This study aimed to investigate the effects of the luting methods on the amount of cement remnants in implant restorations and to determine the restoration surface with the maximum amount of residual cement. Forty abutments and crowns were divided into 4 groups as follows: TB group, luting with zinc oxide-eugenol cement; TBV group, luting with zinc oxide-eugenol cement after application of a separating agent over the transmucosal area of the abutment; PI group, luting with methacrylate cement; and PIV group, luting with methacrylate cement after application of a separating agent. After cementation, all the quadrants of the specimens were photographed, and the amount and location of the cement remnants were statistically analyzed (P ≤ .05). The amount of cement remnants was significantly smaller in the groups with a separating agent. The type of luting material did not significantly affect the results. Cement remnants were more abundant on the mesial and distal sides than on the buccal and lingual sides of the restoration.
The restoration of totally or partially edentulous jaws with implant-supported restorations is a predictable treatment method that has been widely used for several years.1,2 There are 2 types of implant restorations based on the method of connecting the fixed dental prosthesis to the implant fixture: screw- and cement-retained. The main advantage of screw-retained implant restorations is their retrievability.3
To ensure retrievability in cement-retained implant restorations, conventional cements for temporary setting, such as zinc-oxide eugenol (ZOE), were initially widely used.4 Since then, methacrylate (MA) cements specifically developed for cement-retained implant prostheses have also gained popularity due to their easy manipulation and relatively low solubility.5 Some recent studies have reported that these types of luting cements may have different impacts on the peri-implant tissue.6,7
To overcome the aforementioned limitations of cement-retained implant prostheses, several methods for minimizing excess cement have been proposed. Some studies8–10 have reported on the use of a practice abutment fabricated using various methods and materials. Other studies have suggested covering the transmucosal area of the abutment with polytetrafluoroethylene tape11 or a rubber dam.12 Additionally, as an alternative to these types of barrier membranes, petroleum jelly can be used to coat the transmucosal area of abutments, which may be a simpler and easier approach.
However, few studies have compared the effectiveness of using a conventional luting method with techniques for minimizing cement remnants. In addition, the role of various types of luting materials and the location where the majority of cement remnants are found remain unclear. Hence, the present in vitro study aims to investigate the effects of various luting materials and methods on the amount of cement remnants in implant restorations and determine the restoration surface with the maximum amount of residual cement.
Materials and Methods
The master cast of a patient with a single implant on the maxillary left first molar was used. The peri-implant soft tissue of the cast was simulated using vinyl polysiloxane silicone (GumQuick, Dreve Dentamid GmbH, Unna, Germany). The master cast was scanned using a 3D laboratory scanner (D800, 3Shape A/S, Copenhagen, Denmark). A customized titanium abutment was designed using CAD software (Dental Designer, 3Shape A/S) and fabricated from grade 5 titanium alloy. The margin of the abutment was positioned 1 mm subgingivally on the mesial, distal, buccal, and lingual sides. The subgingival area of the abutment was designed with a concave shape such that it would not compress the gingiva (Figure 1a). A crown for the customized abutment was also designed using CAD software (Figure 1b) and then fabricated from cobalt-chromium alloy using a 3D printing laser-sintering technique with the EOSINT M270 machine (Electro Optical Systems GmbH, Krailling, Germany). To ensure easy retrievability, a screw access hole was made on the occlusal surface and knobs were fabricated on the mesial side of the buccal and lingual surfaces (Figure 1c).
In total, 40 abutments and crowns were fabricated using the same procedure. Of these, 20 were cemented with ZOE cement (Temp-Bond, Kerr Corp, Orange, Calif) and 20 with MA cement (Premier Implant Cement, Premier Dental Products Company, Plymouth Meeting, Pa) (Table 1). In each group, petroleum jelly (Vaseline, Unilever, Englewood Cliffs, NJ) was applied as a separating agent using gauze before luting over the transmucosal area of 10 randomly selected abutments (Figure 2). Thus, a total of 4 groups were formed (n = 10 each): group TB (luting with ZOE cement); group TBV (luting with ZOE cement after the application of a separating agent over the transmucosal area of the abutment); group PI (luting with MA cement); and group PIV (luting with MA cement after the application of a separating agent over the transmucosal area of the abutment).
To lute the crown, the abutment was manually tightened on the master cast with or without petroleum jelly application and the crown was cemented on the abutment according to the instructions provided by each cement manufacturer. The cement was loaded within the crown by uniformly coating the axial wall surface with a thickness of about 1 mm. Then, maximum finger pressure was used as the seating force. Once the cement was completely set, a trained prosthodontist removed excess cement using an explorer in a limited timeframe of 30 s. To avoid differences or errors due to variations among practitioners, a single clinician cemented the crown and removed excess cement from all specimens using a motion similar to that used in daily clinical practice with an explorer. After the cement removal procedure, the abutment screw was unscrewed and the cemented crown-abutment complex was unattached from the master cast.
A rectangular, columnar jig was fabricated using putty impression material (Exafine putty type, GC Dental Products Corp, Tokyo, Japan). After binding the crown-abutment complex to the jig, each of the mesial, distal, buccal, and lingual surfaces of the abutments were photographed using a digital camera (D3300, Nikon, Melville, NY). The position of the camera was set at an angle such that only the jig's front surface was visible. Then photographs were taken perpendicularly at a constant magnification of 3:1 (Figure 3). The occlusal view of the master cast was also photographed vertically at 3:1 magnification and at an angle that enabled clear visualization of the internal hex of the laboratory analog. The digital photos were imported and analyzed using Adobe Photoshop CS5 (Adobe Systems Ltd, San Jose, Calif) on a personal computer. The amount of residual cement was visually measured. On the crown-abutment complex images opened in the software, areas with cement remnants were marked using the “lasso tool.” The number of pixels in the selected area was recorded using the histogram option. Next, the photographs of the master cast without the abutment were evaluated. The peri-implant sulcus area was divided into 4 equal quadrants (mesial, distal, buccal, and lingual) on the basis of the internal hexagon of the laboratory analog, and the number of pixels depicting cement remnants in each quadrant was recorded (Figure 4).13,14 Ultimately, the recorded number of pixels measured from the abutment images and master cast images were summed for the mesial, distal, buccal, and lingual surfaces of the restoration as follows:
Total amount of cement remnants (eg, mesial side) = amount of cement remnants on the crown-abutment complex (eg, mesial side) + amount of cement remnants on the master cast (eg, mesial side)
First, the sum of the residual cements in all the quadrants obtained for each tested specimen was analyzed to compare the amount of cement remnants among the groups. Second, the amount of cement remnants in each quadrant was compared regardless of the cement type and/or technique.
The data were statistically analyzed using SPSS version 22 (SPSS Inc, Chicago, Ill) for Windows (Microsoft Corp, Redmond, Wash). First, differences among the groups based on the luting material and method were analyzed. The data were analyzed using one-way analysis of variance (ANOVA). Tukey honest significant difference (HSD) post hoc tests were used for intergroup comparisons, with a significance level of 0.05. Then, the implant surface was selected as the statistical unit. Kruskal-Wallis and Mann-Whitney tests were used to compare the amount of cement remnants among the four restoration surfaces, with a significance level of 0.05. P values were corrected by the Bonferroni test for multiple comparisons (P = .05/6 = .0083). The methodology, results, and conclusions were reviewed by an independent statistician.
The amount of cement remnants (unit: pixels) in each group is summarized in Table 2. There were statistically significant differences among the 4 experimental groups (ANOVA, F = 12.913; df = 39; P < .0001). The Tukey HSD test revealed a significant difference between groups TB and TBV (post hoc test; P = .002) and between groups PI and PIV (P < .0001).
There was no significant difference between groups TB and PI (P = .545). When the abutments were coated with a separating agent, the difference between the luting materials was still not statistically significant (groups TBV and PIV; P = .869). The cleanest surfaces with the least residual cement were observed in group TBV, followed by groups PIV, TB, and PI (Figure 5).
The amount of residual cement was compared among the mesial, distal, lingual, and buccal surfaces of the restorations using the Kruskal-Wallis test (P < .0001). According to the Mann-Whitney and post hoc tests, the amount of cement remnants was significantly greater on the mesial and distal sides than on the buccal and lingual sides (mesial and buccal: P < .0001; mesial and lingual: P < .0001; distal and buccal: P < .0001; distal and lingual: P = .004) (Figure 6).
Cement remnants in the peri-implant sulcus can result in the accumulation of plaque and oral pathogens,15–17 which can lead to peri-implantitis.13,18 According to Wilson,19 cement remnants were found in 81% of patients with clinical or radiographic symptoms of peri-implantitis such as bleeding on probing or pus discharge.
The correlation between the amount of residual cement and the dental prosthesis has been extensively studied. Some studies13,14,20 evaluated the relationship between the position of the restoration margin and amount of residual cement and reported that undetected cement remnants were more likely to be found when the restorations had deeper subgingival margins. In addition, the cervical contour and undercut of the crown have been reported to affect accessibility for excess cement removal.13 Therefore, it appears to be beneficial to design the cervical shape of the crown rather than excessively overcontouring it.
Other studies have evaluated the amount of residual cement in association with the type of luting materials.15,17,21 In a retrospective clinical study, Korsch et al22 reported that residual cement was more likely to be observed when MA cement was used, resulting in complications. In an in vitro study, Behr et al21 reported that no residual cement remained when ZOE cement was used.
In the present study, unlike the findings in previous studies,21,22 a considerable amount of excess cement remained after luting with both the ZOE and MA cements. In the study by Behr el al,21 scaling was performed to remove cement attached to the exposed surface, whereas in the present study, excess cement extruding from the margin, which was situated 1 mm into the simulated peri-implant soft tissue, was removed on the master cast. Therefore, complete removal of the ZOE cement was difficult, even when petroleum jelly was applied. Since the durability of ZOE cement is lower than that of MA cement,23 residual cement was not removed as a lump; only the cement in accessible supragingival regions could be broken, while that in the gingival sulcus remained intact. In contrast, MA cement remnants in the gingival sulcus were often eliminated as lumps during the removal of remnants from the supragingival regions.
Even if a similar amount of excess cement remains with the MA and ZOE cements, the former is less soluble than the latter and, consequently, will remain for a longer time in the oral cavity.23,24 ZOE cement contains eugenol, which has an antimicrobial effect.25 On the other hand, several bacterial species have been reported to adhere more strongly to MA cement.7,16 As a result, more complications may occur with the use of MA cement.
In the present study, the application of a separating agent resulted in a significant decrease in the amount of residual cement irrespective of the type of luting material. Thus, application of petroleum jelly to the transmucosal part of the abutment appears to be a useful technique for minimizing residual cement and decreasing complications. However, the separating agent should be carefully applied because it can seep into the coronal portion of the abutment and weaken crown retention.
We also found that residual cement was more abundant on the mesial and distal sides of the restoration than on the buccal and lingual sides. Owing to the presence of adjacent teeth on the proximal surfaces of the evaluated implants, the approach for residual cement removal was challenging. Moreover, only explorers were used for cement removal, which was difficult to achieve from the proximal surfaces without the use of dental floss. Therefore, in clinical practice, careful attention is necessary during the removal of excess cement from the proximal surfaces of a restoration, and additional instruments, such as floss, should be used.
On the basis of our findings, we suggest the following methods to minimize cement remnants in implant restorations, other than the application of a separating agent. At the time of designing a customized abutment, esthetically sensitive buccal and lingual margins, from which residual cement can be easily removed, could be placed equigingivally or subgingivally for esthetic reasons, whereas esthetically less sensitive proximal margins, from which residual cement is more difficult to remove, should be placed supragingivally.
One strength of this study is the close similarity of the experimental conditions to real-world clinical situations, which was not the case in other in vitro experiments.18,21 The experiments were conducted using the master cast of a patient; furthermore, the posture of the practitioner and the position of the cast were similar to those adopted in actual clinical practice.
However, the in vitro design was a limitation of this investigation. The physical properties of human peri-implant tissue are different from the polyvinylsiloxane-made gingiva of the master cast. Therefore, to apply these results to clinical practice, additional clinical studies that evaluate the amount of cement remnants on real peri-implant tissue are necessary for further validation. Studies investigating the effects of the separating agent on the soft tissue seal around the abutment are also necessary.
Within the limitations of this in vitro study, the following conclusions were drawn. Application of a separating material over the transmucosal part of the abutment prior to cementation of the implant prosthesis can significantly minimize residual cement. The luting materials used in this study may not necessarily influence the amount of cement remnants, and cement is more likely to remain on the proximal surfaces than on the buccal and lingual surfaces of implant restorations when adjacent teeth are present.
These findings suggest that, after the luting of cement-retained implant prostheses, excess cement should be thoroughly and carefully removed from the mesial and distal sides of the restoration to prevent complications caused by residual cement. Furthermore, clinicians should consider carefully designing the vertical position of the abutment margin with the aim of minimizing cement remnants.
The authors have no conflict of interest. The authors self-funded this study, and did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
The authors thank Prof Chan-Uk Oh, PhD, who works for the Institute of Statistics at Korea University, for the review of the statistical analysis.