OUP user menu

Root repair after contact with mini-implants: systematic review of the literature

Matheus Alves Jr, Carolina Baratieri, Cláudia Trindade Mattos, Mônica Tirre de Souza Araújo, Lucianne Cople Maia
DOI: http://dx.doi.org/10.1093/ejo/cjs025 491-499 First published online: 26 April 2012


This systematic review identified and qualified the current evidence of dental root damage and repair after contact with mini-implants. The electronic databases Cochrane library, Ovid, Scirus, Scopus, and Virtual Health Library were used to search original articles from 1980 to December 2011. The inclusion criteria to select the articles were 1. randomized controlled trials and prospective clinical studies based on trials involving humans, 2. randomized controlled studies in animals, 3. use of mini-implants with a diameter less than 2.5 mm, and 4. root contact evaluation associated with the use of orthodontic mini-implants. Two authors independently reviewed and extracted data from the selected studies and a methodological quality assessment process was used to rank the studies classifying them as low moderate or high quality. The searches retrieved 579 citations. After initial selection, 17 studies were considered eligible and their full texts were assessed. Four of those were excluded because root damage was not evaluated and two were excluded because of overlapping samples. Eleven articles, nine in animals and two in humans, fulfilled the inclusion criteria. From these, two studies were ranked as presenting high methodological quality, eight were judged to be of moderate, and one of low quality. The evidence found suggested that the quality of root repair depends on the amount of damage caused by the mini-implant. When the damage is limited to the cementum or dentin, healing and almost complete and repair of the periodontal structure can occur. Mini-implants that injured the pulp were less likely to result in complete repair of the periodontal tissues.


Suggested as a possibility for skeletal anchorage, the mini-implants were first introduced in orthodontics by Creekmore and Eklund (1983) for intrusion of the maxillary incisors. Because of their small size, mini-implants can be inserted in different regions of the oral cavity. Mini-implants are contemporary orthodontic adjuncts, mainly used for situations where orthodontic mechanics such as mass movement of teeth, correction of severe overbite, retraction of anterior teeth with no anchorage loss (Park et al., 2001; Kawakami et al., 2004; Park and Kwon, 2004; Nojima et al., 2006), and molar intrusion for correction of open bites and control of the vertical dimension are required (Umemori et al., 1999; Bae et al., 2002; Park et al., 2003).

However, the use of orthodontic mini-implants can lead to complications such as mini-implant fracture, periimplant mucositis, ulceration of the mucosa, and root injury of the teeth adjacent to the implants. Among these complications, root injury and its sequels may be the most prejudicial for the patient's dental health and this is the most likely reason why clinicians hesitate to use this device. Kuroda et al. (2007) concluded that the proximity of mini-implants to the adjacent tooth root is the major risk factor for their failure. Some animal studies showed complete healing of minor damage to root tissue following implant removal, resulting in normal periodontal structure (Asscherickx et al., 2005; Bae, 2005). However, after more extensive injuries, root tissue did not heal fully (Bae, 2005) and that may lead to ankylosis.

It is therefore important for the clinician to understand the potential risks associated with mini-implant use and how to deal with potential complications. The purpose of this study was to systematically review the current literature looking for root damage after contact with mini-implants.

Materials and methods

The method used in this systematic review was based on the guidelines published in the PRISMA Statement focused on randomized trials and evaluations of interventions (Moher et al., 2009). The Cochrane Library, Ovid, Scirus, Scopus, and Virtual Health Library (VHL) databases were utilized to search original articles.

The search strategy included appropriate changes in the key words (mini-implant, mini-screw, micro-implant, micro-screw, teeth, and root) and followed each database syntax rules.

Furthermore, the following journals were searched manually: American Journal of Orthodontics and Dentofacial Orthopedics, Angle Orthodontist, Clinical Oral Implants Research, and European Journal of Orthodontics. Additionally, the reference lists of the retrieved articles were hand searched for publications that were missed in the database searches.

The inclusion criteria for selection were 1. randomized controlled trials and prospective clinical studies, 2. randomized controlled studies in animals, 3. use of mini-implants with a diameter less than 2.5 mm because larger screws would not be used in the interradicular regions, and 4. root contact evaluation associated with the use of orthodontic mini-implants. The exclusion criteria were technique articles, case reports, opinion articles, and reviews articles. No restrictions were placed on year, publication status, or language.

The titles and abstracts of all potentially relevant studies were reviewed. Search results which did not give sufficient information as to their significance to this study were also reviewed in full. Authors were contacted directly to obtain additional information when necessary. Each article was reviewed independently by two readers (M.A.J. and C.B.) and the information obtained was compared. Interexaminer conflicts were resolved by discussion of the relevant articles.

Articles that fulfilled the inclusion criteria were methodologically assessed for quality according to a modified version described by Feldmann and Bondemark (2006).

The following five variables were evaluated: sample size, study design, selection description, diagnostic methods, and follow-up. Adding up the score of the five variables, each study could maximally score 10 points and be categorized as presenting low (0–5 points), moderate (6–8 points), or high (9 or 10 points) methodological quality (Table 1).

View this table:
Table 1

Quality assessment description according to a modified version described by Feldmann and Bondemark (2006).

1. Sample sizeNumber of affected teeth0–10 = 0 point; 11–20 = 1 point; ≥21 = 2 points
2. Study designRandomized controlled trials (RCTs) and prospective clinical studies (PS) or randomized controlled studies in animalsRCT = 3 points; PS = 1 point
3. Selection descriptionDescription of the evaluated teeth and the characterization of mini-implants (diameter and length)Teeth or mini-screws description = 1 point; teeth and mini-screws description = 2 point
4. Diagnostic methodsDiagnostic methods used to evaluate the tooth after traumaRadiographic = 1 point; histological analysis or scanning electron microscopy = 2 points
5. Post-damage follow-up periodPost-trauma evaluation period<3 months = 0 point; ≥3 months = 1 point


Searches of the electronic databases identified 579 titles and abstracts on mini-implants and root damage, which were entered into a PRISMA flow diagram (Figure 1). Among these, 265 titles were duplicated and were therefore removed. All remaining titles and abstracts (314) were analysed and 294 were found inappropriate and were subsequently excluded. The full texts of 17 studies were assessed and 4 (Cheng et al., 2004; Kravitz and Kusnoto, 2007; Yanosky and Holmes, 2008; El-Beialy et al., 2009) studies were excluded because root damage was not evaluated, although the authors had reported complications associated with orthodontic mini-implants in title and/or abstract. Two studies (Asscherickx et al., 2008; Dao et al. 2009) were excluded because they were based on identical samples/showed results similar to another already selected article. Eleven articles (Asscherickx et al., 2005; Maino et al., 2007; Chen et al., 2008; Kadioglu et al., 2008; Brisceno et al., 2009; Hembree et al., 2009; Kang et al., 2009; Renjen et al., 2009; Lee et al., 2010; Rinaldi and Arana-Chavez 2010; Kim and Kim, 2011) finally analysed (Table 2). Of those 11 articles, 9 were based on animal studies (Asscherickx et al., 2005; Chen et al., 2008; Brisceno et al., 2009; Hembree et al., 2009; Kang et al., 2009; Renjen et al., 2009; Lee et al., 2010; Rinaldi and Arana-Chavez, 2010; Kim and Kim, 2011) and 2 were based on human samples (Maino et al., 2007; Kadioglu et al., 2008).

View this table:
Table 2

Quality assessment of the included studies.

ReferencesSample sizeStudy designSelection descriptionDiagnostic methodsFollow-upTotal scoreJudged methodological quality standard
Asscherickx et al. (2005) 112206Moderate
Brisceno et al. (2009) 2322110High
Chen et al. (2008) 212218Moderate
Hembree et al. (2009) 232209High
Kang et al. (2009) 212218Moderate
Kadioglu et al. (2008) 132208Moderate
Kim and Kim (2011) 212207Moderate
Lee et al. (2010) 211206Moderate
Maino et al. (2007) 012205Low
Renjen et al. (2009) 112206Moderate
Rinaldi and Arana-Chavez (2010) 212207Moderate
Figure 1

Flow diagram of studies selected.

No additional article was found in the manual search. A detailed summary of the final selected studies can be found in Table 3.

View this table:
Table 3

Characteristics of included studies.

ReferencesDiagnostic methodsNumber of animals/subjectsNumber of mini-implantsDiameter × length (mm)Proximity between MI and dental rootMI contact with periodontiumMI contact with dental rootRoot perforation with MITrauma durationFollow-upOutcomes
Asscherickx et al. (2005) Histological analysis5 Beagles dogs201.7 × 6.05 teeth06 teeth00.25 weeks0A total failure rate of 55%. Six mini-implants were identified as being (or having been) in contact with a dental root as observed histologically on the serial sections. One of these was still in situ at the end of the examination period. Formation of separative cementum lining the root could be observed. For the five other implants, which had been in contact with a tooth root and were lost, a defect in a tooth root could be observed. All mini-implants placed in contact with a root surface and less than 1.0 mm away from the marginal bone level failed.
Brisceno et al. (2009) Histological analysis7 Beagles dogs561.8 × 8.00049 teeth7 teeth06 and 12 weeksUnder favourable conditions (no infection or pulpal invasion), root healing occurred in 64.3% of the teeth after damage with mini-implants. In the teeth with normal healing, the percentage of cementum in the defect significantly increased between 6 and 12 weeks. Partial or no healing was evident for teeth with pulpal invasion and inflammatory infiltrate.
Chen et al. (2008) Histological analysis6 Mongrel dogs722.0 × 11.027 teeth045 teeth00, 3, 12, and 24 weeks12 and 24 weeksDuring placement of mini-implants in the alveolar process, increased failure rates were noticed among those contacting adjacent roots. Failed mini-screws appeared to be surrounded with a greater volume of soft tissue. When more inflammation was present, the adjacent roots seemed to experience more resorption. Nevertheless, the created lesion was repaired with a narrow zone of mineralized tissue deposited on the root surface, which was likely cellular cementum, and was mainly filled with alveolar bone, with the periodontal ligament space being maintained.
Hembree et al. (2009) Histological analysis7 Beagles dogs421.8 × 8.011 teeth3 teeth22 teeth6 teeth0, 6, and 12 weeks0The placement of mini-implants could produce 1. immediate and extensive damage to periodontal structures, 2. short- and long-term damage of unloaded mini-implants was similar to immediate damage, 3. short- and long-term healing was evident for mini-implants remaining in contact with the tooth root, and 4. inflammation increases the risk of further damage caused by mini-implants.
Kang et al. (2009) Histological analysis3 Beagles dogs481.8 × 8.524 teeth024 teeth01–8 weeks4–7 weeksAlthough the dental root can be injured by mini-implants, minimal clinical side effects are expected if the injury is not too severe because of the healing potential of surrounding tissues.
Kadioglu et al. (2008) Scanning electron microscopy10 Humans201.5 × 8002004 and 8 weeks4 and 8 weeksThe side root resorptions caused by intentional premolar root—mini-implant contact in this study showed repair and healing within a few weeks after removal of the implants or the tipping springs. The injuries were apparently repaired with minimal, if any, clinical consequences.
Kim and Kim (2011) Histological analysis4 Minipigs201.6 × 8.002 specimens24 specimens6 specimens0, 4, 8, 12, and 16 weeks0When the root resorbs away from the mini-implant, cementum healing occurs in most instances after 12 weeks. When the mini-implant was left in contact with the root surface, mostly due to high force and severe trauma to the root during mini-implant placement, no healing occurred. When the conditions were not optimal, resorption and repair did not occur. The damage was irreversible when the mini-implant ruptured through thicker areas of dentin and into pulp tissue.
Lee et al. (2010) Histological analysis4 Beagles dogs1.6 × 6.024 specimens5 specimens8 specimens7 specimens16 weeks0In the near-root and PDL contact groups, the incidence of root resorption increased when the distance between the mini-implant and the root was less than 0.6 mm. In the root perforation group, root resorption and ankylosis occurred on the side opposite the insertion. Some specimens in the PDL contact and root contact groups had cementum growth or little root resorption in spite of the mini-implant's being close to the root.
Maino et al. (2007) Histological analysis2 Humans41.5 × 8004 teeth01 week27 and 30 daysThe results show that contact between a dental root and a drill, implant or both causes resorptive root damage. After discontinuation of the contact, however, repair begins to occur through the deposition of cellular cementum.
Renjen et al. (2009) Histological analysis3 Beagles dogs602.0 × 10.00011 sites5 sites12 weeks0There was no evidence of inflammatory infiltrate or necrosis in the pulp tissue or along the injured root surfaces. Reparative cementum was present along the periphery of each injured root and along displaced dentin fragments in apposition with the PDL. The presence of woven bone intimately related with mini-implant supported the osseointegration of mini-implants.
Rinaldi and Arana-Chavez (2010) Histological analysis24 Wistar rats481.2 × 1.40048 teeth021, 30, 45, 60, 90, and 120 days0A thin cementum-like layer was formed at longer times after implantation at the areas in which the periodontal ligament was in contact with the implant. In addition, bone formation occurred in the alveolar bone in contact with the implant surface, thus showing that osseointegration actually takes place around orthodontic mini-implants when left for long times.
  • MI, mini-implant; PDL, periodontal ligament.

Overall, the analysed data were based on 56 animals (Asscherickx et al., 2005; Chen et al., 2008; Brisceno et al., 2009; Hembree et al., 2009; Kang et al., 2009; Renjen et al., 2009; Lee et al., 2010; Rinaldi and Arana-Chavez, 2010; Kim and Kim, 2011) and 12 patients (Maino et al., 2007; Kadioglu et al., 2008) and 390 orthodontic mini-implants were used. Only one study (Lee et al., 2010) did not identify the number of orthodontic mini-implants used. The authors of those articles were contacted to obtain the required information, but no reply was received. The data for implant length and diameter were given in the 11 studies included and ranged from 1.4 to 11.0 and from 1.2 to 2.0 mm, respectively. The damage caused by mini-implants consisted of root perforation, root contact, and periodontal contact; mini-implants were also inserted near the roots of adjacent teeth.

Two studies were of high methodological quality. Brisceno et al. (2009) and Hembree et al. (2009) used the same method to evaluate the healing potential of the roots and surrounding periodontium after intentional damage in the mandible and maxilla from seven beagles dogs, respectively. Fifty-six mini-implants were used in the mandible (49 contacted the roots and 7 were drilled into the roots) and 42 in the maxilla (11 were inserted near the root, 3 contacted the periodontal ligament, 22 contacted the roots, and 6 were drilled into the roots). The trauma duration and the follow-up periods ranged from 0 to 12 weeks.

Eight studies were of moderate methodological quality classification (Asscherickx et al., 2005; Chen et al., 2008; Kadioglu et al., 2008; Kang et al., 2009; Renjen et al., 2009; Lee et al., 2010; Rinaldi and Arana-Chavez, 2010; Kim and Kim, 2011). Asscherickx et al. (2005) evaluated the immediate and 25 weeks post-insertion effects of 20 mini-implants that were inserted into the mandible of five beagle dogs. Radiographs were taken and vital stains were administered to posterior histological evaluation. The histological analysis demonstrated that five mini-implants were inserted near the root and six contacted the root. In another study, Chen et al. (2008) evaluated root repair by using 72 mini-implants inserted in six mongrel dogs. The immediate 3, 12, and 24 weeks post-insertion effects of 27 mini-implants inserted near the root and 45 mini-implants that contacted the root were assessed by histologically. Follow-up was at 12 and 24 weeks. In a human study, Kadioglu et al. (2008) evaluated the premolar root surfaces (20 roots contacted) of 10 patients after intentional contact with 20 mini-implants by scanning electron microscopy. The periods of trauma duration and follow-up were 4 and 8 weeks, respectively. Kang et al. (2009) assessed root damage in three beagle dogs caused by 48 mini-implants. Histological investigation revealed that 24 mini-implants were inserted near the root and 24 had contacted the root. The trauma duration ranged from 1 to 8 weeks and the follow-up periods ranged from 4 to 7 weeks. In one study conducted with minipigs, Kim and Kim (2011) assessed root damage caused by 20 mini-implants that contacted the periodontal ligament (n = 2), contacted the roots (n = 24), and drilled into the roots (n = 6). Investigation was by histological analysis and the trauma duration ranged from 0 to 16 weeks. Lee et al. (2010) evaluated the root damage caused by mini-implants inserted near the root (n = 24), in the periodontal ligament (n = 5), on the root (n = 8), and that drilled into the root (n = 7). Duration of trauma was 16 weeks and investigation was by histological analysis. The authors did not reveal the number of mini-implants used. Renjen et al. (2009) evaluated the effects on the pulp and supporting tissues when mini-implants severely damaged the root surface. The authors used 60 mini-implants in three beagles dogs. The histological analysis showed 11 sites with root contacted and 5 with more extensive root damage after 12 weeks.

Rinaldi and Arana-Chavez (2010) described the ultrastructure of the interface between periodontal tissues and titanium mini-implants in rat mandibles of 24 Wistar rats. Forty-eight mini-implants were used for histological analysis and six different periods of trauma (21, 30, 45, 60, 90, and 120 days) were analysed.

One study (Maino et al., 2007) was of low methodological quality. This pilot study investigated the effects of contact between a drill, a mini-implant, or both and the roots of four upper premolars in two adolescent orthodontic patients by means of histological analysis. Four mini-implants contacted four teeth. The trauma duration was 1 week and the follow-up evaluation was 27 and 30 days.


This systematic review utilized a reproducible search strategy to analyse the effects and damage caused by the contact or drilling of mini-implants to the dental root. The first clinical report in the literature using mini-implants for orthodontic anchorage appeared in 1983, when Creekmore and Eklund (1983) used them to intrude maxillary incisors and because of this, we limited our search 1980 to date. To ensure that the most valid and reliable studies were obtained, strict inclusion and exclusion criteria were used.

Of the 11 studies selected, 2 articles were of high methodological quality, 8 were moderate, and 1 study was of low quality. Of the 11 articles selected, 9 were based on animal studies and 2 were based on human samples. A possible explanation for the limited number of human studies are the ethical issues involved as well as the difficulty of the experimental set-up for such a study. All data from the above articles were collected and analysed in order to assess the risks and potential for repair after inadvertently contacting a root during the insertion of a mini-implant.

To strengthen the methodological quality, we added the variables ‘diagnostic methods’ and ‘post-damage follow-up period’ to our methodological assessment. These variables are important in intervention studies in animals that evaluated root damage with orthodontic mini-implants (Asscherickx et al., 2005; Chen et al., 2008; Brisceno et al., 2009; Hembree et al., 2009; Kang et al., 2009; Renjen et al., 2009; Lee et al., 2010; Rinaldi and Arana-Chavez, 2010; Kim and Kim, 2011).

In our systematic review, the two studies (Brisceno et al., 2009; Hembree et al., 2009) that obtained high methodological quality were performed by the same authors using two different sets of beagle dogs. A randomized split-mouth design was used to evaluate the healing potential of the roots and surrounding periodontium [cementum, periodontal ligament (PDL), and bone] after intentional damage during mini-implant placement. Brisceno et al. (2009) evaluated the healing 6 and 12 weeks after intentional root damage. Seven skeletally mature male beagle dogs had mini-implants placed into the roots of eight mandibular teeth (six premolars and two first molars). After root contact had been verified by using insertion torques and radiographs, the mini-implants were immediately removed, and the sites were allowed to heal for 6 or 12 weeks. Damage to the roots and periodontium ranged from mild invasion to the cementum to pulp invasion. New bone, new PDL, and new cementum were observed in 64.3% of the teeth, with significant (P < 0.05) increases in the percentages of the cementum over time. Sequential labelling confirmed healing at both 6 and 12 weeks. Abnormal healing was found in 35.7 per cent of the teeth; it included lack of PDL and bone regeneration, bone degeneration in the furcation area, ankylosis, and no healing associated with inflammatory infiltrate or pulpal invasion. The only other high quality investigation by Hembree et al. (2009) used the same method of Brisceno et al. (2009) and the authors evaluated the immediate, short-term (left for 6 weeks), and long-term (12 weeks) damage on the roots of the maxillary second, third, and fourth premolars of seven mature beagle dogs. Histological analysis showed damage of 73.8 per cent of the teeth, ranging from displacement of bone into the periodontal ligament to invasion of the pulp chamber. Displacement of bone into the periodontal ligament and direct damage to the periodontal ligament occurred in three (7.2 per cent) instances. Damage was isolated to the cementum of eight teeth (19.0 per cent), whereas damage occurred in the dentin of 11 teeth (26.2 per cent). Loss of bone in the furcation area was evident in three teeth (7.2 per cent), and severe damage into the pulp occurred in six teeth (14.2 per cent). No differences in the amounts of damage were evident between the immediate, short-, and long-term groups. Healing often occurred in the cementum around the unloaded mini-screw implant. Unloaded mini-implants that remain in contact with roots of the teeth can show several degrees of healing. In cases that involved perforation of the pulp chamber, some cementum and dentine repair occurred in the short term, despite the position of mini-implants.

Among the eight studies of moderate methodological quality (Asscherickx et al., 2005; Chen et al., 2008; Kadioglu et al., 2008; Kang et al., 2009; Renjen et al., 2009; Lee et al., 2010; Rinaldi and Arana-Chavez, 2010; Kim and Kim, 2011), seven were prospective studies based on different animal models with different animals assess the healing responses after contact or proximity of mini-implants to the roots of the adjacent teeth. These studies (Asscherickx et al., 2005; Chen et al., 2008; Kadioglu et al., 2008; Kang et al., 2009; Renjen et al., 2009; Lee et al., 2010; Rinaldi and Arana-Chavez, 2010; Kim and Kim, 2011) evaluated the consequences of mini-implants inserted near root, in the periodontal ligament, that contacted the roots and that drilled into the roots. Asscherickx et al. (2005), Chen et al. (2008), Kang et al. (2009), and Kim and Kim (2011) reported that when the mini-implants were inserted and removed immediately, evidence of continuous cementum repair was seen along the injured root surface. Asscherickx et al. (2005), Kang et al. (2009), Lee et al. (2010), Renjen et al. (2009), and Rinaldi and Arana-Chavez (2010) showed similar results when the mini-implants were left in contact with the root. Only one study (Kim and Kim, 2011) presented no normal healing response when the mini-implant was left touching the root. In relation to root perforation, Renjen et al. (2009) did not find pulp necrosis, external resorption, and ankylosis when mini-implant had penetrated into the pulp space. The authors observed reparative cementum at each injury site. However, Kim and Kim (2011) and Lee et al. (2010) observed that abnormal healing responses were seen when the pulp tissue was ruptured. Moreover, Lee et al. (2010) observed ankylosis and root resorption on the side opposite the mini-implant insertion. There was only one human study (Kadioglu et al., 2008) among these eight studies classified as moderate methodological quality. The authors performed a split-mouth study design and evaluated dental roots contacted by mini-implants. When the mini-implants drilled the roots by accident, with no repair allowed, there was extensive damage to the root surfaces. In the experimental groups, the roots contacted showed reorganization of collagen fibre and new fibres after the repair period. Although some resorption lacunae were still discernible after 8 weeks, the collagen fibres fully covered the affected areas.

Only one study (Maino et al., 2007) was of low methodological quality. This human study revealed that after discontinuation of the contact, repair begins through deposition of cellular cementum. In the site where the root was damaged by the pilot drill, the original contour of the resorption area was evident as well as incomplete repair of the resorption lacunae with cellular cementum. One of the reasons why this study was ranked as showing low methodological quality was its small sample size.

The evidence from this systematic review of the literature revealed that in both studies (animals and humans), roots, which were contacted by a mini-implant, can experience resorption, but after discontinuation of the contact, the roots are repaired showing absence of significant damage.


Based on the evidence of the high and the moderate quality reports found, this systematic review suggests that the quality of the root repair depends on the amount of damage caused by the mini-implant. Under favourable conditions (no inflammatory infiltrate or pulpal invasion), healing and almost complete repair of the periodontal structure can occur for as long as root damage is limited to the cementum or the dentin. Mini-implants that perforated the pulp space can showed varying degrees of healing.


This study was supported by CAPES.


View Abstract