patients experiencing spontaneous eruption in almost all studies was significantly lower than the average age of patients whose eruption process was arrested (Table 1). Only Qian et al. [22] did not show a statistically significant age difference between the study and control groups, which is evident since they compared DC-associated teeth with those from the other quadrant in the same mandible.
The patient's age is closely related to the stage of root formation; therefore, these two factors may be analyzed simultaneously when assessing their impact on a spontaneous tooth eruption after marsupialization. In the study by Hyomoto et al., an initial stage of root formation (r < 1/2) increased the probability of spontaneous tooth eruption after marsupialization in a statistically significant manner (p < .05). Nevertheless, other authors have not demonstrated such a relationship, proving that eruption is utterly independent of the degree of the root formation (Qian et al .: p = .1; Yahara et al .: p = .6126). Due to the fact that stages of root formation in the study and control groups differed from one article to another, the data is not entirely comparable.
It was impossible to determine the influence of patients' gender on the tooth eruption after marsupialization. The gender distribution in the study and control groups was established in none of the studies.
As for the distance between the top of a tooth cusp and the edge of the alveolar process in a perpendicular projection, Fujii et al. [20] observed significantly lower value in cases of spontaneous eruption after marsupialization than in cases of cystectomy or application of orthodontic traction (p < .0007). On this basis, the authors concluded that the tooth's position at a depth < 5.1 mm is one of the predictive parameters of a spontaneous eruption of a DC-associated tooth. Other authors have reached similar conclusions [21, 23]. In turn, Qian et al. [22] showed that despite shallower location of the teeth in the control group (3.5 ± 2.7), the duration time of their eruption was more prolonged, although it was not a statistically significant difference (p = .137).
As for angulation of the teeth spontaneously erupted after marsupialization, in the study of Fujii et al. [20] and Hyomoto et al. [21], it was two and three times smaller on average, compared to angulation in cases requiring a cystectomy or orthodontic traction (Table 1). On this basis, Fujii et al. [20] determined that angulation < 25° is one of the significant predictive parameters of DC-associated tooth eruption (p = .0149). However, this conclusion was not confirmed in the work of Qian et al. [22] and Yahara et al. [23]; moreover, Qian et al. observed greater angulation values in the group of patients whose teeth erupted spontaneously.
The size of a dentigerous cyst was not found to be a predictive factor of a spontaneous eruption. However, in studies by Fujii et al. [20] and Hyomoto et al. [21], an average cyst size in cases where there was a spontaneous tooth eruption after marsupialization was similar, while Qian et al. [22] provided a much lower value (Table 1). Based on these results, it can be concluded that from an orthodontic point of view, the discussed parameter is not a key one in treatment planning. However, it may have an impact on decisions regarding a surgical procedure.
In the work of Fujii et al. [20], the authors described a difference between groups as statistically significant (p < .05), therefore concluding that a ratio greater than 1:1 is a factor conducive to a spontaneous eruption.
Risks of bias in a pre-intervention domain were considered serious in all works. The reasons were: 1. a failure to use interference elimination methods, such as stratification, regression, matching, standardization or inverse probability weighting, 2. retrospective nature of the research, namely: forming study and control groups after the results have been achieved. Thus, prognostic variables might have been deliberately selected to confirm the assumed hypotheses and 3. selection of participants from a larger group of patients treated in a given unit in a specific timeline, which increases the probability of disturbing the proportion of a primary cohort. This study did not adopt a critical risk because the study group's inclusion criteria were precisely defined in all studies.
Since there was no misclassification of an intervention status, the intervention domain's risk was assessed as low. The study participants were treated routinely, using standard procedures, and medical records were kept. Ongoing records were available entirely to the authors; therefore, a risk of bias due to deviations from the intended interventions and risks of bias due to missing data were considered low. The risks of bias in the measurement of outcomes were defined as moderate because methods of assessing the results were objective and carried out in the same way for the study and control groups. Nevertheless, the lack of randomization made the measurement results potentially dependent on whether the patient belonged to the study or control group.
The risk of bias in selecting the reported results was also moderate because the authors provided all the results without using the data obtained in previous analyzes, despite the studies' retrospective nature.
The risk of bias in all four studies was considered critical. However, despite serious risks of bias of all qualified works, it must not be forgotten that serious risks were recorded only in the pre-intervention domain. The reason was a retrospective nature of the study, which is an obvious disadvantage, however easy to overcome in future studies on this aspect.
As for the qualitative assessment based on QAS, in the "Patient Selection" category, the work by Fujii et al. [20], Hyomoto et al. [21], and Yahara et al. [23] obtained 4 points because the patients were appropriately classified into groups according to a differentiating factor, which was either a spontaneous eruption or its lack after marsupialization. Moreover, there was full documentation available, proving that the studied effects did not occur at the beginning of the research. The article by Qian et al. [22] lost 0.5 points for low representativeness of the study group and another 0.5 points due to the qualitatively weaker control group, which were the participants' healthy teeth from the study group. In turn, such construction of study and control groups by Qian et al. provided no additional differentiating factors. Therefore, this paper awarded 2 points in the section on confounding factors. Articles by Fujii et al. [20] and Hyomoto et al. [21] lost 1 point due to age differences between the control and study groups, while the work of Yahara et al. [23] did not receive points in this category due to significant differences in the size of the control and study groups, which could be a significant factor affecting the studied variables. In the effect assessment section, all articles lost 1 point each due to a lack of evaluators' blinding. The works of Fujii et al. [20] and Hyomoto et al. [21] lost 1 additional point due to applying orthodontic traction sooner than three months after marsupialization or performing cystectomy, which could have influenced an assessment of time needed for a tooth to erupt spontaneously.
Limitations
In this systematic review, the most important database available was searched through using a comprehensive search strategy. However, non-English searches were not undertaken, posing a risk of incomplete yield, which could have had an impact on the obtained results due to the publication bias.
So far, very little research has been conducted on the management of the DC-associated teeth. Apart from merely four retrospective works, which have been selected, only case reports and case series are available [7–19, 25, 26], where the authors present selected methods of treatment, based on individual clinical experiences or protocols adopted in a given unit, which is not in line with the evidence-based medicine.