Healing and long-term prognosis of root fractured permanent teeth: a retrospective longitudinal study

doi:10.21203/rs.3.rs-3313509/v1

Aim

The purpose of the present study was to evaluate permanent teeth with post-traumatic transversal root fractures for their initial healing modality, the effect of candidate predictors and their long term prognosis.

Methodology:

A retrospective longitudinal clinical study was conducted to evaluate records from patients bearing transversal root fractures in permanent teeth in order to radiographically assess short-term healing and non-healing events in the fracture line, their prognostic factors and their relationship with long term outcomes. The inter-fragmentary tissues were classified as healing: hard tissue (HT), connective tissue (CT) or connective tissue and bone (CT + B) and non-healing: interposition of granulation tissue (GT). A competing risk survival analysis was conducted to estimate the hazards of healing and non-healing events and the effect of demographic, clinical and treatment variables was assessed using the subdistribution regression model (Fine & Gray).

Results

Radiographic findings showed 51.9% of healing in the short-term being strongly influenced by the presence and type of concomitant injuries to the coronal fragment. Teeth with concomitant crown fractures (sHR 24.38, 95% CI [3.16–188.3], p = 0.0022), luxations with dislocations (sHR 10.58, 95% CI [1.37–81.9], p = 0.0240) and subluxations (HR 9.66, 95% CI [1.14–81.7], p = 0.0370) were more likely to present non-healing of root fractures in the short-term. The healing rate in the long-term was of 70.1%, most of them with interposition of bone and connective tissue. The only predictive factor for the long-term prognosis of root fractured tooth was the occurrence of short-term healing ((sHR 4.15 [2.15–8.01] p < 0.001).

Conclusion

Healing at the fracture site was the most frequent outcome, both in the short-term and in the long-term. Short-term healing modality was strongly influenced by the presence and type of concomitant injuries to the crown fragment and was also predictive for long-term prognosis.

Clinical relevance:

Post-traumatic transversal root fractures have a positive prognosis in the long-term, supporting therefore, a more conservative approach for these teeth before considering more radical treatments.

transversal root fractures

competitive risks

healing

pulp necrosis

Root fractures account for 1.2–7% of all TDI in permanent teeth. They result from a frontal impact over mature teeth, with complete root development, inducing shearing forces that are dissipated producing the fracture plane. Since the resultant force vector is directed through the fracture line, the apical root fragment is usually not affected. On the other hand, concomitant injury to the coronal fragment, such as crown fractures, subluxation, extrusive or lateral luxation and avulsion are frequent resulting in damage to the mineralized tissues (cementum and dentin), to the pulp, and adjacent supportive tissue (periodontal ligament and alveolar bone) at the line of the fracture and along the coronal fragment [1].

Healing process takes place concurrently and independently, at both aspects of the fracture, the pulp and the periodontal ligament (PDL), and can lead to different outcomes depending upon the extension of injury. If there is minor damage, pulp reactions will create a dentine bridge that unites and stabilizes fragments and is followed by deposition of cementum by cells derived from the adjacent periodontium allowing the repair of the fracture site with hard tissue. In the case of rupture or severe stretching of the pulp at the level of the fracture, a revascularization of the coronal pulp is initiated. In this interim period, periodontally derived cells can reach the fracture site and repair takes place through the interposition of connective tissue and/or bone. However, if bacteria gain access to the fracture line, they shift the course of the healing process promoting pulp necrosis, infection of the coronal root canal and formation of inflammatory granulation tissue between the two root fragments and impairing, therefore, repair of the fracture line [1, 2]. These healing and non-healing events and their resulting tissues, diagnosed through imaging exams, are the base for the classification proposed by Andreasen and Hjørting-Hansen [3] and were confirmed by many clinical studies [4–9]. It was also suggested that these responses are influenced by a number of factors such as the root fracture location [7, 9–12], the stage of root development [5, 9 14–16] and the presence of concomitant injuries in the coronal fragment [5, 10, 12, 14, 17]. Treatment related factors like optimal repositioning [14, 18], splinting type and timing [5, 7, 10, 11, 18] and the use of systemic antibiotic therapy [16] were also implicated in the healing patterns after root fractures with inconclusive results.

The categories propsed by Andreasen & Hørting-Hansen [3] of immediate and short-term responses, as well as their prognostic factors, are well documented in the clinical literature [4, 5, 7–10, 12, 14, 16, 17, 19], but prognostic information regarding their behavior in the long term is still scarce. The available evidence is mainly derived from the material collected at the Eastman Dental Institute in Stockholm and demonstrated high success of conservative endodontic treatment of the coronal fragment of teeth with non-healed fractures in the middle or apical part of the root [8, 9]. Survival of root fractured teeth has been demonstrated to be high in the long term, being the cause of tooth loss dependent on the type of healing and the location of the fracture [7–9, 12, 13]. An understanding of the long-term prognosis of healed and non-healed intra-alveolar transversal root fractured teeth is important to provide accurate prognostic information to support treatment decision-making. The aim of the present study was, therefore, to evaluate permanent teeth with post-traumatic transversal root fractures for their initial healing modality, the effect of candidate predictors and their long term prognosis.

A retrospective cohort study was conducted to evaluate the prognosis of root-fractured teeth from patients referred for treatment at the XXXXXXXX Clinic at the XXXXXXXXXXXX, XXXXXXXXXX from 1995 to 2022. Patients’ records and radiographs were reviewed and written informed consent to use data from the protocols was obtained from all patient’s and/or their caregivers. The study received approval from the Committee on Ethics in Research of the XXXXXX (COEP-XXXXXX − 4.384.144).

Study Sample

The sample included permanent teeth diagnosed radiographically with post-traumatic transversal root fractures. The inclusion criteria for this study were: (a) teeth with complete radiographic records from the time of injury and throughout the follow-up period; (b) teeth with adequate information regarding the traumatic event and emergency care; (c) teeth without radiographic signs of root resorption and periodontal disease prior to the injury. Teeth with incomplete data, previous history of trauma, avulsed and not reimplanted coronal fragment, with emergency treatment performed outside the HMOB, periodontal disease and crown-root fractures were excluded.

Treatment procedures:

Patients were first seen by specially trained oral surgeons from the staff of the Emergency Dental Service of Municipal Hospital Xxxxx Xxxxxx (XXXXX). Emergency care was performed according to a protocol adopted by both centers as follows: reposition of displaced coronal fragments and immobilization with a flexible acid-etch-wire composite splints diameter up to 0.016″. Concomitant crown fractures had the exposed dentin covered with calcium hydroxide cement and/or glass ionomer [1, 2]. After initial treatment patients were referred to the DTC at UFMG where enamel-dentin fractures were restored with composite resin or bonding the fragment to the tooth [20]. Immobilization was removed and patients underwent follow-up for monitoring root fracture outcome. Clinical and radiographic examinations were performed at each visit according to a standard protocol adopted at DTC – SD – XXXX including pulp sensitivity tests (thermal tests with refrigerant spray and hot gutta-percha and electrical tests), horizontal and percussion tests, palpation and mobility tests, visual inspection for identification of crown coloration, swelling and fistula [21]. The standardization of radiographs was based on previously defined criteria [22]. At each follow-up visit, periapical radiographs were taken, with the central beam directed within a maximum range of 15–20° of the fracture plane, using a small film (Kodak Ultra-speed DF 58, size 2), and film holders with a fixed object–focus distance of 33 cm (Cone; Maquira Dental Products, Maringa, PR, Brazil). For digital radiographs a phosphor plate system, VistaScan® (Durr Dental, Bietigheim-Bissingen, Germany), was used. All analog radiographs were taken at 70 kVp, 8 mA (Spectro 70X Electronic Dabi Atlante S/A – Medical & Dental Industry, Ribeirão Preto – SP, Brazil). For digital radiographs a Gendex (765DC, Paris), which operated at 65 KV and 7 mA for 0.2 s. was used. The software used to analyze the digital radiographs was DBSWin (Durr Dental AG, Bietigheim-Bissingen, Germany). Teeth diagnosed with short-term healing were maintained in follow-up, but with more spaced visits.

Non healed teeth due to interposition of GT in the fracture line, underwent endodontic therapy of the coronal fragment using CaOH dressing and/or MTA plug at the level of the fracture line and were filled with gutta-percha [1]. After completing the endodontic therapy these patients were examined at regular intervals to assess long-term outcome classified according to Andreasen’s criteria.

Data collection

Two trained and calibrated authors (XXX and XXX) collected retrospective data from case records: patient’s gender and age at the time of injury, position of the root fracture: cervical, middle or apical third; number of fracture lines (simple or multiple), presence and type of concomitant injury. Injury to the coronal fragment was classified as follows: absent, luxations with no displacement (concussion and subluxation), luxations with extrusive or lateral displacement and crown fractures with and without pulp exposure. Data regarding splinting timing, systemic antibiotic therapy (SAT) prescription and type of endodontic treatment in cases of non-healing was also recorded. The set of periapical radiographs was blindly evaluated by two authors (XXX and XXX) with regard to the stage of root development, position and number of fracture lines. Radiographic findings from follow-up examinations were used to classify short-term and long-term healing/non-healing events at the fracture line and adjacent bone classified as follows: healing with hard tissue (HT) - apical and coronal fragments were in close apposition and the fracture line barely distinguishable. Normality of periodontal ligament space and adjacent alveolar bone (Fig. 1A). Interposition of connective tissue (CT): radiolucent line separating the fragments with normal periodontal ligament space and adjacent alveolar bone (Fig. 1B). Interposition of connective tissue and bone (CT + B): a bony bridge is seen separating the fragments, with a normal periodontal space around both fragments with normal adjacent alveolar bone (Fig. 1C). Non-healing with interposition of granulation tissue (GT): widening of the fracture line with a radiolucent area that extends into the adjacent bone, loss of lamina dura (Fig. 1D) [3]. The onset time for short-term outcomes was calculated using the midpoint between the follow-up visit when the healing status at the fracture line was diagnosed for the first time and the previous one. The long-term fracture-healing type was assessed radiographically at the final follow-up visit. Kappa coefficient was used to assess intra and inter-examiner agreement for each of the measures performed and indicated excellent intra-examiner agreement for both examiners regarding the type of healing (0.90 and 0.94). Inter-examiner agreement was also good (0.78). Disagreements were defined after a consensus discussion.

Statistical analysis

A descriptive analysis was performed to verify frequency distributions of studied outcomes and covariates. A competing risk survival analysis was conducted and cumulative incidence function (CIF) curves were constructed for each of the short-term fracture healing modality. The effect of independent variables in short-term and long-term outcomes was evaluated through univariate analysis using the Fine-Gray subdistribution hazard model. For such purposes two categories were considered: healing: grouping the modalities HT, CT and CT + B; and GT was considered non-healing. Covariates with nominal p-values lower than 0.25 were included in a multivariate regression. The proportional hazards assumption was verified through the Schoenfeld residuals test and the level of significance was set at p < 0.05. Agreement between short and long-term healing modalities was assessed using the Kappa statistics. Statistical analysis was carried out using R software (version 4.2.1, Vienna, Austria, 2014).

Blinding:

The evaluation of the imaging exams and the statistical analysis were blinded to the clinical data.

The final sample, after applying the exclusion criteria, consisted of 85 patients (Fig. 2) (62 males, 71.3%; 25 females, 28.7%) with a mean age of 21years ± 14.7 years (range 7.7 to 61.9 years) at the time of trauma. A total of 104 root fractured permanent teeth were analyzed, including 87 (85.6%) central upper incisors, 2 (1.9%) lateral upper incisors,11 (10.6%) mandibular incisors, and 2 (1.9%) canines. The sample was majorly comprised of mature teeth (95.2%) and only 5 teeth presented incomplete root development (4.8%). Most of cases presented with single fracture lines (76%) and 25 cases presented with multiple fracture lines (24%). Sample distribution regarding the position of fracture line was the following: 36 teeth (34.6%) with apical root fractures, 60 teeth (57.7%) with fractures in the middle third of the root and 8 (9.4%) with cervical root fractures. Concomitant injuries were present in 73 cases being 5 (4.8%) of them concurrent alveolar process fractures and 68 (65.4%) concurrent injuries to the coronal fragment distributed as follows: 19 teeth (18.3%) with lateral luxations; 18 teeth (17.3%) with extrusions; 16 teeth (15.4%) with subluxations,13 teeth (12.5%) with enamel/dentin fractures and 2 teeth (1.9%) with concurrent lateral luxations and enamel dentin crown fractures. Splinting time ranged from 4 weeks to 2 years with a median timing of 59 days. SAT was prescribed in 11.4% of the cases (Table 1). Considering the whole sample, the median follow-up period was 3 years (range from 46 days to 20 years).

Healing modalities in the short-term were assessed in 88 teeth because 7 patients dropped out the treatment before the definition of initial outcome and were right-censored. Other 9 cases were left censored due to wide intervals between the definition of healing and the previous visit. HT was observed in 11 teeth (10.6%), interposition of connective tissue and bone in 23 teeth (22,1%) and interposition of only connective tissue was observed in 20 teeth (19.2%). Non-healing, with interposition of granulation tissue was seen in 34 teeth (32.7%). The distribution of short-term healing modality according to teeth group, stage of root development, concomitant TDI, SAT prescription, number and location of fracture lines and splinting duration is presented in Table 2. Figure 3 illustrates the estimated CIF curves for non- healing and each of the healing patterns in the initial period soon after trauma. The median time for the diagnosis of pulp necrosis (GT) was 2.6 months, with HT was 3.4 months and with CT + B was 3.5 months. The median time for the diagnosis of healing with CT was 5.6 months

From the initial sample of 104 teeth long-term prognosis of root fractured teeth was assessed for 79 teeth since 7 were right-censored, 5 teeth with multiple fracture lines developed GT and were extracted, 6 teeth had new trauma episodes and 7 teeth dropped out follow-ups after initial diagnosis of healing (2 teeth with HT, 3 teeth with CT + Bone and 2 teeth with CT). Long-term healing patterns were distributed as follows: HT was observed in 8 teeth (10.1%), interposition of connective tissue and bone in 40 teeth (50.6%) and interposition of only connective tissue was observed in 9 teeth (11.4%). Non-healing, with interposition of granulation tissue was seen in 22 teeth (27.8%).

The univariate analysis of potential predictors on the hazards of short and long-term prognosis after transversal root fractures in permanent teeth is presented in Table 3. Cases with concomitant alveolar process fractures and with incomplete root development were excluded from such analysis due to the small number of cases. Concomitant injuries to the coronal fragment were grouped into the following categories: absent, luxation with no displacement, luxation with lateral and/or dislocation and crown fractures. Short-term outcome was also tested as a covariate for the long-term prognosis. Risk variables that present nominal p values lower than 0.25 for at least one of the events (healing x non-healing) in the univariate analysis entered the final multivariate subdistribution regression models. For the short-term outcome, the final model demonstrated that the presence of concomitant injuries to the coronal fragment was significantly associated with a lower probability of root fracture healing. The magnitude of such effect was strongly dependent on the type of the injury to the coronal fragment being greater in those teeth with concomitant coronal fractures (sHR 24.38, 95% CI [3.16–188.3], p = 0.0022), followed by luxations with dislocations (sHR 10.58, 95% CI [1.37–81.9], p = 0.0240) and luxations without displacement (HR 9.66, 95% CI [1.14–81.7], p = 0.0370) (Table 4). Such effects are illustrated in Fig. 4. For the long-term outcome, the healing in the shor-term outcome was preditive of long term healing (sHR 4.15 [2.15–8.01] p < 0.001). Such an effect is also evident considering results from Kappa statistics demonstrating an overall agreement of 67.1% between short and long-term fractures healing modalities, in special HT and CT + Bone modalities (Table 5).

Pulp canal obliteration was observed in 9.1% of the cases in the short-term evaluation but was a frequent finding in the long-term evaluation being observed in both fragments in 30.3% of the cases and only in the apical fragment in 28 teeth (35.4%).

Table 1

– Sample distribution to according to demographic, clinical and treatment related covariates
Covariate	N(%)
Gender
Female	25(28.7)
Male	62 (71.3)
Patient’s age at trauma - years
Mean + SD ( range)	21years ± 14.7 (7.7–61.9)
Follow-up period - years
Median ( range)	3y (46 days – 20y)
Teeth group
Central Upper incisors	87 (85.6)
Lateral Upper incisors	2 (1.9)
Mandibular incisors	11 (10.6)
Canines	2 (1.9)
Stage of Root Development
Incomplete	5 4(.8)
Complete	99 (95.2)
Concomitant Injury
None	31 (29.8)
Crown fractures (enamel-dentin)	13 (12.5)
Subluxation	16 (15.4)
Extrusion	18 (17.3)
Lateral Luxation	19 (18.3)
Lateral Lux + Crown Fractures	2 (1.9)
Alveolar Process Fracture	5 (4.8)
Location of fracture line
Apical	36 (34.6)
Middle	60 (57.7)
Cervical (subcrestal)	8 (9.4)
Number of fracture lines
Single	79 (76)
Multiple	25 (24)
*Splinting time- days**
Median ( range)	59 (30–730)
Systemic Antibiotic Therapy**
No	44( 42.3)
Yes	12 (11.4)
• _{*Not applicable for 31 teeth}
• _{** data not available for 48 tetth}

Table 2

– Sample distribution of short-term outcome according to demographic, clinical and treatment related covariates *
		Healing		Non-healing
	HT n (%)	CT n (%)	CT + Bone n (%)	GT n (%)
Teeth group
Central Upper incisors	10 (90)	14 (70)	21 (91.3)	33 (97)
Lateral Upper incisors	0 (0)	1 (5)	0 (0)	0 (0)
Mandibular incisors	1 (10)	5 (25)	1 (4.3)	1 (3)
Canines	0 (0)	0 (0)	1 (4.3)	0 (0)
Stage of Root Development
Incomplete	0 (0)	2 (66.7)	1 (20)	0 (0)
Complete	11 (12.9)	18 (21.2)	22 (25.9)	34 (38.6)
Concomitant Injury
None	6 (27.3)	8 (36.4)	7 (31.8)	1 (4.5)
Crown fractures (dentin)	1 (9.1)	2 (18.2)	0 (0)	2 (72.7)
Subluxation	2 (13.3)	1 (6.7)	6 (40)	6 (40)
Extrusion	0 (0)	4 (26.7)	4 (26.7)	7 (46.7)
Lateral Luxation	2 (11.1)	5 (27.8)	5 (27.8)	6 (33.3)
Lateral Lux + Crown Fractures	0 (0)	0 (0)	0 (0)	2 (100)
Alveolar Process Fracture	0 (0)	0 (0)	1 (20)	4 (80)
Systemic Antibiotic Therapy**
No	5 (17.2)	11 (37.9)	7 (24.1)	6 (20.7)
Yes	2 (18.2)	1 (9.1)	3 (27.3)	5 (45.5)
Location of fracture line
Apical	4 (13.3)	1 (2.8)	9 (30)	16 (53.3)
Middle	7 (13.5)	13 (21.7)	14 (26.9)	18 (34.6)
Cervical (subcrestal)	0 (0)	6 (85.7)	0 (0)	0 (0)
Number of fracture lines
Single	8 (11.8)	12 (17.6)	19 (27.9)	29 (42.6)
Multiple	3 (15)	8 (40)	4 (20)	5 (25)
Splinting time
None	7 (21.9)	10 (31.3)	7 (21.9)	8 (25)
4 weeks	0 (0)	1 (14.3)	4 (57.1)	2 (25)
Up to 4 m	3 (18.8)	3 (18.8)	3 (18.8)	7 (38.9)
¬ 4 m	1 (3.0)	6 (18.2)	9 (27.3)	19 (54.3)
_{**data not available for 7 right- censored teeth and 9 left-censored teeth;}
_{** data not available for 48 tetth}
_{HT: Hard Tissue; CT: Connective Tissue; CT + Bone: Connective Tissue + Bone; GT: Granulation Tissue}

Table 3

Univariate analysis (Gray test)of the effect of clinical an treatment related factors on the hazards of short and long-term outcome after transversal root fractures in permanent teeth

	Healing	Non-healing	Healing	Non-healing
Gender	0.52*	0.76	0.45	0.32
Patient’s age	0.38	0.73	0.41	0.56
Location of fracture line	0.12	-	0.085	-
Number of fracture lines	0.14)	0.17	0.26	0.79
Concomitant Injury	0.003	0.01	0.47	0.19
Splinting time	-	-	0.4	0.53
Short-term healing	NA	NA	< 0.001	0.002
_{*p value}

Table 4

Final multivariate model of the effect of clinical an treatment related factors on the hazards of short and long-term outcome after transversal root fractures in permanent teeth

	Healing sHR (95%CI) p value	Non-healing sHR (95%CI) p value	Healing sHR (95%CI) p value	Non-healing sHR (95%CI) p value
Concomitant Injury
None	reference	reference	reference	reference
Subluxations	0.48 (0.21–1.10) 0.08	9.76 (1.14–81.7) 0.037	0.87 (0.31–2.42) 0.79	2.12 (0.44–10.25) 0.35
Luxation with dislocation	0.41 (0.23–0.74) 0.003	10.6 (1.37–81.9) 0.024	0.98 (0.51–1.88) 0.96	0.83 (0.17–4.08) 0.82
Crown fractures	0.13 (0.03–0.50) 0.003	24.4 (3.16–188.3) 0.002	0.70 (0.26–1.87) 0.48	2.23 (0.55–9.00) 0.26
Short-term healing
Non-healing	NA	NA	reference	reference
Healing	NA	NA	4.15 (2.15–8.01) 0.000	0.04 (0.006–0.313) 0.002
sHR- subdistribution harzard; CI – confidence interval; NA – not applicable

Table 5

Agreement between short and long-term fracture healing modalities.
	Final Healing
Inicial Healing		CT	GT	HT	CT + B	TOTAL
	CT	6	0	1	7	14
	GT	3	16	0	10	29
	HT	0	0	7	2	9
	CT + B	0	0	0	18	18
	TOTAL	9	16	8	37	70
_{CT: Connective Tissue; GT: Granulation Tissue; HT: Hard Tissue; CT + Bone: Connective Tissue + Bone;}

The present study consisted in a retrospective cohort study to evaluate permanent teeth with post-traumatic transversal root fractures for their initial healing modality, their long term prognosis as well as their prognostic factors.

The main results of the present study show that healing of the fracture line was the most likely event to occur in both, the short and the long-term evaluation. Such findings corroborate the results of previous clinical studies, which demonstrated an initial healing rate of up to 77% of cases of teeth with post-traumatic horizontal root fracture [18, 14]. They are also in line with the few studies that demonstrated favorable prognosis of root-fractured teeth in the long-term [8, 9, 13]. Taken together these results highlight the need of a more conservative approach for root fractured teeth contrasting with the common sense that root-fractured teeth have an uncertain or poor long-term prognosis, leading many to view them as suitable candidates for replacement with dental implants [24, 23].

Among the factors that were suggested to affect initial healing events, the present study find out that the presence of concomitant injuries to the coronal fragment had a strong effect confirming previous studies [8, 5, 14, 18]. In addition, the present study demonstrated that the type of injury was also important being the worse prognosis related to the presence of concomitant crown fractures having, followed by luxation injuries with dislocation and subluxations of the coronal fragment, in this order of significance. The association of crown fractures on healing complications was also observed by Welboury et al [12] but was not observed in the study of Andreasen et al [16]: Such difference was explained by the fact that exposed dentin was always covered with temporary materials or restored with composite resins. Although in the present sample coverage of exposed dentin was also performed for all teeth, the time elapsed between trauma showing that healing was not related to crown fractures. An explanation for these differences may on the time elapsed between trauma and emergency care, when exposed dentin was covered, because not all patients had seek treatment immediately after trauma. The role of tooth displacement and increased mobility in worsening the prognosis can be related to the disruption of the PDL and the greater risk of pulp rupture at the fracture line, since it is well known that ideal healing depends on pulp survival [1].

The position of the fracture line was not relevant for short-term outcome in the present sample. Such results confirm previous literature demonstrating that the location of the fracture is not a factor affecting short-term healing outcomes [5, 7, 12, 14, 15, 16]. It's important to mention that the small number of cervical fractures, none of them with GT, precluded to include this covariate when considering the non-healing event during the regression analysis. Therefore, the present results can only be considered for healing events. The same caution should be taken when comparing the present results showing that fracture line position was not important for the long-term outcome. At first, these results may seem contradictory to those from the survival analysis conducted by Cvek et al [9] and Andreasen et al [13] demonstrating a higher rate of tooth loss among non-healed cervical root fractures when compared with apical and mid non healed fractures. However, it should be taken in mind that the present study considered healing and non-healing events in the long term and not tooth loss. Again the small number of cervical fractures and the fact that none of them presented with non-healing event impaired such analysis. On the other hand, the hazards of non-healed apical and mid root fractures healing apical and mid root fractures in the present sample presented considered in the present study did not performed a in the root fractures.

CIF curves were constructed for the short-term outcomes in order to account for the timing of each of the modalities of healing and non-healing observed soon after a root fracture. This information may be critical for clinical decision-making during the follow-up period. In this regard, the present study demonstrated that short-term healing modality could be identified within 6 months what is in line with the findings of Andreasen et al., [13]. It is noteworthy that the median time (5.6 months) until the diagnosis of healing with interposition of connective tissue was longer than that for the diagnosis of non-healing with interposition of granulation tissue, 2.5 months after trauma. Such difference can be explained by the presence of additional clinical signs associated with the pulp necrosis of the coronal fragment such as swelling or fistula, allowing a decisive diagnosis former than for connective tissue healing.

The long-term prognosis demonstrated higher rates of healing indicating a good prognosis even for those teeth initially diagnosed with non-healing and endodontically treated. Differently from the short-term healing events, in the present sample, the only parameter that could be associated with the long-term prognosis was the short-term outcome since no healed teeth changed such status during the follow-up period. Nonetheless, late pulp necrosis has been described in the literature as a result of a second [1]. Such cases in the present sample, had their data collected until the second event because necrosis was not a direct result of the trauma that caused the root fracture. An interesting finding is the fact that while teeth healed with HT and CT + bone kept such status along the follow-up period, teeth healed with connective tissue changed their healing pattern to CT + bone in the long-term, suggesting that CT is an early stage during healing with CT + bone [2]. From the 34 teeth non-healed in the short-term, 5 were extracted due to multiple fractures lines. The other 29 that underwent endodontic therapy of the coronal fragment and in 44.8% of the cases healing with CT + bone or CT was observed. It can be questioned the pertinence of keeping teeth with non-healed fractures, due to the potential chance of further bone loss. However, Cvek et al. [8] demonstrated success rates of up to 86% after using calcium hydroxide (CaOH) as intracanal dressing to induce the formation of a hard-tissue barrier at the apical opening of the coronal fragment. In the present study, different modalities of root filling (CaOH, CaOH + MTA or MTA) were adopted. Both approaches have pros and cons. CaOH therapy demands time and the periodic reapplication of calcium hydroxide, enhancing the risk of recontamination of the root canal. MTA technique, on the other hand is technically more intricate to perform and poses a risk of extruding the barrier material into the tissues between the two root fragments, leading to ongoing inflammation, foreign body reactions, or delayed healing [1]. Considering that failures in the present study were well distributed among these different approaches, and that the retrospective design is not adequate to measure the effect of different treatment strategies, one can only speculate regarding such effect. Other treatment related factors like splinting type and timing [5, 7, 10, 11, 18] and the use of systemic antibiotic therapy [16] could not be confirmed in the present sample. Again, the lack of randomization and great number of missing data might have interfered in the results. Therefore, this retrospective longitudinal study underscores the treatment approach's significance in the initial healing and long-term prognosis of root-fractured permanent teeth. Further controlled prospective studies, aiming to compare different therapeutic strategies are needed.

The current study has other limitations stemming from its retrospective design and the relatively small sample size within certain subsets of covariates. These factors hindered the comprehensive exploration of the research questions and introduced potential biases due to the reliance on patient records instead of tailored study protocols. Retrospective studies, by their nature, rely on historical data, making them susceptible to recall bias and incomplete or inconsistent documentation. Additionally, a limited number of cases in specific covariate categories can restrict the generalizability of findings and decrease the statistical power of the analysis.

To mitigate these limitations A significant number of healthcare professionals were involved in patient care and data entry during the study., the XXX – XX - XXXX implemented standardized forms and protocols for emergency care and follow-up visits since 1995, with consultations supervised by the same experienced professors in dental trauma. Retrospective studies often face the challenge of confounding factors, which can affect the results. Nevertheless, meticulous statistical techniques were employed to minimize their influence. However, it is crucial to recognize that unaccounted confounders may still exist as a limitation in retrospective studies.

In conclusion, the present study demonstrated that post-traumatic transversal root fractures have a positive prognosis in the long-term, supporting therefore, a more conservative approach for these teeth before considering more radical treatments.

Author Contribution: The authors confirm contribution to the paper as follows: Study conception and design: N.O.M Pedrosa, J.V. Bastos; Analysis and interpretation of results: N.O.M Pedrosa, R. A. Santos. Colosimo, J.V. Bastos; Draft manuscript preparation: N.O.M. Pedrosa, S.C. Coste, J.V. Bastos. All authors have revised and approved the final version of the manuscript.

Ethics approval statement: Study approval was obtained from COEP-UFMG - 4.384.144

Patient consent statement: All study participants and/or their legal guardians consented and signed the terms of consent and assent for participation in the study and authorization to use the data in the research. These documents will be made available upon request. Contact info: [email protected]

Data availability statement: The following elements will be made available upon request: data extracted from clinical records, data used for all analyses and the analytic code. Contact info: Natália de Oliveira Murta Pedrosa, [email protected].

Funding statement: No funding was obtained for this study, except for undergraduate and postgraduate students’ scholarship. This study was supported by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), a Brazilian funding agency.

Conflict of interest disclosure: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Permission to reproduce material from other sources: Not applicable.

Clinical trial registration: Not applicable.

Abbott PV (2019). Diagnosis and Management of Transverse Root Fracture. Journal of Endodontics, v. 45. doi: 10.1016/j.joen.2019.05.009.
Heithersay AO, Geoffrey S, Kahler B (2013). Healing responses following transverse root fracture: a historical review and case reports showing healing with (a) calcified tissue and (b) dense fibrous connective tissue. Dental Traumatology 29(4):253-265.
Andreasen JO, Hjørting–Hansen E (1967). Intraalveolar root fractures: radiographic and histologic study of 50 cases. Journal Oral Surgery, Copenhagen, 25:414-26.
Andreasen FM, Andreasen JO (1988). Resorption and mineralization processes following root fracture of permanent incisors. Endod Dent Traumatol, 4:202–14.
Andreasen FM, Andreasen JO, Bayer T (1989). Prognosis of root-fractured permanent incisors-prediction of healing modalities. Endodontics Dental Traumatology, Copenhagen, 5:11-22.
Andreasen (2001). Pulp healing acute following dental trauma: clinical and radiographic review. Pract Proced Sesthet Dent. 13:315-22.
Cvek M, Mejare I, Andreasen JO (2002). Healing and prognosis of teeth with intra-alveolar fractures involving the cervical part of the root. Dent Traumatol, 18:57–65.
Cvek M, Mèjare I, Andreasen JO (2004). Conservative endodontic treatment of teeth fractured in the middle or apical part of the root. Dent Traumatol 20:261–9.
Cvek M, Tsilingaridis G, Andreasen JO (2008). Survival of 534 incisors after intra alveolar root fracture in patients aged 7-17 years. Dent Traumatol 24:379-87.
Yates JA (1992). Root fractures in permanent teeth: a clinical review. International Endodontics Journal, Oxford, 25:150-7.
Caliskan MK, Pehlivan Y (1996). Prognosis of root-fractured permanent incisors. Endodontics Dental Traumatology, Copenhagen, 12(3):129-36. doi: 10.1111/j.1600-9657.1996.tb00111.x.
Welbury R, Kinirons M J, Day P, Humphreys K, Gregg T A (2002) Outcomes for root-fractured permanent incisors: a retrospective study. Pediatr Dent 2:98-102.
Andreasen JO, Lauridsen E, Guerds TA, Ahrensburg SS (2012). Dental Trauma Guide: A source of evidence-based treatment guidelines for dental trauma. Dental Traumatology, 28:142-47.
Cvek M, Andreasen JO, Borum MK (2001). Healing of 208 intra-alveolar root fractures in patients aged 7–17 years. In: Dent Traumatol, 17:53–62.
Feely L, Mackie IC, Macfarlane T (2003). An investigation of root-fractured permanent incisor teeth in children. Dent Traumatol 19:52-54.
Andreasen FM, Andreasen JO, Cvek M, Mejare I (2004). Healing of 400 intra alveolar root fractures. 2. Effect of treatment factors such as treatment delay, repositioning, splinting type and period and antibiotics. Dental Traumatology, Copenhagen, 20:203-11. doi: 10.1111/j.1600-9657.2004.00278.x.
Majorana A, Pasini S, Bardellini E, Keller E (2002). Clinical and epidemiological study of traumatic root fractures. Dental Traumatol. 18:77-80. doi: 10.1034/j.1600-9657.2002.180206.x.
Zachrisson BU.; Jacobsen I (1975). Long-term prognosis of 66 permanent anterior teeth with root fracture. Scandinavian Journal Dentistry Research, Copenhagen, 83:345-54.
Isaksson et al. (2020). Effect of splinting times on the healing of intra-alveolar root fractures in 512 permanent teeth in humans: A Scandinavian multicenter study. Dental Traumatology, Copenhagen, 37:672-676. doi: 10.1111/edt.12683.
Diangelis AJ et al. (2012). International Association of Dental Traumatology guidelines for the management of traumatic dental injuries: 1. Fractures and luxations of permanent teeth. In: Dent Traumatol, 28:2-12.
Bastos JV, Goulart EM, Côrtes MIS. (2014). Pulpal response to sensibility tests after traumatic dental injuries in permanent teeth. Dental Traumatology, 30(3):188-92.
Andreasen FM, Andreasen JO (1985). Diagnosis of luxation injuries. The importance of standardized clinical, radiographic and photographic techniques in clinical investigation. Endod Dent Traumatol, 1:160–9.
Rosenquist B, Grenthe B (1996). Immediate placement of implants into extraction sockets: implant survival. Int J Oral Maxillofac Implants 11:205–9.
Iqbal MK, Kim S (2008). A review of factors influencing treatment planning decisions of single-tooth implants versus preserving natural teeth with nonsurgical endodontic therapy. J endod 34:519–29.

No competing interests reported.

Healing and long-term prognosis of root fractured permanent teeth: a retrospective longitudinal study

Status:

Journal Publication

Version 1

Abstract

Aim

Methodology:

Results

Conclusion

Clinical relevance:

Figures

INTRODUCTION

METHODS

Study Sample

Treatment procedures:

Data collection

Statistical analysis

Blinding:

RESULTS

DISCUSSION

Declarations

References

Additional Declarations

Status:

Journal Publication

Version 1