Early prognostic indicators and outcome prediction model for replanted avulsed teeth.
Abstract: AIM: To identify early clinical variables that are most predictive of treatment outcome and to develop a model that will allow prediction of treatment outcomes based on these variables. STUDY DESIGN AND METHODS: A dental trauma database was used to randomly identify patients who had received treatment for avulsed teeth between 1998 and 2007. A data extraction form was designed and completed for each tooth. Demographic, diagnostic and treatment information recorded in the patient's records, in addition to radiographs, were viewed retrospectively. STATISTICS: The significance and the predictive power for each early clinical variable were assessed using a univariate logistic regression model. Only significant variables (p [less than or equal to] 0.05) were considered eligible for the prediction model and a c-index was then constructed for their respective predictive power (0.5 = no predictive power, 1.0 = perfect prediction). RESULTS: Of the original sample of 213 patients who had received treatment for avulsed teeth between 1998-2007 only 105 fulfilled the criteria for evaluation. Two models ('At first visit' and 'at initial treatment visits') were produced with a total of five variables that were significant and holding the greatest predictive power (high c-index): patient age (p=0.001, c=0.80); stage of root formation (p=0.001, c=0.76); storage medium (p=0.047, c=0.58); tooth mobility after dressing (p=0.001, c=0.70); and tooth mobility after splinting (p=0.003, c=0.70). These variables underwent multi-variate analysis and the final models had good predictive abilities (c-index of 0.80 and 0.74). CONCLUSION: These predictive models based on patient age, stage of root formation, storage medium, tooth mobility after dressing and tooth mobility after splinting were shown to have high predictive value and will enable a clinician to estimate the long term prognosis of avulsed and replanted teeth. It will enable planning for further treatment with a realistic view of outcome at an early stage.

Key words: Dental avulsion, prediction, outcome
Article Type: Report
Subject: Avulsion fractures (Diagnosis)
Avulsion fractures (Care and treatment)
Tooth fractures (Diagnosis)
Tooth fractures (Care and treatment)
Oral surgery (Patient outcomes)
Teeth (Roots)
Teeth (Physiological aspects)
Authors: Rhouma, O.
McMahon, A.D.
Welbury, R.R.
Pub Date: 08/01/2012
Publication: Name: European Archives of Paediatric Dentistry Publisher: European Academy of Paediatric Dentistry Audience: Academic Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2012 European Academy of Paediatric Dentistry ISSN: 1818-6300
Issue: Date: August, 2012 Source Volume: 13 Source Issue: 4
Geographic: Geographic Scope: United Kingdom Geographic Code: 4EUUK United Kingdom
Accession Number: 302113630
Full Text: Introduction

Dental avulsion injuries are distressing and when they occur can affect the daily life of both patients and their parents. Answers to a question from parents and patients about the prognosis for avulsed and replanted teeth remain unclear. Outcome prediction is useful for clinical decision making, family counseling, evaluation of quality of treatment, and medical resource allocation. Many variables and factors have been suggested in the literature that might affect the treatment outcome of these injuries [Cvek et al., 1990; Mackie and Worthington, 1992; Trope, 2002; Pohl et al., 2005a and 2005b], and many attempts have been conducted to study these variables [Andreasen, 1975; 1981; Weinstein et al., 1981; Al-Nazhan et al., 1995; Boyd et al., 2000; Kinirons et al., 2000; Andreasen et al., 2002, 2004; Cardoso et al., 2008; Stewart et al., 2008; Tzigkounakis et al., 2008; Hinckfuss and Messer, 2009; Petrovic et al., 2010]. Prior avulsion studies have presented survival analysis [Barrett and Kenny 1997; Pohl et al., 2005b; Petrovic et al., 2010] however, a mathematical model to estimate the long term prognosis of these injuries has not yet been established. The aim of this study was to: identify and to determine the early clinical variables (indicators) that were most predictive of treatment outcome and to develop a model(s) that would predict the outcome for replanted avulsed teeth based on these early clinical variables.

Material and methods

The study consisted of two parts. The first part was a review of patients' records gathering the target data together with assessment of each radiograph taken throughout the follow-up period. The second part was a review of all early related variables. These were subsequently added together to develop a model(s) that would predict the outcomes for replanted avulsed teeth based on early clinical variables. The outcome in all cases was assessed retrospectively and were categorised into two groups either failure or continued function based on the findings of the clinical and radiographic evaluation after at least one year of follow-up. The criteria in Table 1 are based upon every clinician's, 'goal'; namely to retain an avulsed and reimplanted tooth in the mouth functioning for as long as possible and if possible until full growth is achieved. Maintaining the tooth and its surrounding bone for a few years can be considered a successful treatment in the growing patient [McIntyre et al., 2007; Trope, 2011].

The study sample. The study population comprised patients treated within the Paediatric Department of Glasgow Dental Hospital and School, who had sustained traumatic injuries leading to avulsion of anterior permanent tooth/teeth. The department trauma database was used to identify all patients who had received treatment for dental avulsion in the period 1998-2007. For each patient only one avulsed tooth was randomly selected to be recruited in the study. For inclusion in the study teeth had to have been followed for more than one year after avulsion and replantation or until failure. Teeth were excluded from the study if: they had not been replanted; were lost to follow-up; had been followed for less than a year; had sustained a root or crown fracture; had been autotransplanted; or had sustained subsequent trauma after replantation.

Data elements. Diagnostic and treatment information recorded in each patient's records and on the formatted structured history trauma sheet in the records, in addition to radiographs, were the source of the data for this study. In order to aggregate all the target information from the patients' records, an extraction data form was designed to include all the early clinical variables including the patient's demographics. For each tooth, an extraction data form was completed and given a unique ID number and coded for computer entry. Confidentiality of patient information was considered throughout the study, as each patient's name and contact details were kept concealed. The data was retrieved by one of the authors (OR) and randomly reviewed by a senior consultant under standardised conditions.

Radiographic evaluation. Each tooth was evaluated for the presence of any radiographically detectible pathological changes including root resorption on a standard viewing box. Subsequently, each set of radiographs for each patient was analysed according to the chronological order of the visits until the pathological changes were detected. After determining the type and progress of root resorption, each radiograph was analysed till the tooth had either been retained or lost.

With regard to validity and reproducibility assessment for data extraction and to check for bias in data recorded in this study, a sample from the records was reviewed randomly by a senior consultant under standardised conditions. Consensus agreement was reached for the collection of the data and the clinical and radiographic outcomes.

Statistical methods. SPSS statistical software program 15.0 Standard version was adopted for use in data analysis. The data were subsequently coded, entered, processed and then analysed. Each variable in the study was examined initially by frequency tabulation. The variables were pooled into categories that represented natural grouping, whilst attempting to retain reasonable numbers of subjects in each category. The results for each variable were then expressed in the form of Pearson's Chi-square ([chi square]) test and odds ratios (OR) with 95% confidence intervals. In order to assess the individual significance for each variable a univariate logistic regression model was employed. Only those variables that were significant at the 5% level were considered for further assessment. For each variable the predictive ability was assessed using c-index. The c-index is equal to the area under the receiver operating characteristic (ROC) curve, which is commonly used as a summary index of the model performance [Harrell et al., 1984 and 1996]. Generally the area under the curve ranges from 0.5 to 1.0 and provides a measure of the model's ability to discriminate between those subjects who experience the outcome of interest versus those who do not [Altman and Bland, 1994; Hosmer et al., 2000]. A variable with no predictive power (no discrimination) has a c-index of 0.5 and a variable with perfect predictive power (excellent discrimination) has a c-index of 1.

Those variables that were identified as significant in this analysis were then analysed once more using forward stepwise logistic regression to produce a parsimonious model that can predict treatment outcome. Given the nature and specificity of this type of trauma and its treatment these variables were pooled together to form two groups based on the consequences of the variable in the process (variables at 'first visit' or at 'initial treatment visits'). Therefore two models were produced that could predict the treatment outcome: Model 1- 'At first visit'; Model 2- 'at initial treatment visits'. All cases were included in this analysis apart from those subjects omitted because of a missing subject data point, regardless of the availability of the others.

Results

A total of 213 patients' records were studied. Of these 108 were excluded at the beginning of the study in accord with the exclusion criteria. There were 105 avulsed permanent teeth in 105 patients subsequently recruited. All patients were regular attenders in the period 1998-2007 with 60 male patients (57%) and 45 female patients (43%). The mean of patients ages at time of trauma was 10 years (SD [+ or -] 2.61 yrs), with a range of between 6-16 years. The trauma was most frequent in age group 6-9 years of age.

The most commonly traumatised teeth were the maxillary right and left central incisors (65 teeth, 62% and 31 teeth, 30% respectively). Falls and sporting accidents were the most common causes of trauma (42 teeth, 40% and 47 teeth, 45% correspondingly). Most of the injuries occurred in outdoor locations (67 teeth, 64%), while the second most frequent location was at school (19 teeth, 18%). Some patients sustained a complicated trauma comprising avulsion with enamel dentine fracture or with pulp exposure, while the majority (88 teeth, 84%) was avulsion only. In 42 patients (40%) had avulsed two or more teeth at the same occasion. There were no decoronation cases in the sample.

[FIGURE 1 OMITTED]

The mean follow-up period for the 105 teeth was about 4 years (SD [+ or -] 2.16yrs), ranging between 0.19 (failed) and 10.82 years. The mean number of attendances (visits) was 24 with a maximum of 60. Extra alveolar dry time (EADT) ranged between 0-420 min with a mean of 41 min. EADT was less than 60 min for 71 cases (68%).

The storage medium was varied. The majority of the cases (73 teeth, 70%) were stored in milk, whilst the remainder were stored either dry or in other media. For 48 teeth (46%) there were completely formed apices, 38 teeth (36%) had nearly completed roots and the remainder (19 teeth, 18%) were immature rooted teeth at the time of their presentation. The most common type of splint was composite/wire (73 teeth, 70%). In all, the fixation period for about 43 teeth (46%) of the cases was less than 10 days and 11-15 days for about 22 teeth (23%) of the cases. Mobility after splinting was recorded in 58 teeth.

Pulps were removed within 15 days in 48 teeth (46%) and only 4 teeth did not require pulpal removal as they revascularised. Non-setting calcium hydroxide (NS-Ca[(OH).sub.2]) was the most popular initial intra-canal dressing used (85%) compared with only 15% of cases with Ledermix. Mobility after commencement of root canal dressing was documented in 53 teeth (51%). The frequency and the number of times the root canal dressing were changed depended on both the stage of root formation and the dryness of the root canal.

Teeth with incompletely developed roots and teeth with canals which failed to dry required longer total durations of intra-canal dressings and more changes of dressings. Root resorption was observed in 81 cases (77%) and in more than 58 cases (55%) this was inflammatory root resorption. Replacement root resorption was observed in only 23 cases (22%). Based on the study assessment criteria (Table 1), the failure rate was 36 cases (34%), while continued function was recorded in 69 cases (66%).

Significance of study variables. The significance of all the variables in terms of p value and their predictive ability in terms of c-index are summarised in Table 2 in descending order of statistical significance.

Model production. The significant variables were selected for further analysis to produce an appropriate model for predicting outcome. Overall 82 of 105 teeth were entered into the final stage of analysis (model production process). There were 23 teeth omitted from the analysis because of missing data points in certain variables regardless of the availability of the others.

Model 1: At first visit. This model enables prediction of the outcome at the first day of trauma. Combinations of significant variables were selected and entered into the model (patient age 'age group', storage medium and stage of root formation). These three variables were entered together into one model (Table 3) which showed a good predictive ability (c-index of 0.80, Figure 1). The overall accuracy of the model to predict failure of the treatment was 77%. In summary the major factors influencing the outcome (treatment failure) at this stage of 'first visit' after dental avulsion and replantation would appear to be: younger age; dry storage medium; and immature roots.

Model 2: At initial treatment visits. This model enables prediction of long term outcome from early clinical information collected from 10-45 days after trauma. Combinations of significant variables were selected and entered into this model (tooth mobility after dressing and tooth mobility after splinting). These two variables were entered together into the model (Table 3) and showed good predictive ability (c-index of 0.74, Figure 1). The overall accuracy of the model to predict failure of treatment was 72%. In summary the major factors influencing the outcome (treatment failure) at this stage of 'initial treatment visits' after dental avulsion and replantation appear to be the recording of tooth mobility, either after splinting or after starting root canal dressing.

Discussion

The main observations arising from this study were the magnitude of risk that dry storage medium, immature roots and a younger age exerts on the outcome (treatment failure) at an early stage after dental avulsion and replantation (at 'first visit after trauma'). This is the first study to have looked at the effects of clinical variables on the outcome of avulsed and replanted teeth and to have developed models to predict treatment outcome. Two models were produced from the five variables holding the greatest predictive power (c-index), which underwent multivariate analysis. The final models were developed with good predictive abilities 'c-index' of 0.80 and 0.74 (c-index of 1.0 being perfect prediction), with an overall prediction accuracy of 77% and 72% correspondingly. These levels of prediction abilities and prediction accuracy validate the models and hence their abilities to predict treatment outcome (treatment failure).

The study was designed to pool some of the advantages of previous prospective and retrospective studies. Objective diagnostic information was consistently accessible from patient records and was not subject to a recall bias (reporting bias). Special data extraction sheets were created retrospectively from existing patients records and study subjects were traced until discharge to ascertain whether their tooth was still functioning or whether it had failed based on the criteria in Table 1. Teeth diagnosed as having failed were compared against teeth that were still functioning. The purpose was to determine whether the two groups differed in the proportion of teeth exposed to risk factors.

There is a concern about the bias caused by having the same researcher (one of the authors) performing the analysis and assessment. However the dentist abstracting the records and performing the analysis and assessment was not the treating dentist. The patient's record was reviewed and analysed in entirety according to the chronological order of the visits until the pathological changes were detected. After determining the type and progress of pathological changes, each radiograph was analysed till the tooth had either been retained or extracted. This systematic evaluation, according to the study protocol, ensured consistency of evaluation and eliminated the problem of inter-observer variability. Subsequently, the comparison between the radiographs was made by the main researcher and a random sample of radiographs was independently reviewed by a senior consultant under identical standardised conditions. The findings were compared, and the outcome was assessed, and a final consensus agreement was reached.

This study recruited any anterior permanent tooth sustaining traumatic injuries leading to avulsion. Although these teeth had been treated under three different treatment protocols (BSPD1998 [Gregg and Boyd, 1998]; IADT 2001 [Flores et al., 2001]; IADT 2007 [Flores et al., 2007]), the comparison of outcome rates using these three protocols did not show any significant difference. Nevertheless any avulsed teeth that were lost at the time of the accident, had a short follow-up period, were lost to follow-up, had root fractures, had been autotransplanted, or had sustained subsequent trauma after replantation were all excluded from the study. Although this may have resulted in loss of some power in evaluating coefficients for the logistic regression models, the sample sizes were sufficient to fit valid models.

Post-operative outcomes. Pohl et al. [2005a] based their results on the post-operative healing classified as: functional healing; replacement resorption; and infection-related (inflammatory) resorption. Petrovic et al. [2010] similarly classified post-operative outcomes according to the root resorption). Soares et al. [2008] relied on clinical and radiographic post-operative outcomes as complete success, acceptable success, uncertain success, or failure. During this present study, the clinical diagnosis was based upon whether the tooth pulp and the root had any pathological changes and/or periapical involvement. A tooth/root that was retained and continued functioning, in absence of an abscess or sinus tract of pulpal or endodontic origin, but ultimately planned for extraction (because of root resorption) when patient was in their late teenage years was classified as continuing in function. This agrees with McIntyre's and Trope's definition of success, that maintaining the avulsed and replanted tooth and its surrounding bone for a few years can be considered a successful treatment in the growing patient [McIntyre et al., 2007; Trope, 2011]. This may effectively minimise a positive association between a risk factor and outcome.

Odds ratios (OR) were calculated for all clinical diagnostic variables. Some of the variables showing no association appeared statistically significant such as extra-alveolar dry time, EADT. This is probably due to the effect of the classification system criteria or sample size and identifies the need for further studies to support the results. In the overall data analysis, (EADT) was not a factor associated with the outcomes of the treatment (n=82, p=0.102). However it is an important observation that the majority of successful cases had a short EADT. Given the nature and specificity of this type of trauma and its treatment these variables were pooled together to form two groups based on the clinical consequences. Only variables that were identified as the strongest predictors of outcome at an early stage after avulsion were considered in the first logistic regression model (at first visit) e.g. dry storage medium, immature roots and younger age. In the same way, variables that were identified as the strongest predictors of outcome at the initial treatment visits (10-45 days after replantation) were involved in a second logistic regression model, e.g. tooth mobility, either after splinting or after starting root canal dressing. In this study, recording of tooth mobility greatly reduced a favourable prognosis.

Conclusions

These predictive models based on patient age, stage of root formation, storage medium, tooth mobility after dressing and tooth mobility after splinting were shown to have high predictive value and will enable clinicians to estimate the long term prognosis of avulsed and replanted teeth. It will enable planning for further treatment with a realistic view of outcome at an early stage.

References

Al-Nazhan S, Andreasn JO, Al-Bawardi S, Al-Rouq S. Evaluation of the effect of delayed management of traumatized permanent teeth. J Endod 1995; 21: 391-393.

Altman DG, Bland JM. Diagnostic tests 3: receiver operating characteristic plots. Br Med J 1994; 309: 188.

Andreasen JO. The effect of splinting upon periodontal healing after replantation of permanent incisors in monkeys. Acta Odontol Scand 1975; 33: 313-323.

Andreasen JO. Effect of extra-alveolar period and storage media upon periodontal and pulpal healing after replantation of mature permanent incisors in monkeys. Int J Oral Surg. 1981; 10: 43-53.

Andreasen JO, Andreasen FM, Skeie A, Hjorting-Hansen E, Schwartz O. Effect of treatment delay upon pulp and periodontal healing of traumatic dental injuries--a review article. Dent Traumatol 2002;1 8: 116-128.

Andreasen JO, Andreasen FM, Mejare I, Cvek M. Healing of 400 intra-alveolar root fractures. 2. Effect of treatment factors such as treatment delay, repositioning, splinting type and period and antibiotics. Dent Traumatol 2004; 20: 203-211.

Barrett EJ, Kenny DJ. Survival of avulsed permanent maxillary incisors in children following delayed replantation. Endodont Dent Traumatol 1997; 13: 269-275.

Boyd DH, Kinirons MJ, Gregg TA. A prospective study of factors affecting survival of replanted permanent incisors in children. Int J Paediatr Dent 2000; 10: 200-205.

Cardoso M, Rocha MJ. Identification of factors associated with pathological root resorption in traumatized primary teeth. Dent Traumatol 2008; 24: 343-349.

Cvek M, Cleaton-Jones P, Austin J et al. Effect of topical application of doxycycline on pulp revascularization and periodontal healing in reimplanted monkey incisors. Endod Dent Traumatol 1990; 6: 170-176.

Flores MT, Andreasen JO, Bakland LK et al. Guidelines for the evaluation and management of traumatic dental injuries. Dent Traumatol 2001; 17: 193-198.

Flores MT, Andersson L, Andreasen JO et al. Guidelines for the management of traumatic dental injuries. II. Avulsion of permanent teeth. Dent Traumatol 2007; 23: 130-136.

Gregg TA, Boyd DH. Treatment of avulsed permanent teeth in children. UK National Guidelines in Paediatric Dentistry. Royal College of Surgeons, Faculty of Dental Surgery. Int J Paediatr Dent 1998; 8: 75-81.

Harrell FE, Lee KL, Califf RM, Pryor DB, Rosati RA. Regression modelling strategies for improved prognosticprediction. Stat Med 1984; 3: 143-152.

Harrell FE, Lee KL, Mark DB. Multivariate prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med 1996; 15: 361-387.

Hinckfuss SE, Messer LB. Splinting duration and periodontal outcomes for replanted avulsed teeth: a systematic review. Dent Traumatol 2009; 25: 150-157.

Hosmer DW, Lemeshow S. Applied logistic regression, New York; Chichester, John Wiley & Sons; 2000.

Kinirons MJ, Gregg TA, Welbury RR, Cole BO. Variations in the presenting and treatment features in reimplanted permanent incisors in children and their effect on the prevalence of root resorption. Br Dent J 2000; 189: 263-266.

McIntyre JD, Lee JY, Trope M, Vann WF. Management of avulsed permanent incisors: A comprehensive update. Pediatr Dent 2007; 29: 56-63.

Mackie IC, Worthington HV. An investigation of replantation of traumatically avulsed permanent incisor teeth. Br Dent J 1992; 172: 17-20.

Petrovic B, Markovic D, Peric T, Blagojevic D. Factors related to treatment and outcomes of avulsed teeth. Dent Traumatol 2010; 26: 52-59.

Pohl Y, Filippi A, Kirschner H. Results after replantation of avulsed permanent teeth. II. Periodontal healing and the role of physiologic storage and antiresorptive-regenerative therapy. Dent Traumatol 2005a; 21: 93-101.

Pohl Y, Wahl G, Filippi A, Kirschner H. Results after replantation of avulsed permanent teeth. III. Tooth loss and survival analysis. Dent Traumatol 2005b; 21: 102-110.

Soares Ade J, Gomes BP, Zaia AA, Ferraz CC, de Souza-Filho FJ. Relationship between clinical-radiographic evaluation and outcome of teeth replantation. Dent Traumatol 2008; 24: 183-188.

Stewart CJ, Elledge RO, Kinirons MJ, Welbury RR. Factors affecting the timing of pulp extirpation in a sample of 66 replanted avulsed teeth in children and adolescents. Dent Traumatol 2008; 24: 625-627.

Trope M. Clinical management of the avulsed tooth: present strategies and future directions. Dent Traumatol 2002; 18: 1-11.

Trope M. Avulsion of permanent teeth: theory to practice. Dent Traumatol 2011; 27: 281-294.

Tzigkounakis V, Merglova V, Hecova H, Netolicky J. Retrospective clinical study of 90 avulsed permanent teeth in 58 children. Dent Traumatol 2008; 24: 598-602.

Weinstein FM, Worsaae N, Andreasen JO. The effect on periodontal and pulpal tissues of various cleaning procedures prior to replantation of extracted teeth. Acta Odontol Scand 1981; 39: 251-255.

O. Rhouma, A.D. McMahon, R.R. Welbury Department of Paediatric Dentistry, Glasgow Dental Hospital and School, Glasgow, Scotland.

Postal address: Dr O. Rhouma, Department of Paediatric Dentistry, Level 5, (Prof. Welbury office), Glasgow Dental Hospital and School, 378 Sauchiehall Street, Glasgow, G2 3JZ, United Kingdom.

Email: o.rhouma.1@research.gla.ac.uk
Table 1. Criteria used to determine treatment outcome

Continued function                  Failure

1. A tooth is still in the mouth    1. Tooth extracted before
with normal tooth function.         achieving full growth age.

2. A tooth is retained with no      2. Tooth is still in the mouth
clinical signs of infection up to   with presence of clinical signs
the age of full growth.             of infection and expected to be
                                    extracted before the age of full
3. Tooth extracted after achieved   growth.
full growth age and long term
treatment (Implant/Prosthesis)
already started.

Table 2. Univariate analysis of prognostic variables

                                       Outcomes

                              Continued
Variable                     function (%)   Failure (%)

Age at trauma (age group)
6-9 yrs old                   17 (41.5)      24 (58.5)
10-13 yrs old                 28 (71.8)      11 (28.2)
[greater than or equal to]    24 (96.0)       1 (4.0)
  14 yrs old

Stage of root formation
Immature                       9 (47.4)      10 (52.6)
Nearly complete               18 (47.4)      20 (52.6)
Complete                      42 (87.5)      6 (12.5)

Tooth mobility after
  dressing
No                            41 (83.7)      8 (16.3)
Yes                           24 (46.2)      28 (53.8)

Tooth mobility after
  splinting
No                            32 (82.1)      7 (17.9)
Yes                           30 (51.7)      28 (48.3)

Storage medium
Milk                          52 (71.2)      21 (28.8)
Dry                            6 (37.5)      10 (62.5)
Others                        11 (68.8)      5 (31.3)

Number of injured teeth
One                           37 (58.7)      26 (41.3)
Two or more                   32 (76.2)      10 (23.8)

EADT
<5 min                        19 (86.4)      3 (13.6)
5-15 min                       7 (58.3)      5 (41.7)
16-30 min                     15 (60.0)      10 (40.0)
31-60 min                      7 (58.3)      5 (41.7)
> 60 min                       4 (36.4)      7 (63.6)

Contamination
Not cited                     27 (75.0)      9 (25.0)
Washed off                    39 (60.0)      26 (40.0)
Not completely clean           3 (75.0)      1 (25.0)

Trauma location
Outdoor                       45 (67.2)      22 (32.8)
School                        14 (73.7)      5 (26.3)
Others                        10 (52.6)      9 (47.4)

Time to pulp access
1-15 days                     34 (70.8)      14 (29.2)
16-30 days                    10 (55.6)      8 (44.4)
>30 days                      21 (60.0)      14 (40.0)

Main cause of trauma
Fall                          29 (69.0)      13 (31.0)
Sport Accident                 28 (59.6      19 (40.4)
Others                        12 (75.0)      4 (25.0)

Type of root dressing
NS-Ca[(OH).sub.2]             54 (63.5)      31 (36.5)
Ledermix then                 11 (73.3)      4 (26.7)
  NS-Ca[(OH).sub.2]

Type of root dressing
Maxilla Rt central incisor    40 (61.5)      25 (38.5)
Maxilla Lt central incisor    22 (71.0)      9 (29.0)
Others                         7 (77.8)      2 (22.2)

Additional coronal trauma
Avulsion only                 59 (67.0)      29 (33.0)
Avulsion + crown fracture     10 (58.8)      7 (41.2)

Root canal after dressing
Dry and clean                 57 (65.5)      30 (34.5)
Pus (or any sign of            8 (57.1)      6 (42.9)
  infection)

Use of antibiotics
No                            32 (66.7)      16 (33.3)
Yes                           37 (64.9)      20 (35.1)

Splinting time
0-10 days                     28 (65.1)      15 (34.9)
11-15 days                    15 (68.2)      7 (31.8)
>16 days                      18 (62.1)      11 (37.9)

Variable                         OR (95% CI)        p value   c-index

Age at trauma (age group)                            0.001     0.80
6-9 yrs old                           1
10-13 yrs old                 0.28 (0.11 - 0.71)
[greater than or equal to]    0.03 (0.01 - 0.24)
  14 yrs old

Stage of root formation                              0.001     0.76
Immature                              1
Nearly complete                1 (0.33 - 3.01)
Complete                      0.13 (0.04 - 0.45)

Tooth mobility after                                 0.001     0.70
  dressing
No                                    1
Yes                          5.98 (2.35 - 15.21)

Tooth mobility after                                 0.003     0.70
  splinting
No                                    1
Yes                          4.27 (1.62 - 11.22)

Storage medium                                       0.047     0.58
Milk                                  1
Dry                          4.13 (1.33 - 12.80)
Others                        1.13 (0.35 - 3.64)

Number of injured teeth                              0.068     0.58
One                                   1
Two or more                   0.45 (0.19 - 1.06)

EADT                                                 0.102     0,67
<5 min                                1
5-15 min                     4.52 (0.85 - 24.11)
16-30 min                    4.22 (0.98 - 18.13)
31-60 min                    4.52 (0.85 - 24.11)
> 60 min                     11.10 (1.97 - 62.51)
                                                     0.296     0.52
Contamination
Not cited                             1
Washed off                     2 (0.81 - 4. 93)
Not completely clean           1 (0.92 - 10.87)

Trauma location                                      0.368     0.54
Outdoor                               1
School                        0.73 (0.23 -2.29)
Others                        1.84 (0.65 - 5.18)
                                                     0.415     0.60
Time to pulp access
1-15 days                             1
16-30 days                    1.94 (0.64 - 5.95)
>30 days                     1.62 (0.65 - 4 .06)

Main cause of trauma                                 0.452     0.54
Fall                                  1
Sport Accident                1.51 (0.63 - 3.64)
Others                        0.74 (0.20 - 2.75)

Type of root dressing                                0.466     0.51
NS-Ca[(OH).sub.2]                     1
Ledermix then                 0.63 (0.19 - 2.16)
  NS-Ca[(OH).sub.2]

Type of root dressing                                0.487     0.58
Maxilla Rt central incisor            1
Maxilla Lt central incisor    0.66 (0.26 - 1.65)
Others                        0.46 (0.09 - 2.34)

Additional coronal trauma                            0.515     0.51
Avulsion only                         1
Avulsion + crown fracture     1.42 (0.49 - 4.12)

Root canal after dressing                            0.545     0.55
Dry and clean                         1
Pus (or any sign of           1.43 (0.45 - 4.49)
  infection)                                         0.850     0.55

Use of antibiotics
No                                    1
Yes                           1.08 (.05 - 2.43)

Splinting time                                       0.902     0.51
0-10 days                             1
11-15 days                    0.87 (0.29 - 2.60)
>16 days                      1.14 (0.43 - 3.03)

Table 3. Variables in Model 1 and Model 2

                                   Outcomes

                        Continued
Variable               function (%)   Failure (%)       OR (95%CI)

Age at trauma
  (age group)
6-9 yrs old             10 (32.3)      21 (67.7)             1
10-13 yrs old           22 (73.3)      8 (26.7)     0.17 (0.06 - 0.52)
>14 yrs old             20 (95.2)       1 (4.8)     0.02 (0.00 - 0.20)

Storage medium
Milk                    41 (69.5)      18 (30,5)             1
Dry                      6 (37.5)      10 (62.5)    3.80 (1.20 - 12.04)
Others                   5 (71.4)      2 (28.6)     0.91 (0.16 - 5.14)

Stage of root
  formation
Immature                 7 (41.2)      10 (58.8)             1
Nearly complete         12 (42.9)      16 (57.1)    0.93 (0.28 - 3.07)
Complete                33 (89.2)      4 (10.8)     0.09 (0.02 - 0.35)

Tooth mobility after
  splinting
No                      33 (82.5)      7 (17.5)              1
Yes                     19 (45.2)      23 (54.8)    5.71 (2.06 - 15.78)

Tooth mobility after
  dressing
No                      35 (85.4)      6 (14.6)              1
Yes                     17 (41.5)      24 (58.5)    8.24 (2.84 - 23.91)
Yes                      37(64.9)      20(35.1)      1.08 (.05 - 2.43)
Gale Copyright: Copyright 2012 Gale, Cengage Learning. All rights reserved.