The management of teeth with poor prognosis is challenging in pediatric dentistry. Major causes of dilemma in both children and adolescents include tooth impactions in remote sites or ectopically positioned teeth that require immediate intervention and even teeth with severe traumatic injuries on the verge of exfoliation [1-5]. Because conventional prosthetic treatments involving osseointegrated implants are inappropriate for growing pediatric patients, autogenous transplantation is often adopted in pediatric dentistry [6]. It is a surgical technique of transferring teeth from one place to another in the same individual, offering remodeling and growth of the alveolar bone induced by periodontal membranes. It further allows the formation of natural marginal gingiva efficiently [7,8].

A wide range of success rates of autotransplantation, from 62% to 100%, had been reported in the systematic review by Almpani et al. [9]. Such large variations may be the result of differences in radiographic and clinical criteria, and characteristics of the study samples among literatures; for instance, age groups, donor tooth types and observation periods [9]. Previous findings were more focused on conventional autotransplantation in late adolescence and adulthood with indications of hypodontia or loss of tooth due to trauma and periodontal diseases [9-11]. Therefore, results from these previous reports have limited information for impaction cases, which are major causes of dilemma in children and adolescents. Moreover, past research has mostly involved multi-rooted molars as the donors. Donor teeth with a cone-shaped single root are known to be more favorable for autotransplantation because the furcation area is known to be more prone to an increase in pocket depth and the loss of alveolar bone [12,13]. Such uncontrolled variables were not considered in the previous studies. Furthermore, more than one dental surgeon participated in the majority of studies, conveying the possibility of human errors [11]. Finally, given that most of the study participants were based in Western countries, there seems to be a need for further research that evaluates the possible gaps that can arise from ethnic differences. For example, the direction of displacement of impacted maxillary canines differs between the Western and Asian populations in that a greater prevalence of buccally displaced canines is noted in the Asian population [14]. Such differences may influence the prognosis of transplanted teeth. These aforementioned limitations observed in the previously reported studies have yet to be discussed in depth. Ultimately, studies on the prognosis of autransplantation in Asian children and adolescents with indications of impactions, performed by the same surgeon, are still lacking. Evaluations on how previously known contributing factors such as the developmental stage before autotransplantation, surgical technique, and extraoral time of the donor teeth differ in Asian children and adolescents would provide new insights [11,15-17].

This study aimed to retrospectively evaluate the postoperative sequelae and root growth of single-rooted transplanted teeth and assess factors affecting the outcomes of autotransplantation performed by the same surgeon in Korean children and adolescents.

Autotransplantation is a versatile surgical technique with high potential that can be used to solve problems such as aplasia, impaction, tooth loss due to traumatic injuries, periodontal diseases, or other unfortunate events in all age groups. On many occasions, autotransplantation is performed to replace the missing tooth in the case of aplasia in adults. According to Lundberg’s study on 278 autotransplanted teeth[10], the majority of the indications for autotransplantation was aplasia (60.43%), followed by caries and associated diseases (23.38%), impaction (8.27%), and trauma (3.24%). Similarly, in the studies by Kvint et al. [18], autotransplantation indicators were mainly aplasia (48.84%), displacement (21.86%), and caries-associated diseases (12.56%). Conversely, in this study, the major indication was tooth impaction. The overall frequency of individual-based permanent tooth eruption disturbances is reported to be approximately 20%[30]. Specifically, the maxillary canine is the second most prevalent tooth in terms of ectopic eruption, affecting approximately 2% of the population [31,32]. This explains why transalveolar transplantation of the canine had the highest percentage among other types of donor teeth (79.45%, Table 1). These data provide a glimpse of the clinical potential of autotransplantation in tooth impactions in children and adolescents. In a rare case, conventional transplantation of an erupted premolar to the incisor site (1.37%) was performed to replace central incisors with congenital root malformations. The reliability of such conventional transplantation was demonstrated by Czochrowska et al. [25] who successfully used premolars as a donor to replace central incisors lost from trauma. Similar to studies by Plakwicz et al. [27], we included transplanted teeth with an observation time of at least 6 months to determine legitimate and reliable results.

All of the transplanted teeth, even premolars, were single-rooted in our study. This coherence regarding root morphology is significant since the furcation area of multi-rooted teeth may be more susceptible to inflammation [12,13]. In fact, from the age above 30 years, at least 1 furcation involvement was observed in 50% of molars [33]. Differences in the degree of healing in single-rooted and multi-rooted teeth should be controlled so that the weight of other prognostic factors can be accurately determined. In this manner, our study opted out the candidate factor related to furcations, increasing the reliability of our results.

Large variations in the success rates of the transplanted teeth exist, as confirmed in various articles [9]. This variation may be due to the differences in the radiographic or clinical criteria, sample size, age groups, tooth types, and observation periods. In this study, the success criteria involved the absence of both root resorption and ankylosis and teeth with a crown-to-root ratio of < 1 based on well-established criteria. Kokai et al. [21] reported a success rate of 71% with 5.8-year follow-up for 100 teeth with closed apex. Czochrowska et al. [22] examined 33 teeth with a success rate of 79%, with 26.4-year follow-up. Similarly, Kallu et al. [11] revealed a success rate of 68%, with an observation period of 3.8 years for 273 teeth. In the present study, the success rate was 73.97% for 73 teeth, with a mean observation period of 3.2 years, falling in the range reported previously (Table 2). The lower success rate was noted for incisors (50%) and premolars (55.6%) as the donor teeth. Because 93.2% of the donor teeth were impacted, surgical maneuvers to proclaim the teeth from the impaction site were inevitable in our studies. Possible damage on the root surface of the donor teeth might have affected the prognosis. Conversely, a greater success rate was noted for canine autotransplantation (83.3%) compared with the results of Kallu et al. [11] (51%). Possible explanations for these differences lie in the anteroposterior position of the canines. Palatally displaced canines are approximately two times more frequently observed, except in Asian populations where more buccally displaced canines are seen [14,34-37]. Palatal impactions make surgical accessibility more challenging [38]. In general, buccally displaced canines are easier to access in terms of surgical exposure, minimizing root surface damage.

The orthodontic traction of deeply impacted teeth or teeth with unfavorable orientations has a poor prognosis. In this case, the extraction of the impacted teeth may be the only option apart from transalveolar autotransplantation, which is considered the last resort [39]. Even in the presence of postoperative sequelae such as resorption and ankylosis after surgery, transplanted teeth are likely to survive without exfoliation in children and adolescents [40]. Therefore, 26.03% of “failed” cases do not literally refer to failures in our study.

The eruption timing of incisors, canines, and premolars ranged from 7 - 8, 11 -12, and 10 - 12 years, respectively [41]. Deviation from this range would make parents question the “missing” teeth, and they would seek treatment if necessary. Although some patients may coincidentally notice the impaction during a routine checkup at a dental clinic, the parents are highly likely curious about the reason for its absence. This partly explains why the high percentage of transplantations of incisors, canines, and premolars were performed at the ages of 7 - 9, 11 - 13, and ≥ 13 years, respectively (p= 0.001).

The tooth developmental stage of the donor teeth is an important factor that must be first considered because it influences the prognosis of the transplant. Specifically, when the root formation of two-thirds to three-quarters is achieved, it is adequate for optimal autotransplantation [16,42-44]. Thus, in this study, the tooth developmental stage at the time of operation was mostly Demirjian stage G for canines and premolars. Conversely, approximately half of the transplanted incisors were at stage F (p= 0.044). This may be because the impacted incisors with their crown facing the nasal floor usually accompany root underdevelopment and dilacerations [45].

Prolonged retention of primary predecessors is usually associated with the presence of local factors such as the supernumerary teeth and the ectopic eruption of the impacted teeth. Henklein et al. [46] noted a higher incidence of primary predecessors for impacted canines and premolars, whereas none were present for the incisors. In accordance with these previous studies, the presence of predecessors at the time of autotransplantation was significantly associated with the canines and premolars but not with the incisor site (p= 0.01). Prolonged retention of primary incisors may be rarely observed because the neighboring permanent central or lateral incisors, which are approximately 1.3 times wider in the mesiodistal width, take up the available space, leading to premature exfoliation [47]. The primary predecessors of canines and premolars are usually retained well without root resorption unless pulp treatment or restorative treatments are performed beforehand [46].

The transalveolar transplantation of maxillary canines, the second most prevalent tooth for impaction, was predominant in our study [32]. This explains the finding that most of the recipient sites were in the maxilla (p= 0.002). Premolars usually have sufficient space for eruption owing to the presence of a leeway space [48]. Thus, pre-orthodontic treatment for space acquisition was relatively less likely to be needed for premolars. The need for space acquisition and esthetic alignment of the anterior teeth might increase the need for orthodontic treatment (p= 0.008). Finally, the observation period of canine and premolar transplantations was longer than that of incisor transplantation (p= 0.041). Incisors as donors only take up a small percentage of 73 cases (5.48%); therefore, the statistical significance observed among the donor teeth regarding the follow-up periods may not be clinically meaningful.

The resorption rate (19.18%) was relatively higher than those reported in previous studies by Lundberg and Isaksson [10] (7.91%), Kallu et al. [11] (21.3%), and Kokai et al. [21] (10%). The preservation of periodontal tissues on the root surface of the donor teeth is paramount. Because the accessibility of the deeply impacted donor teeth is usually limited, root surface damage upon extraction might be inevitable [38]. However, this statement cannot explain the reason why the percentage of teeth with signs of ankylosis (5.48%) was lower than that in previous studies: Kallu et al. [11] reported 13.6%, and Kokai et al. [21] revealed 15%. Long-term rigid splinting and the lack of occlusal stimuli increase the incidence of ankylosis [49,50]. As a postoperative intervention to minimize the risk of ankylosis, the orthodontic loading after the healing period has shown positive effects [51,52]. In the present study, orthodontic biologic loading was employed in over two-thirds of cases (72.6%). This may partly explain the comparably lower incidence of ankylosis.

Pulp healing in the form of varying degrees of calcification was observed. In our study, 62.5% of the transplanted teeth falling into the category of ≤ 1-year recall period showed no changes in the size of the pulp canal. This shows that signs of canal calcification within a year were only evident in about one-third of cases. This result was in agreement with the previous findings that signs of canal calcification were evident 6 months after autotransplantation, and such phenomena are not considered pathological sequelae [16,28]. Between 1 and 5 years of recall period, 63.7% of teeth revealed radiographic signs of obliteration. The connective tissue originating from the local periodontal ligaments may be incorporated into the pulp canal, which is responsible for this repair. Despite being unidentical histologically from the pulp tissue, it induces the formation of reactive or reparative dentin, leading to gradual canal obliteration [53]. The transplanted teeth with a closed apex usually undergo root canal treatment within 2 weeks after surgery [21]. In this study, the percentage of teeth with a closed apex was 31.5%, whereas the percentage of teeth that have undergone endodontic treatment was 21.92%. Clearly, not every fully developed tooth received root canal treatment after autotransplantation. Only teeth showing clinical symptoms to percussion and palpation were endodontically treated. This implies that teeth with a closed apex may not necessarily need root canal treatment in children and adolescents, suggesting some differences in healing potential compared with adults. The transplanted teeth evaluated at > 5-year recall period had conveyed that some of the transplanted teeth had received root canal treatment at some point during the observation period. Not every transplanted tooth with an open apex sustained pulp vitality throughout the observation period. According to Denys et al. [16], it is advisable to perform root canal treatment on teeth with an open apex even in the absence of clinical symptoms due to the plausibility of pulpal necrosis and inflammatory resorption at any time. Ten teeth (20%) among the transplanted teeth with open apex had undergone root canal treatments since radiographic signs of periapical rarefaction appeared. This emphasizes the importance of periodic examination of the transplanted teeth. Based on our results, transplanted teeth with an open apex are not advised to be endodontically treated until clinical or radiographic signs appear.

The percentage of the transplanted teeth with a crownto-root ratio of ≥ 1 was similar to those reported by Kallu et al. [11] (5.5%) and Czochrowska et al. [22] (6.06%). However, none of the transplanted teeth reached the final root length (contralateral control teeth as a reference), similar to the results reported by Andreasen et al. [42]. Certain reductions in root length may be due to damage on Hertwig’s epithelial root sheath. Furthermore, the lag period before the full vascularization at the root apex of the donor teeth might lead to a delay in nutrition supply. The ectopic positioning of the donor teeth to the recipient sites to ensure infraocclusion may also influence revascularization [42]. Moreover, similar to studies by Westerveld et al. [26], no significant results were revealed among the developmental stage and the mean achievement in the final tooth length (Table 6). Determining the stage of tooth development with the Demirjian stage is difficult when the root formation lies in between two discrete stages; therefore, it is considered only an estimate. Further information such as the width of the root apex would be a more reliable indicator in the determination of the stage of tooth development [26]. The transplanted teeth with a wider apical width and root development > 50% of the final length have a longer final root length [26,42]. Thus, root development stages must always be considered before autotransplantation.

Clinically, the amount of root growth after autotransplantation must be determined. These findings aim to reveal if the amount of root development differs among the transplanted teeth at different Demirjian stages. The transplanted teeth with an open apex (Demirjian stages F and G) both showed a significant increase in root length when compared to the teeth with a closed apex (Demirjian stage H), as depicted in Table 6. All transplanted teeth with an open apex showed root development (none of them ceased to grow). This result from our study was in agreement with the results published by Lucas-Taulé et al. [54], who reported that over 84.1% of teeth with open apex showed root formation. Data on Demirjian stage H was evaluated to take into account of the possible errors in measurements because it represents teeth with complete root development with a closed apex. The mean root growth of 0.11 ± 0.81 mm was set as a reference for ceased development. No significant differences in the amount of root growth were observed between the transplanted teeth at Demirjian stages F and G. These findings were in accordance with studies reported by Andreasen et al. [42] and Slagsvold and Bjercke [43] that an increase in root length was observed when autotransplantation was performed between one-half and three-quarters of the expected root length. No significant differences were observed within this length. It can be concluded that significant root growth can be expected in teeth with open apex.

Similar to previous studies [27], the crown-to-root ratio of the transplanted teeth was greater than that of the control teeth (Fig. 4, p < 0.0001). Only 6 transplanted teeth had a crown-to-root ratio of ≥ 1 (Fig. 3). According to Plakwicz et al. [27], as the difference only represents roots that are 1 - 1.25 mm shorter for transplanted teeth, it may not be clinically significant.

Bone grafts are widely used to promote the healing of intrabony defects via periodontal regeneration. In the case of implantation of titanium fixtures in regions with large defects, bone grafts can act as pillars for primary stability [55]. In such cases, bone grafts are crucial since titanium fixtures do not possess periodontal ligament cells, whereby bone-tissue regenerating potential is absent. Conversely, autogenously transplanted teeth have periodontal ligament cells on the surface of the root capable of inducing bone formation. These differences had given rise to a controversial issue regarding the necessity for the use of bone graft upon autotransplantation. According to Suwanapong et al. [56], even in the case of excess bone removal during recipient site preparation, complete trabeculation of alveolar bones was seen within 12 months without any signs of inflammation. The healing of transplanted teeth was not dependent on the amount of remaining bone at the recipient site. Since trauma to the donor root surface due to insufficient recipient site preparation acted as greater threat, extensive bone removal was recommended. Moreover, the use of bone grafts did not significantly improve healing in terms of bone regeneration, as highlighted by Bauss et al. [57] and Miura et al. [58]. Instead, bone grafts disturbed the stability of the transplant. In autotransplantation where immediate revascularization is paramount, the presence of bone grafts might hinder the supply of nutrients to the transplanted donor teeth. This explains the higher rates of root resorption, ankylosis, and a lower success rate with the use of bone grafts. Intriguingly, the rate of bone formation was faster in the patients under 18 years, relative to older patients [56]. These results further emphasized the lack of need for bone grafts upon autotransplantation in children and adolescents. Thus, the criteria for the use of a bone graft are still unclear. In our study, a bone graft was utilized in the case of severe dehiscence, approximately 3 to 4 times greater than the width of the crown of the donor teeth. Such dehiscence was often observed in impacted canines with hyperplastic follicles. Based on the results from our study, bone grafts should not be used even in the presence of large bony defects. Preservation of periodontal ligament cells is more crucial without exerting trauma upon transplantation, followed by flexible splinting at infraoccluded state.

Because the same oral surgeon performed the surgical operation with identical protocol, determining how one’s experience in autotransplantation influences the prognosis of the transplanted teeth is possible. Schwartz et al. [17] reported higher success rates as the operator gained more experience. However, Jakobsen et al. [59] assured that surgeon experience is not a critical factor. Autotransplantation performed by two senior surgeons and six unexperienced junior surgeons did not significantly affect the survival rate of the transplanted donor teeth. The survival rate differs from the success rate because the transplanted teeth showing any signs of postoperative sequelae such as signs of resorption or ankylosis are all considered “survived” as long as these teeth are not extracted. In the present study, 37 teeth transplanted in the latter half of the total cases (50.7%) showed a significantly higher success rate and a lower incidence of root resorption. Because every transplantation was performed by the same surgeon, our results are more persuasive in terms of experience.

The splinting period was not a significant factor that influenced success rates in our study. Preoperative orthodontic treatment allows for the acquisition of sufficient space at the recipient site. Transplantation of the donor teeth at the vertically and mesiodistally ideal position allowed bracket positioning at which passive ligation of the archwire was possible. Consequently, direct orthodontic loading without splinting was engaged in 7 cases. However, not every case with preoperational orthodontic treatment led to direct loading. Sometimes, in the case where the transplanted teeth were positioned either too submerged or buccally faced to avoid occlusion, direct orthodontic engagement would exert excessive force. Such active force may lead to root resorption or alveolar bone loss and direct loading is discouraged [60]. Despite the recommended composite-wire splinting period between 4 and 8 weeks for the transplanted teeth with a closed apex reported by Kokai et al. [21], 9 cases were splinted for more than 8 weeks. In the areas of severe dehiscence, especially for canines in our study, primary stability was delayed and therefore removal of splints was postponed accordingly. According to Kim et al. [61], radiographic infrabony dehiscence observed right after extraction of mandibular third molars recovered to a normal range after 6 months. Similarly, the increase in bone density subsequent to mandibular cyst enucleation was reported to be 37% within 6 months and 42.27% after 12 months [62]. The rate of spontaneous bone healing may be greater in the first few months, but individual differences in the healing potential should be taken into account [61]. The longest splinting period of approximately 9.7 weeks, an outlier, was recorded in our study. The rest of the cases falling into the category of ≥ 8 weeks of splinting period were within 9 weeks. This clearly demonstrated that alveolar bone formation in areas of large dehiscence to achieve optimum stability takes at most 10 weeks. However, there was insufficient evidence to support associations among the splinting periods and the success rates in our study. This highlights the possibility that as long as initial stability with signs of radiographic bone formation and physiologic mobility is achieved, splints can be removed at any time, from a 1-week postoperative stitch out to 10 weeks. This statement was supported by Lundberg and Isaksson [10] and Kokai et al. [21] who carried out fixation for varying periods of 1 to 3 weeks and 4 to 8 weeks, respectively. Due to the fact that prolonged fixation may result in ankylosis, it would be advisable to stop splinting as soon as the primary stability of the transplanted teeth is achieved [52].

The skewed distribution in age groups and donor tooth types was noted in this retrospective study. Moreover, limitations such as patients’ and parents’ demands, which might have influenced the overall treatment plan, were not controlled. Thus, a predesigned prospective study with more evenly distributed age groups and tooth types is warranted.

In this study, the majority of indications were tooth impactions. The location and depth of the impaction may influence the difficulty of donor tooth extraction. For instance, palatally displaced maxillary canines are more challenging to extract than buccally displaced canines [38]. Complications may be encountered during surgery, and technically challenging extraction may damage periodontal tissues, which are strongly associated with root resorption or ankylosis [18,63]. Three-dimensional information on the mesiodistal angulation and buccolingual position of the ectopically positioned teeth can be acquired in further studies to reflect the possible adversity of surgical procedures.

Perioperative factors such as the tooth extraoral time and storage media may affect the overall treatment outcome [15]. Although such information was limited in this study, variations in these perioperative factors were partly controlled and minimized. The same surgeon performed all surgical procedures, and the extraoral time was minimized by placing the donor teeth into the impaction site for the preservation of periodontal tissues on the root surface.

The influence of types of splinting on success rates was opted out of this study due to the matter of multicollinearity. It is known that flexible splinting is crucial in the case of splinting traumatically avulsed teeth. A wire with a diameter of 0.4 mm, or 0.014 inches, and nylon wire are known to be ideal flexible splints [64]. In this study, all forms of splinting involved nylon or wire splints such as Cu-NiTi and NiTi of 0.014-inch diameter. Since all these types of splints fall into the category of flexible splinting, the weight of their significance on the success rates may be low. Prospective studies on comparing rigid and flexible splinting, or suture and wire splinting may be required in the future.

Linear measurements using ImageJ software on panoramic radiographs have some limitations. The patient’ s head position may cause changes in the occlusal plane, leading to deviations from the actual measurements. According to Stramotas et al. [65], linear vertical measurements and calculations of the ratio of the same patient at different times revealed consistent accuracy. Absolute measurements from panoramic radiographs by themselves may be different from the actual measurements. Calculating differences in vertical lengths at two time points or ratios has been confirmed to be both reliable and accurate. In the future, using post-operative CBCT to measure root length would be more accurate.

This study only involved radiographic treatment outcomes to evaluate the success criteria. As mentioned in previous studies, clinical treatment outcomes such as color, mobility, and pocket depth can provide further information required for proper evaluations of autotransplantation outcomes [11,21,22]. A discrepancy in radiographic and clinical outcomes may be observed, conveying the possibility of localized inflammation within the pulp or on periodontal membranes. Further studies with more standardized clinical and radiographic criteria are recommended.

In this study, 54 (73.97%) of 73 transplanted teeth met the success criteria, with 100% survival during the observation period. In addition, 16 (21.9%) of 73 transplanted teeth had received root canal treatment. For vital teeth, pulp healing in the form of canal calcification (58.9%) was observed. A significant increase in root growth was noted for transplanted teeth with an open apex, and none of the root growth ceased clinically. These results highlight the predictability and clinical potential of autotransplantation in children and adolescents, particularly for tooth impactions.

The root developmental stage of the donor teeth was not a critical factor in the success of autotransplantation. Instead, being technically more experienced on the surgical procedures and prohibiting the use of bone grafts may lower the probability of root resorption, ankylosis, or root development for proper tooth function. Even in the presence of the aforementioned postoperative sequelae, the transplanted teeth can be functionally and esthetically pleasing without exfoliation in children and adolescents.