Clinical outcomes of tooth autotransplantation: a systematic review and meta-analysis of survival
Article information
Abstract
Objectives
Autotransplantation is a procedure that involves the extraction and transplantation of a tooth from one site to another within the same individual. This systematic review and meta-analysis aimed to investigate how clinical outcomes of autotransplanted teeth evolve over time and the principal reasons for extraction.
Methods
Studies were identified from five databases. A meta-analysis was performed to estimate the survival rates in the short-term (1 to ≤4 years), medium-term (>4 to ≤8 years), and long-term (>8 years) periods. Subgroup analysis was performed for age and root development. Risks of bias, reasons for extraction, and patient-reported outcome measures were evaluated.
Results
Of the 3,941 reports initially identified, 46 were included. The estimated short-, medium-, and long-term survival rates were 96.31% (95% confidence interval [CI], 94.80–97.82), 88.23% (95% CI, 85.59–90.87), and 84.80% (95% CI, 76.70–92.91), respectively. There were no significant differences in outcomes between age and root development groups. The most common reason for tooth loss was root resorption. High patient satisfaction rates were reported.
Conclusions
Autotransplanted teeth exhibit high survival rates in the short- to medium-term. Minimizing root surface damage and excluding pulpal contaminants may promote longevity. The procedure appeared equally successful for teeth at different stages of root development and across various age groups.
INTRODUCTION
Autotransplantation is a surgical procedure that involves the extraction and transplantation of a tooth from one site to another within the same individual [1]. It serves as a viable option for the replacement of teeth that are missing or need to be extracted due to traumatic injuries, poor restorability, and/or severe developmental defects [2,3].
Due to the preservation of the periodontal ligament (PDL), autotransplantation offers distinct advantages over other types of fixed prostheses, particularly for younger patients undergoing skeletal growth and developmental changes [4], as well as those who require orthodontic treatment [5]. Other biological advantages include the preservation of proprioception and the ability to remodel in conjunction with the alveolar bone [6]. Whilst autotransplanted immature teeth can benefit from revascularization and continued root development, autotransplantation procedures have also been carried out using mature teeth [7].
Considerations for the donor tooth include a favorable root form to facilitate atraumatic extraction, extent of root development, and sufficient periodontal support prior to extraction [1]. Patient factors, such as age, as well as clinician-related factors, such as surgical techniques, have similarly been reported to influence autotransplantation outcomes [2]. Conversely, damage to the root surface and PDL may result in complications, including external inflammatory root resorption (EIRR) and external replacement root resorption (ERRR), potentially compromising the clinical outcome [8].
Clinical outcomes can be evaluated by determining the healing events, i.e., ‘success,’ and/or functional retention, i.e., ‘survival,’ of the autotransplanted tooth. Whilst ‘success’ typically takes into account biological complications, ‘survival’ refers to the functional presence of the autotransplanted tooth in the mouth [9]. Functional retention of these teeth is particularly relevant to patients and is an important aspect of patient-reported outcome measures (PROMs) [10].
The outcome of autotransplantation has been examined across various time frames; however, the impact of short-, medium-, and long-term follow-up remains unaddressed in prior studies [11–13]. Furthermore, the reasons for extraction and PROMs related to autotransplantation have been largely overlooked in previous reviews [11,12,14,15]. This study aimed to evaluate the survival rate of autotransplanted teeth over different time frames, reasons for extraction, PROMs, and explore key potential influencing factors, namely age and root development.
METHODS
This review was performed in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. The protocol was registered on the PROSPERO database (ID: CRD42024513534). The PICO question was: what are the short-, medium-, and long-term (C) survival rates (O) of autotransplanted teeth (I) in patients (P)?
Search of the literature
A literature search was conducted to identify relevant studies from the databases PubMed, Embase, Scopus, Web of Science, and Cochrane Central Register of Controlled Trials. The search was conducted in January 2025 with no initial time limit. Table 1 outlines an example of the search strategy employed.
Search strategy used for PubMed
After removal of duplicates, two reviewers (JW and EF) independently screened titles and abstracts to assess their suitability using predefined inclusion and exclusion criteria in Covidence systematic review software (Veritas Health Innovation, Melbourne, VIC, Australia). Full texts were then reviewed. Additional studies identified from references were also considered. Disagreements were resolved by consulting a third reviewer (GC) to achieve a consensus.
The inclusion criteria were: (1) intervention was autotransplantation, (2) ≥1 year follow-up, (3) permanent dentition, (4) reported survival rate and/or number of teeth extracted, (5) outcomes determined by clearly defined clinical/radiographic criteria, (6) English language, and (7) clinical studies involving humans. The exclusion criteria were: (1) in-vitro/animal studies, (2) case reports/series, (3) intervention included intentional replantation, and (4) review/opinion articles. If multiple studies had investigated the same or overlapping sample populations, only one representative study, i.e., the one with the largest sample size, was included.
Risk of bias assessment
Two reviewers (JW and EHWF) assessed the risk of bias of the included studies using the Risk Of Bias In Non-randomized Studies of Interventions, ver. 2 (ROBINS-I V2) tool [16]. A third reviewer (GC) was consulted to resolve any disagreements.
Data extraction
Survival rate, defined as the proportion of autotransplanted teeth present at follow-up, was extracted from the included studies. If not explicitly stated, it was calculated based on the number of teeth lost compared to the initially recruited sample. Average follow-up time, or observation period, was recorded. When not stated, the survival rates corresponding to specific time points were recorded instead. Other data collected included the mean/median ages of patients undergoing autotransplantation, the type of donor tooth, and the proportion of teeth with complete or incomplete root development. When declared, PROMs and reasons for extraction were recorded and categorized. If multiple reasons were stated for a single tooth, this would be reflected in the applicable categories.
Statistical analysis
A meta-analysis was performed in Stata software ver. 16.1 (StataCorp LLC, College Station, TX, USA) using the metaprop_one command [17] to estimate pooled survival rates. Study-specific confidence intervals (CIs) were calculated using the exact method, incorporating a continuity correction of 0.5 to prevent the exclusion of studies reporting 0% or 100% survival rates and ensure computational stability. Pooled survival estimates were derived using Wald CIs by random-effects meta-analysis [18], with significance set at p < 0.05. A random-effects model was selected a priori due to the expected clinical and methodological heterogeneity across studies, including differences in patient populations and study designs, and the intention of generalizability. To account for temporal variations, separate meta-analyses were performed, stratified by follow-up duration. Follow-up periods were grouped into short-term (1 to ≤4 years), medium-term (4 to ≤8 years), and long-term (>8 years). Pooled survival estimates were calculated for each period. Data lacking clear follow-up times were excluded from meta-analyses.
Whenever appropriate data were available, meta-regression and subgroup analysis were conducted using Stata’s metareg and metaprop one commands, respectively, to evaluate the following potential confounders: patient age (≤18 years vs >18 years) and root development (completed vs incomplete). For the latter, only those studies that had involved all donor teeth with either completed root formation or incomplete root development were included in the subgroup analysis to prevent excessive heterogeneity of results. Substantial heterogeneity among studies was quantified using the I2 statistic (I2 >50%) or the chi-square test with a p-value of <0.01 [19]. The random-effects model was used to estimate survival rates [19], with significance set at p < 0.05. All statistical analysis was performed under the guidance of a qualified statistician (K.Y.L).
RESULTS
A total of 3,941 records were identified. After removal of duplicates, retracted studies, and corrupted files, 2689 records remained. Initial screening with the addition of hand-searched articles resulted in 224 articles eligible for full-text screening. Of these, 178 studies were excluded, whilst 46 studies were included in this review (Figure 1).
Flow diagram of study selection.
The characteristics of the included studies are summarized in Table 2 [3–10,20–57]. The risk of bias of individual studies is shown in Figure 2. The majority were associated with a moderate (n = 21) to serious (n = 20) risk of bias.
Included studies and their main characteristics
Risk of bias of included studies.
The reported overall survival rates ranged from 67.9% to 100%. The average follow-up period ranged from 1 to 26.4 years, with an average patient age between 10 and 39 years. Most studies were classified as having a short-term observation period of 1 to ≤4 years (n = 20), while fewer studies observed autotransplantation outcomes for medium-term (n = 12) and long-term (n = 14) observation periods. The short-term survival rate was 96.31% (95% CI, 94.80–97.82). The medium- and long-term survival rates were 88.23% (95% CI, 85.59–90.87) and 84.80% (95% CI, 76.70–92.91), respectively (Figure 3). Substantial heterogeneity was observed between studies across all time periods (I2 > 50%, p < 0.01).
Forrest plots of overall estimated survival rate of autotransplanted teeth in the (A) short-term, (B) medium-term, and (C) long-term. ES, effect size; CI, confidence interval.
Subgroup analysis using single-factor meta-regressions revealed no significant association between age and the survival of autotransplanted teeth (p > 0.05). The proportion difference, CIs, and p-values are reflected in Table 4. Half of the studies investigated autotransplanted teeth in patients ≤18 years (n = 23). The short-, medium-, and long-term survival rates for these patients were 96.92% (95% CI, 94.79–99.05), 87.56% (95% CI, 84.74–90.38), and 89.77% (95% CI, 83.37–96.17), respectively. For those over 18 years, survival rates were slightly lower, i.e., 95.92% (95% CI, 93.40–98.44), 88.94% (95% CI, 84.62–93.25), and 82.20% (95% CI, 69.97–94.44), respectively, although these differences were not statistically significant (Figure 4).
Proportion difference, 95% confidence intervals, and p-values for subgroup analysis of age (≤18 years vs >18 years)
Comparison of estimated survival rates of autotransplanted teeth in patients 18 years or younger compared to patients over 18 years old, in the (A) short-term, (B) medium-term, and (C) long-term. ES, effect size; CI, confidence interval.
Similarly, the extent of root development was not significantly associated with the survival of the autotransplanted teeth (p > 0.05). The proportion difference, CIs, and p-values are reflected in Table 5. Short, medium, and long-term survival rates for mature teeth were 91.38% (95% CI, 85.29–97.47), 92.14% (95% CI, 89.10–95.18), and 85.30% (95% CI, 73.78–96.81). For teeth with incomplete root development, survival rates were 99.50% (95% CI, 98.31–100.70), 88.59% (95% CI, 83.76–93.43), and 96.39% (95% CI, 92.77–100.00), respectively. Except for medium-term outcomes, which had the fewest applicable studies for both groups, teeth with incomplete root formation showed slightly better survival (Figure 5).
Proportion difference, confidence intervals, and p values for subgroup analysis of root development (completed vs incomplete)
Comparison of estimated survival rates of autotransplanted teeth with completed root development compared incomplete root development, in the (A) short-term, (B) medium-term, and (C) long-term. ES, effect size; CI, confidence interval.
Substantial heterogeneity was observed amongst all subgroups (I2 > 50%, p < 0.01), except for studies investigating the medium-term survival rate for patients 18 years of age or younger (I2 = 0.0%, p = 0.92) and the short-term survival rate for teeth with incomplete root development (I2 = 0.0%, p = 0.51).
Regarding donor tooth type, almost half of the studies (n = 22) investigated a variety of donor teeth, whilst the remainder evaluated a single tooth type. Molar teeth were most commonly studied (n = 11), followed by premolars (n = 9) and canines (n = 2). Although this factor was not included in the subgroup analysis, the overall impression from qualitative evaluation suggested that donor tooth type was not a major prognostic factor. Few studies reported that specific tooth types, i.e., molar teeth [30], mandibular teeth [7], and multirooted teeth [5], could be associated with poorer survival rates.
Of the 259 teeth that had been extracted after autotransplantation, nearly half were extracted due to various types of root resorption (n = 128), the most prevalent form being EIRR (n = 36), followed by ERRR (n = 34). A minority (n = 6) had suffered external cervical root resorption, and the resorption type in the remaining teeth was unspecified (n = 52). About one-third were lost due to periodontal reasons (n = 84). Other factors, such as compromised coronal structure (i.e., cracks or caries beyond restoration) (n = 5), periapical infection (n = 2), and root fracture (n = 8), contributed to less than 10%. Only 5% failed due to immediate post-surgical complications (n = 13), including failure of initial healing and acute infections (Figure 6).
Proportion of autotransplanted teeth extracted (n = 259) for various reasons.
Eight studies evaluated PROMs. Various short-form questionnaires were used to evaluate patient satisfaction, functionality, and/or symptoms (Table 3 [21,22,40,43-45,47,49]). The surveys were either designed specifically for the study or were modified from existing instruments [27,58,59]. The majority of patients experienced minimal intra- and postoperative discomfort associated with the surgery [10,40,43,45]. Patients were generally pleased with the aesthetic outcome [22]. One study remarked that the most common postoperative inconvenience was food trapping [40]. Overall, studies reported high levels of patient satisfaction, and most autotransplanted teeth were functional and well-received [10,43,45].
Studies reporting patient-reported outcome measures (PROMs) and a summary of the findings
DISCUSSION
Over the last decade or so, the increased interest in tooth autotransplantation has led to the publication of multiple studies [9,41,45,58], literature reviews [60,61], as well as a position statement by the European Society of Endodontics [62]. Given the considerable body of evidence, summarizing and analyzing the research pertaining to clinical outcomes would enhance our understanding of this treatment modality. The primary objective of this study was to synthesize the available data regarding survival rates of autotransplanted teeth via meta-analysis, whilst supplementing these findings with subgroup analysis of the key influencing factors, i.e., age and root development, as well as a qualitative evaluation of the reasons for extraction and PROMs.
In total, 46 studies were included in this review, which suggests tooth autotransplantation is a well-documented treatment modality. The studies identified for this review spanned various time frames, with almost half classified as having a short-term observation period (i.e., 1 to ≤4 years) [20,21,24,26], whilst the remaining were generally evenly distributed between medium- and long-term observation periods. A few large-scale, long-term observational studies were also identified [3,41]. Nevertheless, each study represents an important contribution to this systematic review because it facilitated a comprehensive analysis of how survival rates evolve over time.
This review demonstrated that autotransplantation of teeth yields favorable survival rates, with 96.31% of teeth retained in the oral cavity for up to 4 years. This dropped to 88.23% when autotransplanted teeth were observed for a period between 4 and 8 years, and decreased further to 84.80% when observed for more than 8 years. Unsurprisingly, a reduction in survival rates corresponded with longer observation periods, as more complications are likely to occur over time [20]. The survival rates were comparable to other forms of surgical endodontic treatment, such as intentional replantation [63] and apical surgery [64]. Hence, with careful case selection and adherence to evidence-based practices, autotransplantation should be considered as a viable and predictable treatment option. It is important to note that survival rates express the functional retention of teeth, but do not reveal information about specific clinical outcomes, such as the continuation of root development, revascularization, presence or absence of resorptive defects, and other pathologies [6,8]. Nevertheless, functional retention is a key component of the patients’ experience [65,66]. Criteria for ‘success,’ i.e., the supposed lack of biological complications, do not only vary widely across different studies [22], but may also be too strict to judge the outcome from a patient’s perspective [27]. Hence, survival analysis remains a meaningful clinical determinant of treatment effectiveness [67].
Whilst the characteristics of the individual studies differed across a variety of dimensions, such as tooth type, observation periods, surgical techniques, and age (Table 2), it was noted that the majority of studies were observational in nature, therefore exhibited a moderate to serious risk of bias due to the effect of uncontrolled confounders. The authors acknowledge that this limits the generalizability and reliability of the findings of this review, a shortcoming of the study. However, recent efforts to standardize surgical protocols [60,62] and core outcome sets [68] may improve the consistency and quality of future research.
Despite the application of select inclusion and exclusion criteria, high levels of heterogeneity were observed amongst studies, especially regarding the tooth type, age, root development status, and the surgical technique [7,9,30]. There appeared to be a large variability in the reported parameters [3,12], with most studies primarily focused on reporting the outcomes, with secondary evaluation of selected variables [6,23,26,31]. This heterogeneity complicates meta-analyses, risking potentially unreliable and unrepresentative results. Furthermore, stratifying the period of observation into short-, medium-, and long-term for this review further limited the number of studies within each subgroup. Subgroup analysis and meta-regression were attempted to account for the heterogeneity present and investigate potential influencing factors. However, a comprehensive analysis of all confounding variables was not feasible, which is a recognized limitation of this study. That said, after evaluating the available data, this review was able to analyze the effect of two key factors: root development and age [2,6,12,42]. Although most studies included in our analysis did not report within-study survival rate data stratified by age groups or root development types, precluding a meta-analysis of effect sizes (e.g., odds ratios or hazard ratios) for these subgroups, the approach of comparing groups across studies via subgroup analysis and meta-regression nevertheless allowed for the investigation of the influence of age and root development using the available study-level covariates. However, this approach may result in reduced precision compared to within-study comparisons due to potential ecological bias, which is a limitation of our review. Future studies should consider reporting stratified survival data to enable more robust subgroup analyses.
The extent of root development has been suggested to affect autotransplantation outcomes for several reasons. Shorter root-forms allow for easier manipulation during surgery [1], and with the presence of the Hertwig epithelial root sheath, immature teeth have greater potential for continued root growth and regeneration of periodontal support [69]. Several studies found that teeth with incomplete root development were less likely to be extracted or experience complications compared to their fully developed counterparts [3,30]. A previous review concluded that the stage of root development significantly influenced the outcome of autotransplanted teeth, but the authors cautioned the interpretation of the results due to substantial inconsistencies amongst the pooled data [12]. Hence, to facilitate subgroup analysis and minimize heterogeneity, root development was dichotomized into ‘incomplete’ and ‘completed’ and was analyzed according to each observation time frame. The results of the present study indicated that root development was not associated with survival of the autotransplanted teeth (p > 0.05). Nevertheless, biological considerations may still steer clinicians towards using immature teeth for autotransplantation, if available.
Patient age may have an impact on the survival of autotransplanted teeth, as studies suggest younger patients experience fewer biological complications [7,38,48]. Younger individuals exhibit higher regenerative potential [2,48] and more favorable bone mineral density for extractions [7]. However, some studies have found that age did not influence the outcome [9,57]. The results of this review suggested that patients aged 18 years or younger seemed to experience slightly better outcomes compared to older individuals, although the association was not statistically significant (p > 0.05). An ideal age for autotransplantation has not been identified per se, but autotransplantation should be considered as an effective treatment option for younger patients, especially given the limitations of prosthodontic replacement options in these age groups [9,60].
Although our study was unable to investigate the influence of donor tooth type on survival rates in subgroup analysis, given the heterogeneity of the data, multirooted donor teeth and mandibular appeared to be associated with poorer outcomes, which was attributed to difficulties during extraction and more challenging endodontic treatment [5,7,30]. Although not necessarily a major prognostic factor, careful case selection focusing on ease of extraction and appropriate tooth dimensions can provide a more favorable environment to promote healing [1,8,38].
Regarding the reasons for loss of autotransplanted teeth, about half of the teeth were extracted due to various forms of resorption, the most common being EIRR, followed by ERRR. Bacterial ingress into the root canal system, coupled with root surface damage, is responsible for the onset of EIRR [70,71]. Although limited evidence suggests that systemic antibiotics may have a protective effect against EIRR [9], current guidelines emphasize that antibiotics should be prescribed only to medically compromised patients [62,72,73]. Therefore, maintaining an aseptic field, reducing surgical trauma, and timely root canal treatment may improve survival outcomes [1,26,34].
Unlike EIRR, which tends to manifest as an early complication [3], the physiological processes leading to ERRR are more gradual, especially in older individuals with lower cell turnover rates [74]. Hence, the affected teeth can be functionally retained for many years [75]. However, particularly in children undergoing rapid developmental changes, ERRR can quickly lead to stunted dentoalveolar growth and tooth loss [4,74]. Ultimately, ERRR is still fundamentally a pathological process and should therefore be mitigated whenever possible. Preservation of viable PDL is universally considered a major protective factor [76]. Extra-oral time within 15 minutes [8,9], digitally assisted techniques, use of three-dimensional printed replicas [21,31,60], and appropriate functional loading have been suggested to be beneficial for the regeneration of periodontal tissues and for reducing the risk of ERRR [3,6].
Periodontal attachment loss was the next most significant reason for extraction. Untreated periodontal disease is a well-established risk factor for tooth loss [77,78], and this is no different for autotransplanted teeth. Deficient recipient bone level and poor initial stability were associated with periodontal defects and poorer healing outcomes [8,9]. Endodontic and restorative-related conditions, such as periapical infections, root fractures, and unrestorable caries, accounted for around 10% of teeth lost. Lapses in surgical technique and post-surgical complications played an equal, yet minor role in survival outcomes. Therefore, as resorption appears to be the most significant reason for the extraction of the autotransplanted tooth, followed by periodontal attachment loss, future research could prioritize strategies to reduce the likelihood of root resorption and encourage periodontal healing.
A comprehensive understanding of clinical outcomes is incomplete without the evaluation of PROMs [66,68]. The present review found only eight of the 46 studies assessed PROMs [10,21,22,40,43,45,49]. The studies utilized a variety of custom instruments, making cross-study comparisons challenging. To circumvent this, studies could choose to implement previously validated and widely used instruments to improve the comparability between studies, particularly regarding the impact of different treatment options on patients’ well-being [59,67,79,80]. Nevertheless, there were high levels of satisfaction reported in all eight studies. Several studies reported that minimal discomfort was experienced by patients during and after the surgery [10,21,43]. One study reported that although achieving optimal aesthetic outcomes could be highly technique-sensitive, patients were still largely satisfied with their appearance [22]. Overall, patients were satisfied with the function of the autotransplanted tooth, with minor inconveniences, such as food trapping [21,40,43]. A recent review has echoed the significant evidence gap concerning PROMs in autotransplantation treatments, underscoring the need to standardize future clinical research by incorporating appropriate and validated instruments [80].
CONCLUSIONS
The survival rates of autotransplanted teeth are 96%, 88%, and 84% in the short-, medium-, and long-term, respectively. The longevity of autotransplanted teeth decreases with longer periods of observation. The procedure seemed equally successful for teeth with different stages of root development and age groups. However, the heterogeneity amongst studies limits the generalizability and reliability of these findings. Patients are generally satisfied with the outcome and the functionality of their transplanted teeth. The most common reason for extraction was root resorption. Curating strategies to reduce the risk of resorption, such as ensuring minimal damage to the PDL, atraumatic extraction, and excluding any pulpal contaminants, could optimize outcomes and improve the survival of autotransplanted teeth.
Notes
CONFLICT OF INTEREST
No potential conflict of interest relevant to this article was reported.
FUNDING/SUPPORT
The authors have no financial relationships relevant to this article to disclose.
ACKNOWLEDGMENTS
We would like to thank Ms. Chloe Cheung for her valuable assistance in formulating the search strategy for this review. The authors deny any conflicts of interest related to this study.
AUTHOR CONTRIBUTIONS
Conceptualization, Methodology: Wong J, Cheung GSP. Investigation, Data curation: Wong J, Li KY, Fok EHW. Resources: Wong J, Li KY. Supervision: Cheung GSP. Project administration: Zhang C, Cheung GSP. Writing - original draft: Wong J. Writing - review & editing: Wong J, Cheung GSP, Zhang C, Fok EHW. All authors read and approved the final manuscript.
DATA SHARING STATEMENT
The datasets are not publicly available but are available from the corresponding author upon reasonable request.