Retrospective study of combined splinting restorations in the aesthetic zone of periodontal patients

Evaluates the clinical performance of of a fiber-reinforced, composite-resin bonded splint; a bridge; a fiber post; and/or a resin veneer for the restoration of periodontally-involved anterior teeth. Shows that a combination of approaches is a good choice for minimally invasive prosthodontic treatment in the aesthetic zone in periodontal patients.


INTRODUCTION
The loss of a single anterior tooth may be a catastrophic event for a patient. 1 In advanced periodontal disease such a loss is unfortunately often accompanied by adjacent tooth mobility due to severe bone loss. Furthermore, periodontal disease also chronically dislodges teeth due to the masticatory movement of the mobile teeth, which seriously affects anterior aesthetics, especially for the upper anterior teeth. Therefore, prosthodontic rehabilitation is often necessary to restore aesthetics and function in periodontally compromised dentition. 2,3 Prosthodontic treatment for periodontally-involved anterior teeth usually includes dental implants, partial fixed dental prostheses (PFDPs), resin-bonded partial fixed dental prostheses (RBPFDPs), or partial removable dental prostheses (PRDPs). 2,4 Implant restoration is the first choice for the treatment of suitable systemic and local conditions. However, as Donos et al. reported, the patient and site-risk profile should be assessed in combination with Aims This retrospective study evaluated the clinical performance of combining a fiber-reinforced, composite-resin (FRC) bonded splint; bridge; fiber post; and/or resin veneer for the restoration of periodontally-involved anterior teeth. Methods Sixty-three patients with periodontally-involved anterior teeth each received one of the following splinting restorations: FRC-bonded splint-bridges (SB), a combination of an FRC-bonded splint and resin veneer (SV), or FRC-bonded splints (S). Six patients with healthy periodontal teeth received orthodontic retainers with FRC-bonded splints (OS) as a control. Modified USPHS criteria were used to evaluate the effects of rehabilitation, and the periodontal pocket depth (PPD) and clinical attachment level (CAL) of the natural teeth were measured. Results All restorations resulted in good aesthetic outcomes. SV, S in combination with fiber posts and OS all resulted in 100% acceptable ratings for each category that was directly evaluated at each follow-up. The mean PPD was 3.5 ± 1.0 mm at baseline, decreased to 3.3 ± 1.0 mm after one year (p <0.05), and subsequently increased to 3.5±1.0 mm again after four years (p >0.05). These changes were similar to those in the mean CAL. Conclusion A combination of a FRC-bonded splint, bridge, fiber post and/or resin veneer for minimally invasive prosthodontic treatment in the aesthetic zone is a good choice for periodontal patients.
chronic dislodgement is often a dilemma when the extraction of dislodged teeth seriously affects anterior aesthetics. This challenge is especially common for severe labial or lingual dislodgement because the effect of rehabilitation is uncertain, irrespective of the extraction of dislodged teeth. Recently, adequate orthodontic and periodontal treatment has been sought to improve the periodontal condition and anterior aesthetics. [18][19][20] However in such cases, periodontal breakdown causes pathologic migration of teeth, making the orthodontic treatment more complicated. [21][22][23] At present, few clinical studies have examined the aesthetic restoration of periodontally-involved anterior teeth with mobility and chronic dislodgement. The development of silanised and resin-impregnated FRC materials has provided potential new approaches for stabilising hypermobile teeth or replacing teeth in a conservative manner. 24,25 In this study, a combination of an FRC-bonded splint, resin veneer and/or fiber post was used to restore anterior teeth with mobility and chronic dislodgement without tooth extraction.
Therefore, this study systematically and retrospectively investigated anterior teeth restoration cases of periodontally treated patients restored with a combination of an FRC-bonded splint, bridge, fiber post and/or resin veneer to summarise the clinical operation and evaluated the clinical performance. Patients with periodontally-involved anterior teeth and anterior teeth mobility who had been diagnosed with chronic periodontitis according to the 1999 APP (American Academy of Periodontology) Classification 26,27 were admitted to the study unless they met the following exclusion criteria: (1) they were younger than 18 years old; (2) they were unable to read and sign the informed consent document; (3) they presented with concurrent psychological disorders; (4) they suffered from general health-compromising conditions; (5) they were pregnant; (6) they were smokers; (7) they had history of restoration or whitening treatments for the teeth to be evaluated; (8) they could not or were unwilling to return for follow-up; (9) they had lost two or more teeth in the anterior segment of one jaw; or (10) one canine tooth was missing or both canines in the anterior segment on one jaw were mobile. According to the 1999 International Workshop Classification, 27,28 63 patients were diagnosed with severe chronic periodontitis. These patients had to undergo initial periodontal therapy, including oral health instructions, scaling, root planning, and local drug treatment. Scaling and root planning was implemented at intervals of four to six weeks until the patient's oral hygiene was stable, which was indicated by the absence of bleeding on probing. 29,30 The patients were educated about oral hygiene protocols and recalled for periodical follow-up controls annually after splinting. All teeth with FRC bonding that presented with a mobility of MII or more [mobility -classified according to Miller's classification] 31 after initial periodontal therapy or severe facial-lingual dislodgement received root canal treatments and fiber-post restorations.

Patient population
In addition, six patients (three females and three males, between 18 and 28 years old) who were free of periodontal disease received orthodontic retainers with FRCbonded splints to compare the effects of clinical rehabilitation for splinting restorations between chronic periodontitis and a healthy periodontium. In addition to being free of periodontal disease, none of these patients were missing teeth in the nine units of the anterior segment of one jaw. The patients abided by the same exclusion criteria as the periodontally-involved patients. Each unit of restoration was applied from canine to canine at the anterior segment of one jaw.   Operating procedure

Groups of restorations
Each unit of restoration was applied from canine to canine in the anterior segment of one jaw. As shown in Table 1

Tooth preparation
The teeth for splinting were cleaned with a pumice using a prophylactic brush. When the restoration was planned on the maxilla, the lingual surfaces needed to be ground to allow for a 0.5 mm interspace. Teeth that had undergone root canals were treated with fiber-post restorations. For SV, the configurations of the coronals of the teeth with severe facial-lingual dislodgement were modified to restore a normal arch form with fiber posts and dualcure resin core materials (Paro Core, coltène/ whaledent, Swiss). For SB, extracted teeth with rimmed roots and pulp spaces were obturated with resin core or commercial acrylic resin (Heraeus, German) with grooves (0.5 × 3 mm) on the lingual surface to inlay the splint and process the area for a pontic. The pontic was prepared for a modified ridge lap-type with a slight concavity that conformed to a residual ridge 32 and was temporarily fixed in the suitable position by a thin layer of flowable composite resin (Filtek™ Z350 XT Flowable Restorative, 3M ESPE) which was photo-polymerided for 20 seconds without adhesive. A silicone rubber template was then prepared by placing silicone rubber on the lingual and incisal surfaces of all bonded teeth and covering the labial surface of the pontic. For the SV, S and OS treatments, the silicone rubber template was placed on the lingual and incisal surfaces of all bonded teeth.

Bonding procedure
All splints were applied from canine to canine and prepared under a rubber dam. The enamel surfaces on the palatal or lingual side were subsequently etched with 35% orthophosphoric acid (Gluma Etch, Heraeus, Genman) for 60 seconds. After rinsing with water and air drying, the corresponding intermediate adhesive (One-Step Plus filled universal dental adhesive, Bisco Inc., USA) was applied to the surfaces using a microbrush and photo-polymerised for 10 seconds. With the aid of corresponding silicone rubber template, a previously measured length of the single-layer quartz splint and the flowable resin (Quartz splint resin, RTD) were then placed on the lingual surface of the tooth to continuously align the FRC. For SB, the extracted tooth or commercial acrylic resin tooth for the pontic was placed in the silicone rubber template to be adhered to quartz splint and neighbouring teeth in the correct position. After a preliminarily photopolymeridation for 20 seconds, the silicone rubber template was removed to allow the entire FRC to fully polymerise from the lingual aspect. The exposed surfaces of the FRC were then covered with another thick flowable resin composite (Filtek™ Z350 XT Flowable Restorative, 3M ESPE) and photopolymerised again for 40 seconds from all aspects. Direct resin veneers were applied for SV when the facial view needed to be improved. Finally, after the removal of the excess resin composite and occlusal adjustments, the composite surfaces were polished with coarse, medium, fine, and ultrafine finishing disks (Sof-Lex™, 3M ESPE) in sequence in order to not expose the fibers. The restorations were cleaned and polished at each recall and repaired in case of fracture or the debonding of the splints or resin.

Clinical evaluation
The patient satisfaction was assessed only at baseline. However, the clinical rehabilitation and periodontal condition were first assessed at baseline (immediately after splinting procedures) and thereafter annually during the follow-up visits according to the written protocol. The patients were informed about possible complications and instructed to call upon experience of a failure that could occur before the actual annual follow-up appointment.

Patient satisfaction survey
Sixty-nine patients were interviewed regarding their satisfaction with their restorations (regarding general satisfaction) using a visual analogue scale (VAS) of 100 mm with the endpoints extremely dissatisfied (0) and extremely satisfied (100). 33

Effects of clinical rehabilitation
The modified USPHS criteria 34 (Table 2) were used to evaluate retention, marginal integrity, marginal discolouration, anatomic form and secondary caries in all groups (SB, SV, S and OS) at baseline, 1, 2, 3 and 4 years No visible gap in which the explorer will penetrate There is visible gap, the explorer will penetrate or catch The explorer penetrates the gap and dentin or base is exposed The restoration is mobile, partially or totally fractured or lost Marginal discolouration No discolouration Discolouration is present but has not penetrated along the margin Discolouration has penetrated along the margin Anatomic form Restoration is continuous with existing anatomic form Restoration is discontinuous with existing anatomic form, but dentin or base is not exposed Sufficient material is lost to expose dentin or base Secondary caries No caries is present at the margin of the restoration There is evidence of caries at the margin of the restoration after splinting. The mobility of the natural teeth in the SB, SV and S groups was evaluated according to Miller's classification 31 at baseline, 1, 2, 3 and 4 years after splinting.

Periodontal evaluation
The periodontal pocket depth (PPD) and clinical attachment level (CAL) were measured at six sites (mesiobuccal, buccal, distobuccal, distolingual, lingual and mesiolingual sites) on each natural tooth of each restoration unit in the SB, SV and S groups at baseline, 1, 2, 3 and 4 years after splinting. The SB group included 30 sites per restoration, while the SV and S groups included 36 sites per unit. Williams periodontal probes with 1, 2, 3, 5, 7, 8, 9 and 10 mm markings 35 were used.

Statistical analysis
A statistical analysis was performed using the SPSS statistical package 13.0 (IBM Corporation, Armonk, NY, USA). A restoration unit comprised each study unit. The effects of clinical rehabilitation were analysed with a chi-square test. The PPD and CAL data are expressed as means ± standard deviation (SD). The PPD and CAL were analysed with a factorial-designed analysis of variance (ANOVA). The differences were considered to be significant when the p-value did not exceed 0.05.

Periodontal evaluation
The total mean PPD for the splinted natural teeth was 3.4 ± 1.0 mm, and the total mean CAL was 5.9 ± 1.2 mm. The mean PPD and CAL of the different groups are shown in Figure 5a, and their changes over time within the same group are shown in Figure 5b. The mean PPD was 3.5 ± 1.0 mm at baseline and then decreased to 3.3 ± 1.0 mm at the one-year recall (p <0.05). It then increased to 3.5 ± 1.0 mm again at the four-year recall (p >0.05). This change was similar to that of the mean CAL. Neither the PPD nor the CAL differed between baseline and the two-year, three-year recall or four-year recall (p >0.05).

DISCUSSION
Aesthetic restoration for periodontallyinvolved anterior teeth is a challenge. In this retrospective study, we applied a combination of an FRC-bonded splint, bridge, fiber post and/or resin veneer in prosthodontic treatment for periodontally-involved tooth mobility, chronic anterior tooth dislodgement or anterior tooth loss. Under the limits of this retrospective evaluation, this approach resulted in acceptable outcomes for all groups and minimally invasive approaches for the restoration in the aesthetic zone of periodontal patients.
In our study, the periodontally-involved anterior teeth presenting with hypermobility (MII or above) after initial periodontal therapy received root canal treatments to prevent combined periodontic-endodontic lesions after restoration and sensitivity caused by gingival recession. The periodontic lesions of these teeth were usually in the apical third of the root, and most lesions were close to the root apical region. This approach constituted an empirical clinical therapy that blocked the possibilities of periodontic-endodontic infection. Traditionally, periodontally involved teeth do not usually need root canal treatments except when associated with endodontic lesions because keeping the pulp as vital as possible was a basic principle for dental treatments. Many studies showed endodontically treated teeth without mobility were generally weaker due to loss of tooth structure and removal of the pulp, and consequentially fractures often occurred in pulpless teeth. [36][37][38] However, we had not reviewed the related literature comparing the tooth longevity of periodontally-involved natural teeth with and without root canal treatments. In our opinion, tooth fracture would not have occurred among periodontally involved teeth presenting with hypermobility, so the tooth longevity would not be affected by whether or not root canal treatments were received. Based on this opinion, another aim of root canal treatments in this study was to provide convenience for the aesthetic restoration. The teeth with severe facial-lingual dislodgement received root canal treatments because they required significant tooth reduction. Posts are often required to restore endodontically treated teeth in order to retain and provide resistance for the core material, to provide corono-radicular stabilisation and to facilitate the transmission of the dental loads from the coronal structure to the root. 39 In our study, glass fiber posts were applied because of their advantageous features, including their aesthetic effects and elastic behaviour similar to that of dentine, which permitted a uniform stress distribution within the root. 40,41 The fiber posts connected the tooth dentin and the FRC-bonded splint and/or resin veneers to retain and provide resistance for the splinting restorations. The splinting method was convenient and distinctive when employed with the aid of a self-made silicone rubber template of putty or heavy body, which has been rarely reported. The silicone rubber for the template, which might be transparent or non-transparent, was placed on the lingual and incisal surfaces of all bonded teeth and covered the labial surface of the pontic for impression in the SB group, while it was only placed on the lingual and incisal surfaces of all bonded teeth in the SV, S and OS groups. The splint was preliminarily photo-polymerised from the labial and proximal aspects of interdental spaces if the silicone rubber template was nontransparent. The silicone rubber template not only laminated the bonded lingual surfaces of the splinting teeth with the splint, but also ensured that the pontic adhered to the splint and to neighbouring teeth in the correct position. These advantages simplified the FRCbonded splint-bridges or splint procedures.
Although the patient satisfaction with all combined splinting restorations was generally high, which could be partly because of this being single visit treatment, the clinical rehabilitation outcomes measured in our study were all satisfactory. All restorations resulted in good aesthetic outcomes. SV1, SV2, S1 and OS all showed 100% acceptable ratings in each category that was directly evaluated at each recall; this indicated that periodontally-involved anterior teeth were successfully rehabilitated for up to four years using an FRC-bonded splint and resin veneer combination or an FRC-bonded splint and fiber post combination. The FRC-bonded splints also successfully performed as orthodontic retainers in the healthy periodontal negative controls. This success was closely related to the good flexure and rigidity of the quartz fiber splints. In our study, all restorations were made of single-layer woven quartz splints, whose flexural strength exceeds 120 MPa. Moreover, a study of the rigidity of quartz-fiber splints by Berthold et al. showed that reinforcement materials significantly influenced splint rigidity. Woven quartz splints are rigid quartz fiber splints that can be used to treat horizontal root fractures and alveolar process fractures. 42 However, the rigidity of single-layer woven quarts splints could not satisfy all of our needs for splintbased restoration, especially for FRC-bonded splint-bridges. In our study, SB2 received the lowest value of acceptability ratings. Thus, we will consider the use of two or more layers of woven quartz splints in future studies to increase the rigidity if needed. Here, the thickness of a single-layer never exceeded 0.3 mm, while quartz splints could be placed on both the labial and lingual sides in combination with resin veneers for smaller interspaces.
In this study, FRC-bonded splints without fiber posts for periodontally involved teeth received 100% acceptable ratings for retention and anatomic form but 76.9% ratings for marginal integrity and marginal discolouration. When used as orthodontic retainers for healthy periodontal controls, FRC-bonded splints received 100% acceptable ratings for each category. This difference indicated that tooth mobility significantly negatively impacts the clinical success rates of splinting restorations. Our evaluation of clinical rehabilitation showed that these unacceptable ratings all appeared in the maxilla, which indicated that the FRC materials on the lingual surfaces of periodontally-involved maxillary teeth were easily affected by occlusion, that is, easily wore off, debonded or fractured after improper chewing.
The acceptable rating value of splinting restorations with fiber posts exceeded that of the splinting restorations without fiber posts, which indicated that combining the restoration with fiber posts might help to improve the clinical success of the splinting restorations in the aesthetic zone by providing retention and resistance for the splinting restorations. For the combination of an FRC-bonded splint with a resin veneer, the acceptable ratings were 100% for all outcomes, irrespective of the presence of a fiber post. We attributed this success to the increase in the area of resin adhesion by the resin veneer, which strengthened the entire prosthesis and may have improved clinical success rates of the splinting restorations.
In a pilot study, Kumbuloglupilot et al. evaluated the performance of FRC splints for the treatment of periodontally-involved mandibular anterior teeth and concluded that direct tooth splinting with E-glass FRC material performed successfully for up to 4.5 years; the periodontal status of the splinted teeth showed decreased PPD and CAL values. 25 In contrast, the periodontal status of the splinted teeth in our study showed decreased PPD and CAL values only at the one-year recall. Others have reported that splinting could increase patient comfort during chewing when used to connect periodontally compromised teeth; however, these splints also hinder oral hygiene procedures. 43 PPD and CAL are important indicators of oral hygiene. In our study, decreased PPD and CAL values were observed only at the one-year recall; these values did not differ between the baseline and the two, three-or four-year recall. We believe that these parameters were controlled by oral hygiene. In general, the patients paid great attention to maintaining their oral hygiene and protecting the restorations within the first year of splinting, which was supported by the results of our clinical and periodontal evaluations. Over time, the patients required reminders to maintain enhanced oral hygiene measures, regularly perform periodontal maintenance, and prevent improper chewing. Quartz FRC-bonded splint-bridges or splints did not show an increased risk of PPD and CAL when the periondontally involved teeth were annually evaluated at follow-up.
In general, combinations of an FRC-bonded splint, bridge, fiber post and/or resin veneers for prosthodontic treatment in the aesthetic zone of periodontal patients resulted in good clinical rehabilitation as well as good periodontal status without an increased risk of PPD and CAL. However, the key to maintaining these clinical effects up to four years following treatment was high patient attention to oral hygiene and the protection of the restorations. Combining a fiber post and resin veneer may help improve the clinical success rates of splinting restorations in the aesthetic zone when using self-made silicon rubber templates, which simplify splinting procedures. In future studies, we will consider the use of two or more layers of woven quartz splints to increase the rigidity of splinted restorations if needed. For smaller interspaces, the quartz splints may also be placed on both the labial and lingual sides and combined with resin veneers. In addition to the treatment of periodontal patients, a combination of an FRC-bonded splint, bridge, fiber post and/or resin veneer for prosthodontic treatment may also be used as a prosthetic after orthodontic correction, to treat anterior teeth trauma, etc.
In conclusion, combining an FRC-bonded splint, bridge, fiber post and/or resin veneer for minimally invasive prosthodontic treatment in the aesthetic zone is a good choice for periodontal patients.
Author contributions Authors X. Liu, Y. Zhang, Z. Zhou contributed equally to this work.