Ten‐Year Survival of Single Implants With Veneered Porcelain on Zirconia or Titanium Abutments: A Retrospective Analysis

ABSTRACT

Objective

This study aimed to compare the technical and biological complications of single implants with veneered porcelain on either titanium or zirconia abutments after 10 years in function.

Materials and Methods

All single‐implant surgeries performed at a specialist prosthodontic clinic between 2011 and 2013 were reviewed. After applying the inclusion criteria, 174 implants in 132 patients were included for further analysis. Marginal bone loss relative to baseline and technical complications, including porcelain fractures, were recorded from patient data.

Results

A total of 65 patients with 87 implants were followed until the 10‐year evaluation, including 29 implants connected to a veneered zirconia abutment and 58 to a veneered titanium abutment. One implant with a zirconia abutment was lost due to peri‐implantitis after 7 years, while all other implants survived. Marginal bone loss did not differ significantly between the two abutment groups. At 10 years, implants with titanium abutments showed a mean marginal bone loss of 0.31 mm, whereas implants with zirconia abutments had a mean marginal bone loss of 0.28 mm. Peri‐implant mucositis was observed in 34% of implants with zirconia abutments, compared to 16% of those with titanium abutments. Technical complications were noted in 10.3% of the implants at the 10‐year follow‐up. The most common complication was the chipping of the veneering porcelain, followed by abutment screw loosening.

Conclusion

Implant survival did not differ significantly between the two veneered abutment materials. Both abutment types exhibited low marginal bone loss at the 10‐year follow‐up. However, peri‐implant mucositis was more frequently observed around zirconia abutments. Technical complications were moderate, primarily presenting as chipping of the veneering porcelain. The success rate for implants with zirconia abutments was 93%, compared to 82% for those with titanium abutments.

1. Introduction

The dental implant restoration is composed of several components enabling an aesthetically pleasing and functional replacement of lost teeth. The abutment supporting the crown material has traditionally been made of titanium. In recent years, ceramic abutments made of zirconia have allowed for metal‐free abutments; although recently, a metal base has been introduced to support zirconia crowns. Five‐year follow‐up studies on implants with ceramic abutments have reported promising implant survival rates of over 90% of the included single implants [1, 2, 3]. A small sample study on 27 patients receiving all‐ceramic crowns on zirconia abutments found high survival rates after an 11‐year period [4]. As for long‐term studies on implant survival in general, the literature is scarce. A limited number of studies were presented in a systematic review by Hjalmarsson et al. in 2016 and presented a high survival rate of 95% on the implant level [5]. Kowar et al. followed 97 patients with 129 implants in a Swedish cohort for up to 15 years and found similar high implant survival rates above 95% [6]. However, there is a lack of long‐term evaluation reports on implants with zirconia‐based supraconstructions [3] and only limited knowledge of implant survival more generally.

Dental implant complications can be divided into technical or biological failures. Technical complications, such as porcelain fractures and retention loss, are quite common. In a recent article, the authors found technical complications to occur for 24.6% of the patients followed during a mean time of 5.3 years [7]. Kowar et al. found technical complications on 15% of the implants, with crown detachment being the major complication closely followed by veneer fractures [6]. A systematic review by Pjetursson et al. in 2018 found similar technical complication rates for metal and ceramic abutments (around 11%); however, abutment fractures were more common among ceramic abutments [8]. Furthermore, Pjetursson et al. reported a higher frequency of biological complications for single implants with ceramic abutments than for implants with a metal abutment [8]. Similar findings were presented by Hosseini et al., who in a recent five‐year randomized controlled trial reported a biological success rate of only 77% for zirconia abutments compared to 93% for titanium abutments [9]. One recent area of concern in dental literature is the release of wear particles from the implant‐abutment junction [10]. Zirconia abutments have been proven to cause more wear on the titanium implant head compared to titanium abutments in experimental tests [11, 12]. Concerns regarding the metal wear particles and possible distribution into mucosa and bone tissues around dental implants have been raised, positing the risk of bone loss as a result of wear particle‐induced inflammation [10], like that seen around medical implants [13].

The objective of this study was to compare the biological and technical complications associated with single implants featuring veneered porcelain on zirconia versus titanium abutments after up to 10 years of function. The hypothesis posited that there would be no significant difference between the two abutment materials.

2. Material and Methods

2.1. Patients and Data

In total, 132 patients with 174 implants were included in this retrospective study. Patient description, inclusion criteria, surgical protocol, and prosthetic procedure have been described in detail in a previous article [14]. In brief, patient records of patients receiving single implant surgeries during the years 2011–2013 were opened and read. Inclusion criteria were external platform single implants with screw‐retained crowns made of a veneered porcelain on a zirconia or titanium abutment. Patients were excluded if severely ill or had a bone degenerative illness compromising the implant position. The patients received single implant surgery during the years of 2011–2013 (Nobel Brånemark Implant, Brånemark System, Nobel Biocare, Zurich, Switzerland) and received implant‐supported crowns with either a screw‐retained titanium or zirconia abutment with porcelain veneer. The abutments were primarily manufactured by Procera (Nobel Biocare AB, Zürich‐Flughafen, Switzerland) (113/65%), Biomain I‐Butment angled Zr abutment (Biomain, Helsingborg, Sweden) (8/4, 5%), and one Ti abutment by Cara (Heraeus Kulzer, Hanau, Germany) (1/0, 5%). All patients were treated and followed up at one specialist clinic in Gothenburg, Sweden (the Brånemarkclinic, Folktandvården, Västra Götalandregionen) by several surgeons and prosthodontists with a variation in experience level. Further details on patient description, inclusion criteria, surgical protocol, and prosthetic procedure are found in a previous article [14].

Patient data records were analyzed from the baseline appointment through to the 10‐year follow‐up. All records were reviewed up until May 2024. Patient demographic information, including age at surgery, gender, and number of implants, was recorded. Implant‐specific details such as implant length and diameter, surgery protocol (one‐ or two‐step approach), implant placement, bone augmentation, and the reason for tooth loss (if noted in the data records) were also documented. Technical complications occurring over the years, including fractures, chipping, and loosening of the abutment screw, were noted. Biological complications, as documented in the patient records, including implant loss, peri‐implantitis, or peri‐implant mucositis identified at each follow‐up, were also included in this retrospective study. Radiographic images from the baseline and 10‐year follow‐up were downloaded from the Romexis radiographic records (Planmeca, Helsinki, Finland) and analyzed for marginal bone loss using calibrated millimeter scales in the ImageJ image analysis tool (Fuji, ImageJ Wayne Rasband, NIH, USA). To minimize potential bias, all images were coded and measured without prior knowledge of the implant abutment material used.

2.2. Statistical Methods

Descriptive statistics, including mean values, standard deviations, and range, were calculated using Stata (STATAcorp, USA).

Marginal bone loss values were analyzed using a multilevel mixed‐effects linear regression. Binary values such as the presence of peri‐implant mucositis were analyzed using a logistic regression in Stata (STATAcorp, USA). Missing information was omitted in the statistical method.

Cumulative survival rates (CSR) and success rates and 95% confidence interval (CI) were calculated using a Kaplan–Meier curve with the log‐rank test in Stata (STATAcorp, USA) on the patients first implant. Success was defined as no technical complications registered at the clinical appointment, and survival was defined as no implant loss.

The statistical significance level was set below a p‐value of 0.05.

2.3. Ethical Approval

The study was approved by the regional ethics Committee of Gothenburg, Sweden (2019‐00830/1205‐18).

This observational cohort study followed the STROBE guidelines [15].

3. Results

In total, 63 patients and 87 implants were followed up until the 10‐year control. A total of 56 patients attended both the 5‐year and 10‐year appointments. Patient data are summarized in Table 1. Of the implants, 28 were connected to zirconia abutments and 59 to titanium abutments. For more details on the implant data, see Table 1.

TABLE 1.

Description of patients and implants at 10‐year control.

		Titanium abutments	Zirconia abutments	Total
Patients	n	42	21	63
Age at surgery	Mean age	50 years	41 years	47 years
Gender	Male	21	11	32
	Female	21	10	31
No. of implants	1	31	13	44
	2	8	7	15
	3	2	1	3
	4	1	0	1
Implants	n	59	28	87
Implant position	Maxilla	21	25	46
	Incisors and canines	8	23	31
	Premolars	11	2	13
	Molars	2	0	2
	Mandible	38	3	41
	Incisors and canines	2	0	2
	Premolars	18	3	21
	Molars	15	0	15
Reason tooth loss	Aplasia	19	14	33
	Trauma	4	12	16
	Fracture	9	0	9
	Caries	1	0	1
	Periapical lesion	2	0	2
	Other	2	0	2
	Unknown	22	2	24
Augmentation	Yes	4	13	17

Only one implant with a zirconia abutment was removed after 7 years in function due to severe peri‐implantitis and was therefore not included in the 10‐year follow‐up. The other reasons for exclusion from the 10‐year follow‐up are listed in Table 2.

TABLE 2.

Reason not included.

Moved	Declined	Unknown	Other a
6	28	23	10

^a Too many cancellations (1), examination planned at a later date (9).

The cumulative survival rates (CSR) at 10 years for implants with veneered titanium abutments were 100%, and for implants with veneered zirconia abutments, 97.9% (p‐value: 0.1859) (Graph 1). Success rates for implant crowns on titanium abutments were 82%, compared to 93% for implant crowns on zirconia abutments (p‐value: 0.1401) (Graph 2).

GRAPH 1.

Kaplan–Meier analysis rates on implant survival (implant loss). Implant survival rates for implants with a titanium or zirconia abutment. Survival is defined as no implant loss.

GRAPH 2.

Kaplan–Meier analysis rates on implant crown success (technical complications). Implant crown success rates for implants with a titanium or zirconia abutment. Success is defined as no technical complications registered at clinical appointments.

The mean follow‐up time was 9.98 years, ranging from 7.20 to 11.58 years among the patients. The differences in appointment time with respect to marginal bone loss were statistically analyzed using linear regression and found to be not statistically significant (p‐value: 0.810, 95% CI: [−0.0004717, −0.0003687]).

3.1. Biological Complications

At the 10‐year follow‐up, 10 implants with zirconia abutments (34% of the total number of zirconia abutments) and nine implants with titanium abutments (16% of the total number of titanium abutments) were diagnosed with peri‐implant mucositis. The difference between the abutment groups was not statistically significant (p‐value: 0.072, 95% CI: [0.903, 8.27]), with an odds ratio of 2.73 for zirconia abutments. However, when controlling for placement in the upper jaw, a statistically significant difference was observed (p‐value: 0.045, 95% CI: [1.032, 18.736]), with an odds ratio of 4.39 for zirconia abutments compared to titanium abutments.

3.1.1. Marginal Bone Loss

Comparing abutment types, there were only non‐significant differences in marginal bone loss from baseline to the 10‐year follow‐up. Zirconia abutments had a mean difference of 0.28 mm, and for titanium abutments, the mean difference was 0.31 mm (Table 3). Nine implants had more than 1 mm of bone loss (two connected to zirconia abutments, and seven to titanium abutments). Out of these, only two titanium abutments had more than 2 mm of bone loss compared to baseline. Figure 1 displays radiographic images of marginal bone loss on an implant with a titanium abutment and one implant with a zirconia abutment at baseline and 10‐year control.

TABLE 3.

Marginal bone loss at 10‐year control compared to baseline.

Bone loss Y‐10	N	Mean	SD	Min	Max
Titanium abutment	58	0.31	0.65	0	3.13
Zirconia abutment	29	0.28	0.58	0	2.47
Total	87	0.30	0.63	0	3.13

FIGURE 1.

Radiographic images at baseline and 10‐year control.

Compared to baseline, no statistically significant effect on marginal bone loss at year 10 was seen comparing the two abutment materials (p‐value: 0.799 [−0.314–0.242]). Similarly, placement in augmented bone did not affect marginal bone loss at the 10‐year control (p‐value: 0.215 [−0.114–0.506]). For 61 implants, there were records of the bone loss at both 5 and 10 years, allowing for comparison (Table 4). Progression of bone loss between the 5‐ and 10‐year control was seen for 11 implants (18%), a majority of which had a titanium abutment (seven implants). Mean values of bone loss between 5 and 10 years were slightly higher for implants with zirconia abutments (n: 26; 0.08 mm; SD: 0.25 mm) compared to implants with titanium abutments (n: 35; 0.06 mm; SD: 0.15 mm). Bone loss values were higher between the first 5 years compared to the latter 5 years up to the 10‐year control, as seen in Graph 3.

TABLE 4.

Bone loss values between the 5‐ and 10‐year controls.

	N	Mean	SD	Min	Max
Titanium abutments	35	0.06	0.16	0	0.63
Zirconia abutments	26	0.09	0.26	0	1.17
Total	61	0.07	0.20	0	1.17

GRAPH 3.

Marginal bone loss at 5‐ and 10‐year controls compared to baseline.

3.2. Technical Complications

In total, 15 implant crowns displayed a fracture or chipping of surface porcelain from baseline to the 10‐year control (13 veneered titanium abutments and 2 veneered zirconia abutments); the difference between groups was not statistically significant (p‐value: 0.071 [−2.882616–0.1205109]). Loosening of abutment screws was seen for seven implants (five titanium abutments and two zirconia abutments); this difference was not statistically significant (p‐value: 0.638 [−2.048382–1.255909]).

On five implants, a technical complication occurred between the 5‐year control and 10‐year control (two zirconia abutments and three titanium abutments). Two implants had both a loosening of the abutment screw and chipping of surface porcelain recorded (one zirconia and one titanium abutment); see Table 5. In one patient (20 years old) when receiving an implant in the frontal region, a difference in height between the implant crown and contralateral tooth was seen.

TABLE 5.

Technical complications between 5‐ and 10‐year control.

Five to 10‐year control	Zirconia abutments (n: 29)	Titanium abutments (n: 58)
Loosening of abutment screw	1	1
Chipping	1	3
New crown a	1

^a One new crown was fabricated on an implant with a zirconia abutment due to a mobile crown that was not able to be retightened.

Success rates of implant crowns with zirconia abutments were slightly higher compared to titanium counterparts, as seen in the Kaplan–Meier survival analysis in Graph 2, where technical complications were used as outcome values. Success rates during the 10‐year period for implant crowns with titanium abutments were 82.2%, and for zirconia abutments, 93.8% (p‐value: 0.1401).

4. Discussion

The results of this long‐term follow‐up analysis on biological and technical complications on single implants with a moderately rough surface with veneered titanium or zirconia abutments display low marginal bone loss values after 10 years in function. It further finds high implant survival and success rates for both abutment types. Technical complications were reported on 10.3% of the implants during a 10‐year period, and most occurred during the first 5 years in function. The most common complication seen was chipping of covering porcelain on titanium abutments.

Compared to the results from a previous study from our group where single implants with zirconia abutments displayed slightly higher mean marginal bone loss, the opposite occurred in the present 10‐year follow‐up, where slightly more marginal bone loss was observed in implants with titanium abutments. The different outcomes between 5‐ and 10‐year results might be due to not all patients attending both control appointments, as only 61 patients attended both the 5‐ and 10‐year appointments. Two patients with zirconia abutments and pronounced marginal bone loss values did not attend the 10‐year appointment, which might have influenced the mean values. An interesting finding is that between 5 years and up to the 10 years control, the marginal bone loss rate seems to be lower than during the first 5 years in function, indicating a time‐dependent effect on marginal bone loss. This is in line with findings from a recent study on short implants in posterior positions in both jaws, which concluded higher bone loss rates during the first 5 years than during the following five [16]. Previous studies on periimplantitis have suggested an early onset of the disease, usually seen during the first 3 years in function [17]. Furthermore, wear on implant head and abutment components has been discussed as a contributing factor to marginal bone loss. Interestingly, Klotz et al. detected a decline in wear rate after 500 000 cycles for zirconia abutments [12], which could explain the reduction of marginal bone loss with time in the zirconia abutment group.

In the present study, mean marginal bone loss was below 0.5 mm for both abutment types at the 10‐year appointment. A slightly lower value as compared to a study by K. Gotfredsen on single implants in the frontal maxilla, where marginal bone loss for single implants with metal abutments and cemented metal‐ceramic crowns displayed mean marginal bone loss ranging from 0.64 to 0.86 mm after 10 years in function [18]. In similarity, Roccuzzo et al. found low marginal bone loss values in a recent 10‐year retrospective analysis on grafted compared to non‐grafted bone (−0.56 mm) [19]. In contrast, Zembic et al. found slightly higher bone loss values (1.6 mm) in a prospective study on all‐ceramic implant crowns cemented on zirconia abutments 11 years [4].

In the present study, a statistically significant difference in mucositis between the two abutment groups was found in the maxilla. On one third of the implants with zirconia abutments, mucositis was observed, compared to only one in six of the implants with a titanium abutment. Since only a very small sample of zirconia abutments was placed in the mandible (n: 3), a comparison between the two material groups with regard to mucositis was not possible. Similar mucositis outcomes were reported by Hosseini et al., who compared titanium abutments to zirconia abutments in a randomized controlled trial and found more peri‐implant mucositis around implants with zirconia abutments after 5 years in function [9]. Peri‐implant mucositis was stated as the precursor of peri‐implantitis in the 9th European Workshop on Periodontology in 2014 [20]. However, in the present study, a slightly increased marginal bone loss for implants with titanium abutments was observed, which is a somewhat contradictory finding and needs to be further studied.

Implant survival rates were high for both abutment types, reaching above 95% after a 10‐year period. This finding corroborates previous studies. In a previous study with 20 patients receiving single implants in the frontal maxilla, a 100% survival rate was seen during a 10‐year period [18]. In contrast, Rossi et al. followed 35 patients with 40 short single‐positioned implants (=6 mm) placed in posterior mandibulae and maxilla and found a slightly lower 10‐year survival rate of 91.7% [16]. From a larger cohort of 10 871 implants, treated by one periodontal surgeon in Canada, French et al. retrospectively analyzed implants in various positions of the jaw and noted a high 10‐year survival rate of 98.5% [21]. In a recent systematic review and sensitivity meta‐analysis, the overall 10‐year implant survival values were 96.4%, well in line with the findings of our study [22].

The technical complication rate was moderate: 10.3% of the total number of implants followed during the 10 years. A slightly lower value than expected. A recent article on technical complication rates in a large cohort of approximately 2000 patients in Sweden found complications to occur for 24.8% of the patients during a 5‐year period [7]. In terms of success, as defined as the absence of any technical complications, zirconia performed better than titanium abutments in a Kaplan–Meier survival analysis. Implants with a zirconia abutment demonstrated success rates of 93% after 10 years, and titanium abutments only 82%. Much of this is due to the occurrence of porcelain chipping on the titanium abutments. Compared to previous studies, Gotfredsen et al. found that single implants with metal abutments in the frontal maxilla had a 90% success rate after 10 years [18]. The latter is in line with what Zembic et al. found for single implants with ceramic crowns in the frontal‐premolar regio, where the 11‐year success rate was 90% [4].

This study has several limitations inherent to its retrospective design. Firstly, the patient sample was drawn from a single specialist clinic, which may limit the generalizability of the findings to broader populations or different clinical settings. The analysis relied on existing patient records, and the accuracy of the data is subject to the quality of documentation and potential biases in record keeping.

Moreover, due to a 10‐year follow‐up design, the retrospective nature of the study implies that long‐term data such as changes in patients' systemic health or external factors influencing implant success were not consistently available. Furthermore, the lack of randomization may have introduced selection bias, as patients receiving different abutment materials may have had underlying clinical differences or preferences that influenced the treatment choice.

Another limitation is the reliance on radiographic assessments, which can be subject to interpretation bias or technical inconsistencies across different imaging devices. Lastly, while statistical analyses were conducted to assess the differences between groups, the relatively small sample size and the observational nature of the study may limit the power to detect smaller, yet clinically relevant, differences between implant types.

5. Conclusion

In this retrospective study, both veneered zirconia and titanium abutments demonstrated high implant survival rates over a 10‐year period, without statistically significant differences in marginal bone loss between the two materials. However, the study revealed a higher incidence of peri‐implant mucositis around zirconia abutments, particularly in the upper jaw. Despite the higher frequency of biological complications, zirconia abutments showed comparable success rates to titanium abutments. Technical complications, primarily related to porcelain chipping, were observed in both groups to the same extent.

Although this study offers valuable insights into the long‐term performance of veneered zirconia and titanium abutments, further prospective, multi‐center studies with larger sample sizes are needed to confirm these findings and improve the knowledge about the impact of abutment material on long‐term implant outcomes. Moreover, further investigations of other factors that might have an impact on implant success and complication rates are still needed.

Author Contributions

Julia Olander: concept/design, data analysis/interpretation, drafting article, critical revision of article, approval of article, statistics, data collection. Victoria Franke Stenport: concept/design, drafting article, critical revision of article, approval of article.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.