1 INTRODUCTION

Avulsion of permanent teeth is one of the most serious traumatic dental injuries. Avulsed permanent teeth should ideally be replanted as soon as possible,1-3 and long-term dry storage should be avoided as it will damage the periodontal ligament (PDL) cells and compromise normal healing.1 If an avulsed permanent tooth cannot be replanted at the site of accident, it is important to store the tooth in a suitable storage medium to preserve the viability of PDL cells prior to replantation.1 Numerous storage media such as saliva, saline, balanced salt solution and milk have been recommended and used in guidelines.1-8 Several studies have recommended milk as the best first choice.2, 4, 5, 7, 8 In addition to these storage media several alternative media such as propolis, cling film, rice water, coconut water, cell culture media and rehydration salts have also demonstrated some efficacy in preserving cell viability.9-19 A storage medium that is always available at the place of accident is the patient's saliva. It has been recommended as a storage medium and has been included in guidelines for several decades.2 Saliva, however, does not have the same optimal osmolality as milk or saline,4, 13 and it contains microorganisms that may damage the PDL in case of long-term storage. In contrast, saliva has been reported as a suitable storage medium for shorter periods of time in the patient’s mouth.4, 6, 9-11, 13, 16 However, iorder to avoid the risk of aspiration or swallowing the avulsed tooth, the patient can spit into a cup and the tooth can be placed in that cup.2, 3

In vitro studies have clarified how cells react to different storage conditions in general.1, 3, 5-9, 20 Other in vitro studies have suggested that storage in saliva may contribute to the healing of PDL cells.4, 6, 11, 13, 21

Several in vivo studies have been conducted in various experimental animals. It was found that storage in saliva for up to 2 h was as effective as storage in saline for the prevention of ankylosis.16 Other replantation studies in monkeys have reported the same protective effect.11 In another study in monkeys, 2 h storage in saliva prior to replantation resulted in PDL healing.9

Although experimental in vitro and in vitro studies are important to learn about the healing of cells and tissue, clinical studies on tooth replantation in humans are important to achieve a higher level of evidence and should be encouraged.22 Very few clinical studies have been conducted on human teeth stored in saliva prior to replantation.21, 23, 24 Case reports have reported that teeth stored in saliva for several hours after the accident healed without root resorption after long-term follow-up.14, 15 A long-term clinical study from Copenhagen on 400 replanted teeth indicated a potential for long-term survival of teeth stored in different media.25 A recent comprehensive review showed that there is insufficient evidence to recommend saliva versus alternative solutions for temporary storage of an avulsed tooth.7 Hence, more specific clinical studies on human teeth stored in saliva prior to replantation are needed. Clinical situations are typically characterized by an initial dry storage period followed by storage in a medium, so many factors must be taken into consideration. However, healing with normal PDL is less likely when the PDL has been severely damaged - for example, due to long-term drying.1, 3, 16, 23, 26-40 In such situations, bone will grow in contact with the root, and the tooth will be fused with the alveolar bone, so-called ankylosis.1, 36-40 This is a progressive condition resulting in gradual replacement of the tooth by bone, a process termed replacement resorption. Replacement resorption will progress until the tooth is completely replaced by bone. Ankylosis and replacement resorption cannot be treated, and hence, this condition cannot be stopped. For this reason, ankylosis has been shown to be the best predictor of unfavourable healing and an unfavourable outcome.22, 40

The aim of this study was to evaluate the risk of ankylosis for avulsed human teeth stored in saliva.

2 MATERIAL AND METHODS

The material for this study was extracted from an existing database at the Department of Oral and Maxillofacial Surgery, University Hospital, Copenhagen, Denmark, which included 400 replanted permanent incisors from 322 patients. Data were collected from 1965 to 1988. The permanent teeth were replanted and followed for up to 20 years.41 The methodology and results of this large material have been published elsewhere.22, 41-44 In the following, a summary of the methods of relevance for the present study is presented.

The standardized clinical protocol for avulsed teeth at the University Hospital of Copenhagen was followed. The protocol stated that the avulsed teeth were to be placed in physiological saline (0.9%) as soon as the patient arrived at the emergency ward. If the tooth was obviously contaminated, it was cleansed with a flow of saline from a syringe. The tooth was then replanted into its socket by digital pressure. No effort was made to remove the coagulum before replantation. Splinting was performed with either cemented orthodontic bands united with acrylic (rigid) or acid etched labial crown surfaces united with a flexible resin (Protemp® 3 M ESPE, Germany). Splinting periods ranged from 2 to 6 weeks. Phenoxymethylpenicillin, adjusted to the weight of the patient, was administered orally for 5 days, starting on the day of the injury. In teeth with complete root formation, endodontic treatment was initiated by placing calcium hydroxide inside the root canal within 14 days after the injury and the canals were later filled with gutta-percha followed by a coronal composite filling. Teeth with immature root development were observed, and endodontic treatment was performed if there were signs of pulp necrosis and infection or infection-related (inflammatory) root resorption. One week after the injury, the patients and their parents were contacted to check and further supplement the information given during the initial emergency service to achieve the highest possible accuracy of data about the injury conditions, such as storage conditions and storage time. Exclusion criteria were previous endodontic treatment, previous trauma, destruction of the crown caused by dental caries or restorations, the presence of a concomitant crown fracture with pulp exposure, root fracture or alveolar fracture. The standard follow-up programme included clinical and radiographic reviews after 1 week, 2–3 weeks, 4 weeks, 6 weeks, 6 months and 1, 2, 3, 5, 10, 15 and 20 years. In case of suspected root resorption, an additional control was performed after 4 weeks. The mean follow-up period was 5.1 years, ranging from 7 months to 20 years. At the time of the injury, the following parameters were recorded: gender, age, cause of injury, date and time of injury, number of injured teeth, the condition of the supporting tissue, fractures of the teeth, type of storage medium and length of storage. Extra-oral dry time was defined as the time interval from when the tooth was avulsed from its socket to the time when the tooth was placed in saliva. For each tooth, clinical information from the time of injury and from follow-up examinations included tooth colour, tenderness to percussion, percussion sound, tooth mobility and tooth position in relation to the contralateral tooth.41-45 The stage of root development was determined by evaluation of radiographs from the initial examination and classified into one of the following six stages described by Moorrees et al46: ¼ root formation, ½ root formation; ¾ root formation; full root formation with wide open apex; full root formation with ½ closed apex; and full root formation with closed apex. The material in the present study was divided into two sub-groups: immature root development (stages 1–5) or mature root development (stage 6). Ankylosis-related resorption was diagnosed when one or more of the following clinical signs were present: high-pitch percussion sound, absence of mobility and infra-position of the tooth in growing individuals. Radiographic signs of replacement resorption later verified the clinical diagnosis of ankylosis. Mobility was measured using a modified Mühlemann periodontometer.47

For the present study, which focused on storage conditions and ankylosis, the following inclusion criteria were applied: tooth stored in saliva prior to replantation, presence of tooth-specific clinical information and radiographs from the time of injury and the subsequent controls according to a standardized protocol. Furthermore, a minimum of 1-year follow-up was carried out unless early complications had necessitated extraction before that time. All data used in the present study were obtained in a clinical context as part of a standardized treatment regime with full acceptance from the parents (oral consent) and are fully in line with the World Medical Association's Declaration of Helsinki from 2013. By Danish law, this study is considered a ‘quality assurance follow-up study’ (i.e. all data were obtained in a clinical context and/or as part of a standardized treatment protocol), and the study thus did not qualify for evaluation by a research ethics committee. In the present data set, it was not possible to track any individual patients.

Statistical methods: Survival analysis was applied to analyse the risk of ankylosis because the follow-up period varied among patients. The risk of ankylosis was estimated using the Aalen–Johansen method48, 49 accounting for the competing risk of tooth loss. The 3-year risk of ankylosis with 95% confidence limits is reported separately for teeth with immature and mature root development and in relation to the dry-storage period. It has previously been reported that the two most decisive factors for ankylosis are the length of dry time and the stage of root development. Due to the limited number of patients in the study, it was therefore decided to only include these two factors in the multivariate analysis. Risk factors for ankylosis were analysed by cause-specific Cox regression models for the hazard rate of ankylosis and the hazard rate of tooth loss without ankylosis, respectively.50 Reported were hazard ratios and 3-year risks of ankylosis according to all combinations of risk factors. Risk factors included in the analysis were the length of dry time and the stage of root development. The effect of increasing the wet storage time on the hazard rate of ankylosis was analysed in the subset of the data including only teeth for which the dry storage time was below 6 min. The level of significance was set at 5%. All analyses were performed with the statistical software RR Core Team 2020.49

3 RESULTS

The material comprised 74 (54 male and 20 female) patients with 89 avulsed and replanted teeth that had all been placed in saliva prior to replantation. The median age of the patients was 13.0 years, ranging from 6 to 36 years. Table 1 shows the distribution of the saliva-stored teeth related to patient, age, gender and tooth type. Most teeth were maxillary central and lateral incisors (84 teeth). The period the avulsed tooth was placed in saliva (wet storage) ranged from 1 to 150 min, with a mean period of 35 min. The distribution of mature and immature teeth in relation to dry-storage period is presented in Table 2. Mature teeth comprised 82% of the material.

TABLE 1. Distribution of patient age and gender Male n (%) Female n (%) Total n (%) p-value Age 0–10 years 19 (28.4) 5 (22.7) 24 (27.0) Age 11–15 years 21 (31.3) 9 (40.9) 30 (33.7) Age more than 15 years 27 (40.3) 8 (36.4) 35 (39.3) .7013 TABLE 2. Distribution of teeth in relation to stage of root development and dry time period Immature n (%) Mature n (%) Total n (%) p-value Dry time period 0–5 min 9 (56.2) 44 (60.3) 53 (59.6) 6–15 min 0 (0.0) 12 (16.4) 12 (13.5) >15 min 7 (43.8) 17 (23.3) 24 (27.0) .0952

Figure 1 shows the overall risk of ankylosis over time. Overall, 40% of the teeth that had been stored in saliva prior to replantation showed healing without ankylosis. In most cases, ankylosis was diagnosed during the first year after replantation. However, some teeth were not diagnosed as having ankylosis until 2–3 years after the trauma. After 3 years, no new cases of ankylosis were diagnosed. Progress of ankylosis with replacement resorption over time was seen in all cases where ankylosis was diagnosed and no cases were reversible.

The overall risk of ankylosis (y-axis) of avulsed permanent teeth stored in saliva prior to replantation. Number of remaining teeth by follow-up time (x-axis). N = 89

Figures 2–4-2–4 show examples of cases where teeth which had been stored in saliva healed without ankylosis even after the teeth had been stored in saliva for long periods of time.

Ten-year-old boy who fell and avulsed the left mandibular incisor, which fell to the floor and was picked up and placed in the mouth where it remained until replantation 40 min later (A) Radiographs show the condition immediately after replantation; (B), atreviews 1 and 5 years after the trauma; (C, D), there is no sign of ankylosis, radiographically or clinically

Six-year-old boy who fell from a bicycle and avulsed two erupting central incisors (A and B). They were avulsed into the oral cavity and remained there until replanted 150 min later and splinted (C). After 1 month, the left central incisor showed radiographic signs of pulp necrosis and infection; (D) the tooth was opened, and half of the pulp was found to be necrotic and therefore extirpated and the root canal was filled with calcium hydroxide. Following apical closure, the root canal was subsequently filled with composite and a bonding material. At the 7-year clinical and radiographical control, there was no sign of ankylosis of the two incisors (E)

Ten-year-old girl who fell and avulsed both central incisors. The central incisors adhered to the oral gingiva, but the roots were exposed to the oral cavity, and the left lateral incisor was not retrieved (A). The two central incisors were replanted and splinted for 2 weeks. The radiographs were taken after (B) 1 month, (C) 1 year, (D) 5 years, (E) 10 years and (F) 20 years. There are no radiographic or clinical signs of ankylosis

Table 3 shows the risk of ankylosis estimated after 3 years of observation in relation to dry-storage and stage of root development. Figure 5A shows the risk of ankylosis over time in relation to the stage of root development and Figure 5B shows the risk of ankylosis over time in relation to the dry-time period.

TABLE 3. Cox regression analysis Hazard ratio CI 95% p-value Dry time period 0–5 min Ref 6–15 min 2.25 [1.03;4.89] .0417 >15 min 3.40 [1.81;6.37] <.001 Stage of Root development Immature Ref Mature 2.37 [1.02;5.46] .0436

A. Risk of ankylosis (y-axis) of avulsed permanent teeth stored in saliva prior to replantation related to root development. Number of remaining teeth over follow-up time (x-axis). N = 89. B. Risk of ankylosis (y-axis) related to dry time storage (0-5, 6-15, >15 min) prior to placing the tooth in saliva. Number of remaining teeth over follow-up time (x-axis). N = 89

Table 4 shows the Cox regression analysis including the risk factors dry-storage period and stage of root development. Increasing the dry storage time from 5 min or less to 6-20 min significantly increased the risk of ankylosis (HR: 2.25 (95% CI:1.03–4.89) p = .04). Increasing the dry time to more than 20 min increased the risk even more (HR: 3.40 (95% CI: 1.81–6.37) p < .001). Furthermore, teeth with mature root development were significantly more frequently affected by ankylosis than teeth with immature root development (HR: 2.4 (95% CI: 1.0–5.5), p = .04).

TABLE 4. Estimated risk of ankylosis after 3 years Root development Dry time (min) Risk 3 years (95% CI) Immature 0-5 23.7 [8.6;42.9] Immature 6-15 NA Immature >15 60.5 [29.6;81.2] Mature 0-5 47.4 [32.8;60.7] Mature 6-15 76.8 [45.7;91.5] Mature >15 89.3 [68.0;96.7]

A separate analysis of the effect of wet time was conducted for 53 teeth that had been stored dry for 5 min or less before being placed in saliva. Table 5 shows the Cox regression analysis including wet time and stage of root development. A linear trend was found, showing that one additional minute of wet time increased the hazard rate for ankylosis by approximately 1% (CI = [0%, 2%], p = .052. Figure 6 illustrates the trend of development of ankylosis between teeth with mature and immature root development over time in this sub-group.

TABLE 5. Cox regression of teeth with dry time less than 5 min (n = 53) Hazard ratio CI 95% p-value Wet time 1.01 [1.00;1.02] .0516 Stage of Root development Immature Ref Mature 2.67 [0.61;11.72] .1923

Trend of development of ankylosis during 3 years after replantation related to wet time in saliva for teeth with a dry time of 5 min or less. N = 53

4 DISCUSSION

The present study shows that saliva is a realistic storage medium for avulsed teeth before replantation.

One strength of the present study is that the material is the largest follow-up material reported in the literature on saliva-stored human teeth prior to replantation. Another strength is thatthe patients included in the study were treated and followed according to a standardized protocol. To ensure that the information on storage medium and storage period was as accurate as possible, the patient was recalled, generally after 1 week, for a review of the information provided at the time of the accident. This was the responsibility of a single person. In this manner, the number of misunderstandings and amount of incorrect information were reduced. Another strength is that the healing of the replanted tooth was followed for an extended period. In most teeth, ankylosis was diagnosed during the first year, and all teeth with ankylosis were diagnosed within the first 3 years. This is in line with reports from previous studies.1 Replacement resorption progressed with time, which is also in accordance with the literature.1, 45

A weakness of a long-term study such as the present one is that the recommended treatment may have changed over time – for example, the fixation period for the traumatized tooth varied from 6 weeks at the beginning of the study period in the 1960-s and 1970-s to later recommendations of 1–2 weeks. It has been shown that functional stimulation during the first weeks of healing may influence the development of ankylosis, and short-time splinting has hence been suggested.1 However, this variation in splinting period affected the groups equally in this study.

The present study focused only on ankylosis. Hence, other factors such as age, splinting type and tooth type were not included in the analysis. However, these factors have been analysed and reported in other studies originating from the same database.25, 41-43

The clinical diagnosis of ankylosis was carried out by registering high percussion sound, absence of mobility and radiographic signs of replacement resorption. In the early stages, percussion sound and absence of mobility were seen. Later, radiographs revealed replacement resorption. The validity of these methods for diagnosis of ankylosis has earlier been assessed and documented.38 These diagnostic methods of ankylosis are used in clinical practice and have been reported in clinical studies.27, 45 Progress with replacement resorption could be seen in all cases where ankylosis was diagnosed.

There may be different reasons why saliva is a good storage medium. One unique factor related to saliva appears to be its general positive effect upon wound healing.51 A recent experimental study showed better wound healing in extraction sockets when saliva was present.21 Moreover, the ability of PDL cells to form colonies (chlonogenic capacity) has been studied in an in vitro study, which found that PDL cells stored in saliva showed a reduced chlonogenic capacity over time.6 If the chlonogenic capacity was higher than 3%, there was a chance of PDL healing. After 30 min, the capacity was 7.6%. At 60 min, the chlonogenic capacity was 1.5% after storage in saliva.6 Another factor may be a high number of bacteria which can possibly make saliva a less suitable storage medium during the extra-alveolar period for the tooth before replantation.5 Saliva contains four times more aerobic bacteria and two times more anaerobic bacteria than milk.5 In the same study, saliva-stored monkey teeth showed many adhering bacteria on the root surface, regardless of the time they were stored in saliva, while teeth stored in milk had few adhering bacteria on their periodontal ligaments.5 A third factor is osmolality which has been shown to be an important factor in PDL cell survival. Saliva has a higher osmolality than water but not as high as milk. In experimental in vivo studies, Blomlöf et al showed that monkey teeth stored in milk prior to replantation can heal with normal PDL when the replanted teeth were stored in milk for up to 6 h. In saliva, the time was much shorter, and Blomlöf et al suggested that saliva may serve as a storage medium for up to 2 h.10 The results of the present study suggest that saliva can be recommended as an immediate temporary storage medium until a better storage medium - such as milk or physiological saline - is found near the accident site. However, the findings of the present study also show that saliva has a limited ability to preserve the PDL for longer periods. Already after 5 min, the risk of ankylosis is increased. Hence, the results of this study indicate that saliva should be considered as a temporary storage medium that should be replaced as soon as possible by a better, more physiological medium such as milk or saline. The results of the present study indicated a linear trend with ankylosis increasing by approximately 1% per minute when the tooth was placed in saliva prior to replantation. The results of this study indicate that avulsed teeth in humans have a higher risk of ankylosis than what has been reported earlier in studies of animal teeth when they were stored in saliva prior to replantation.10, 16 This may be due to the studies in laboratory animals being conducted under more controlled conditions than studies conducted in human clinical settings. Factors other than osmolality that promote healing may influence the findings in replanted saliva-stored teeth. Many animals lick their wounds, and this may possibly have a positive effect on wound healing. One study suggested that saliva is the primary factor in accelerating the oral wound-healing process.52 Moreover, saliva creates a humid environment in the oral mucosa that prevents oral tissues from dehydrating, which yields improved access to nutrition and increases the survival of the cells involved in the wound healing process. All these factors together accelerate re-epithelialization. Saliva also contains many different molecules that improve healing in the oral cavity, including proteins and peptides.52 One novel discovery is that saliva contains high amounts of tissue factors that probably contribute to wound healing by speeding up blood clotting.53 Growth factors are found in saliva, (e.g. epidermal growth factor), but it has been shown that most growth factors in human saliva occur in very small amounts and sometimes in an inactive form compared with saliva in animals.51, 54-56 Histatins and secretory leucocyte protease inhibitors have also been found in saliva and these help accelerate wound healing.51 Moreover, it has been shown that saliva is important for rapid healing of extraction sockets.2

The current results show that storage in saliva prior to replantation reduces the risk of ankylosis compared to dry storage. These results are consistent with some previous experimental animal studies and in vitro studies.4, 6, 9, 11, 16 In other studies, in which teeth were stored dry for 1 h or more before replantation, ankylosis was reported.9, 10, 16, 23, 32, 45 One study with replanted dog teeth showed that it did not matter if the teeth were stored dry for 20, 60 or 90 min before replantation because the PDL cells were already very damaged after 20 min, and ankylosis was reported.47 In another in vivo study with replanted monkey teeth, the results showed a clear association between ankylosis and teeth stored dry for 30 min prior to replantation.16 Hence, ankylosis must be expected in many teeth when the dry storage period is prolonged. However, a very recent study on replanted teeth stored dry for more than 60 min prior to replantation showed that ankylosis may be expected in as many as 85% of the teeth.22

Within the imitations of this study, with a low number of teeth in the material, the results indicate that teeth with immature root development may have a lower risk of ankylosis than teeth with a closed apex. The reason for this may be that, unlike teeth with completed root development, immature teeth have a thicker PDL layer.1, 22, 25 Another factor may be that immature teeth have an open apex allowing more vessels to provide a better blood supply. However, a prerequisite for this is that the pulp is not infected. In cases with an infected pulp, infection-related root resorption can rapidly destroy an immature tooth due to the wider root canals and dentine tubules compared to mature teeth.1, 28 Infection-related root resorption may have an influence on healing in the short term (e.g. during the first year) but can today be prevented or treated, while ankylosis is the main factor influencing the long-term prognosis and it is not possible to treat.

The dry storage time before the tooth was placed in saliva was shown to be a critical factor. Dry storage for 5 min or less before saliva storage resulted in