Assessment and aesthetic impact of a long‐term vertical discrepancy between the single anterior maxillary implant‐supported crown and adjacent teeth: A retrospective cross‐sectional study

To assess the vertical discrepancy between implant‐supported crowns and adjacent teeth in the maxillary anterior region at least 8 years after implant placement and to evaluate the influence of this discrepancy on the level of aesthetic awareness of patients.


| INTRODUCTION
In adults, the replacement of a missing anterior maxillary permanent tooth is often performed using an implant-supported crown. This treatment is not performed in growing patients because of the infraocclusion occurring at the implant site due to the vertical development of the alveolar process and the continuous eruption of the adjacent teeth (Johansson et al., 1994;Odman et al., 1991;Thilander et al., 2001). Thilander et al. (2001) recommended delaying the treatment of young patients until the end of their growth or to consider an alternative treatment such as adhesive fixed partial bridges. However, it has been shown by Bernard et al. (2004) that despite the recommendation mentioned above, the development of infraocclusion of an implantsupported crown might occur in adults. A vertical discrepancy may develop in both young and mature adult subjects 4 years after the placement of the implant. Similar results were reported by Vilhjálmsson et al. (2013) who also observed that female patients showed a higher risk than male subjects of developing infraposition of the implant-supported crowns with respect to the adjacent teeth. The infraposition of the implant-supported crowns was also observed by Schwartz-Arad and Bichacho (2015) who claimed that the older adults presented a minor degree compared to the younger adults.
As shown by Christou and Kiliaridis (2008), large variability among individuals was reported, which might be due to the variation in the amount of secondary growth. Facial morphology and the intraocclusal forces may regulate the amount of continuous eruption (Kiliaridis et al., 2019;Winitsky et al., 2021).
Despite the huge number of implants that have been placed over the last three decades, and quite a few of them in the anterior maxillary region as single implants, the impact of the vertical discrepancy on the patients' aesthetic perception of their smile some years after the implant placement was not reported.
The aim of this retrospective observational study was to quantitatively assess the vertical discrepancy between the implantsupported crowns and adjacent teeth in the anterior region in patients treated for at least 8 years. Furthermore, the aesthetic impact of this discrepancy was evaluated as well as the patients' awareness of this situation.

| Subjects
The database of private clinics (Ardentis Clinique Dentaire) was searched for patients treated with a single implant in the anterior maxillary region at least 8 years after implant placement.
The inclusion criteria were the following: adult patients with an anterior central or lateral maxillary incisor implant-supported crown for at least 8 years; the presence of the adjacent teeth for control; no active pathology (chronic periodontal disease, peri-implantitis); no replacement of the initial implant-supported crown and no abnormal wear of the contralateral tooth.
The sample size calculation was performed using Stata/IC 16.1 and was based on the information of the primary outcome, that is, the vertical discrepancy between the implant-supported crown and adjacent anterior maxillary teeth, as obtained from a pilot study we performed on 10 subjects. The standard deviation in the vertical displacement in this pilot sample was 0.4 mm. The calculation was based on the assumption that 0.3 mm vertical discrepancy is the least clinical significant difference in the subjects treated with an implant-supported crown to replace an anterior maxillary tooth. A sample size of 21 patients was expected to have 90% power (Power (1−ß) = 0.9) to detect this 0.3 mm difference between the initial and the follow-up periapical radiographs using the paired t-test, with a 0.05 two-sided significance level (Alpha (α) = 0.05). To compensate for possible imperfections in the measurement procedure, we increased the sample by 10%.
Thus, the total sample was planned to be 23 subjects.
To reach this number, a total of 81 patients were contacted by phone or email. Twenty-nine patients could not be reached, nine were not willing to participate, eight were excluded because of orthodontic treatment, restoration of the adjacent teeth or crown replacement, two had relocated, and one passed away. Thirty-two patients accepted a complimentary examination in Ardentis Clinique Dentaire.
Previous clinical data (initial and intermediate radiographs, photographs, and periodontal assessments) and information on crown replacement were available from the patients' medical files.
From the initial sample, nine patients were excluded because of missing initial apical radiograph taken immediately after crown placement, leading to the final sample of 23 individuals required for the study. The dental implants received by the included patients (Straumann AG, Basel, Switzerland) were placed by two experienced oral surgeons. Six implants were placed immediately after extraction of the tooth and in 12 cases, a GBR (guided bone regeneration) was performed using autogenous bone graft (three cases) and xenogenous bone graft (nine cases).
This study was approved by the Ethics Committees of the University Hospitals of Lausanne (Switzerland) for human research under protocol reference number 2017-01986. Informed consent was obtained from every subject before entering the study.

| Clinical examination
A periodontal assessment of the maxillary anterior region was performed. It included probing depth and bleeding index to exclude individuals with chronic periodontal disease or peri-implantitis.
The pink and white aesthetic score as modified by Belser et al. (2009) (PES/WES) was used to assess the aesthetics of the implant restoration several years after implant placement. The five gingival parameters were assessed from 0 (worst outcome) to 2 (best outcome). The maximal possible score was 10.

| Patient satisfaction and awareness
A questionnaire was used to evaluate the perception of the patients on their maxillary anterior teeth condition (Table 1). General questions concerning their satisfaction with the aspect of their smile were asked as well as details on aesthetical characteristics like color, shape, position, and height of the teeth/gingiva. The questionnaire was mainly composed of questions from previous studies evaluating the aesthetic outcome and the patient satisfaction of individuals with anterior implant-supported crowns (Belser et al., 2008;Graber & Lucker, 1980;Vilhjálmsson et al., 2011). The patients were also asked how they perceived the changes in these features (color, shape, position, height) over time. Questions 1 to 7 were answered on a 1 to 5 scale for which 1 was considered being very unsatisfied and 5 very satisfied whereas questions 8 to 11 were binary yes/no questions.

| Radiological assessment
A periapical radiograph of the implant and the adjacent teeth was taken using the "Parallel technique" with an "X-Mind AC" or an "X-Mind System Image X" Satelec ® machine.
The radiograph taken immediately after the implant restoration (baseline T0) found in the file of the patient and the radiograph taken at the time of control (T1) were printed with high quality and 10 times magnification.
The internal calibration of each radiograph was done by comparing the radiographic implant size with that given by the manufacturer. If the implant was not completely visible on the radiograph, the interthread distance defined by the manufacturer was used as a reference.
An assessment of the eruption was performed using the implant as a stable structure ( Figure 1). Reproducible landmarks on the adjacent teeth were selected on T0 and T1 images. If no clear landmarks were found on one adjacent tooth, this tooth was excluded to minimize error. The distance between the projection of the landmark on the implant's long axis (not the crown) and the implant's neck was measured. The difference in this value measured on T0 and T1 radiographs expressed the discrepancy between the implant-supported crown and adjacent teeth.

| Statistics
The primary outcome concerning the vertical discrepancy between the implant-supported crown and adjacent anterior maxillary teeth was evaluated with a paired t-test calculation between the initial and the follow-up radiograph. Linear and logistic regression analyses were used to test the impact of the discrepancy and the objective aesthetic score (PES/WES) on the patients' awareness of this situation as perceived by the questionnaire. The level of statistical significance was set at 5%.
The error of the method was calculated by measuring twice, 1 month apart, the vertical discrepancy between implants and adjacent teeth in 15 randomly selected cases. The random error was calculated using the Dahlberg formula (Dahlberg, 1940)   Questionnaires were fulfilled by all the included patients (Table 3).

Twenty
Patients were generally satisfied with the long-term aesthetic outcome of their smile (Q1, mean: 3.9) and more specifically with the implantsupported crown (Q6, mean: 3.9). To a lesser extent, they estimated the aesthetic of their smile "about the same as everybody else" (Q3, mean: 3.0). Furthermore, patients were less satisfied with the color of their anterior teeth (Q5, mean: 3.1).
The gingival aesthetics were satisfactory for the patients. They were generally pleased with the overall aspect of their gingiva (Q7, mean: 3.3).
Seven of them had noticed changes in height of the gingiva during time After implementing linear and logistic regression analysis to test the impact of the PES/WES on the questionnaire's answers, the only F I G U R E 1 Radiograph taken at prosthesis placement (reference T0) and at the long-term control (T1). The implant length was measured on each radiograph and compared to that given by the manufacturer for internal calibration. Reproducible landmark (green) on the adjacent teeth was selected on T0 and T1 radiographs. The projection of the T0 landmark (green) on the long axis of the implant only (red) and not the crown established the baseline distance X1. The same measurement was made on the T1 image. X2 was then subtracted to X1 to determine the amount of vertical discrepancy between T0 and T1.
T A B L E 2 Sample description, implant position, follow-up period, and mean vertical discrepancy between the implant-supported crown and the adjacent tooth The patients who noticed the vertical discrepancy in their maxillary anterior region were not those who presented the higher value measured on radiographs.
The sex and age of subjects were tested as a predictor of the perception by the patient of the vertical discrepancy, but the results did not reveal any statistical significance.

| DISCUSSION
The present study has shown that all patients included in this study and followed for 12.5 years on average presented a vertical discrepancy between the implant-supported crown and adjacent teeth in the maxillary anterior region. The mean value of this discrepancy was 0.62 mm with a large range between the individuals (0.15-1.63 mm). Our results are in line with studies by Bernard et al. (2004) and Vilhjálmsson et al. (2013), and also with a recent metaanalysis performed by Papageorgiou et al. (2018) who calculated a mean implant infraposition of 0.58 mm (range: 0.33-0.85 mm).
Neither the age nor the gender of patients influenced the value of the vertical discrepancy. Literature concerning the difference between males and females is controversial ;Jemt et al. (2007) and Andersson et al. (2013) found that females had a higher risk of having more severe implant infraocclusion than males whereas Bernard et al. (2004), Winitsky et al. (2021), and Brahem et al. (2017) did not find any differences. However, the studies cited above had relatively small sample sizes. In a metaanalysis based on these studies, Papageorgiou et al. (2018) found that females had a higher risk of infraocclusion.
In the present study, the patients' age at implant placement was not associated with the severity of the infraocclusion. Similar results were found by Andersson et al. (2013), Chang and Wennström (2012), Cocchetto et al. (2019), and Bernard et al. (2004). In a previously published study, Schwartz-Arad and Bichacho (2015) showed a difference in the severity of the infraocclusion between a young group (<30 years) and a more mature group (>30 years) but the measurements were done solely on the final photographs and were not measured on the absolute vertical discrepancy but evaluated as a percentage of the crown size.
In our sample, the length of the observation period does not seem to influence the amount of vertical discrepancy, possibly due to the small number of cases with big variation among individuals.
In the past, it was thought that the placing of an implant in adult patients was stable as soon as growth had stopped. After that, multiple studies have shown that remaining vertical growth occurred (Behrents, 1985;Ghislanzoni et al., 2017;Thilander et al., 1999). This vertical growth in adult subjects was sometimes called secondary growth.
The vertical discrepancy that occurs due to the continuous eruption takes place throughout life (Christou & Kiliaridis, 2008) with a big variability. It has been shown that facial morphology and the intraocclusal forces may regulate the amount of continuous eruption (Kiliaridis et al., 2019).
Unfortunately, neither evaluation of the facial morphology of the patients nor the functional capacity of the muscles were available for this sample.
Individuals with implants in the anterior upper region were generally satisfied with the treatment result at the long-term followup. It has been shown that most of them were satisfied with the longterm result (Q1, mean: 3.9) and considered their smile aesthetically pleasing compared to the general population despite the replaced tooth. Previous studies found similar results concerning long-term satisfaction (Derks et al., 2015;Pjetursson et al., 2005;Simonis et al., 2010). In a systematic review regrouping 11 articles, Arunyanak et al. (2017) found that patients were satisfied with a mean range score of 43%-93% for peri-implant soft tissue and 81%-96% for T A B L E 3 Patient satisfaction and awareness questionnaire: Answer distribution Multiple studies showed that implant restoration was rated more satisfactory than peri-implant mucosa (Cosyn et al., 2012;den Hartog et al., 2013;Meijndert et al., 2007). Our results fall in the same range and proportion, patients were also more pleased with the implant restoration (Q6, mean: 3.9) than the soft tissue (Q7, mean: 3.3). The subjective aesthetic evaluation made by the patients was confirmed by the calculated PES/WES. The general PES at long-term evaluation was acceptable (mean: 6.69, SD: 1.7). A previous study (Belser et al., 2009) found a mean of 7.8 ± 0.8 but the follow-up period was shorter (2-4 years), which may explain the difference in our results. In our study, the PES/WES was evaluated by an orthodontist, who has been shown to be clearly more critical than other professional observers (Fürhauser et al., 2005). In our sample, the patients who had noticed a change in the height of their gingiva during time were those that had significantly lower PES scores.
Interestingly, the measured severity of the vertical discrepancy was professionals. The patients are generally less critical than dental professionals (Hartlev et al., 2014;Palmer et al., 2007). The understanding of the patients' thoughts is the key point. Therefore, recent studies are trying to develop and validate new reliable aesthetic indexes that are better associated with the patient's consideration (Li et al., 2019).
In the present study, the age at implant placement can predict neither a higher patient satisfaction nor a lower awareness of the vertical discrepancy. These results contrast with those of Derks et al. (2015) who found that older patients perceived in a more positive manner the long-term results of their implant-supported restorations (Ghislanzoni et al., 2017). Our results might be underpowered to detect similar findings.
The limitation of this study was the relatively small sample size. Nine patients were not willing to participate in the study; these refusals might introduce a selection bias since they could possibly conceal an outlier.
On the other hand, the strengths were the long-term follow-up (12.5 years) and the standardized radiological recordings that permitted measurements of the vertical discrepancy detected on the adjacent teeth.
Even with a vertical discrepancy evolving over time up to 1.63 mm, it does not elicit demands for review or re-intervention by patients.
Nevertheless, general dentists and oral surgeons should inform patients about this potential long-term complication. Currently, dental research has not identified clear factors predictive of a higher risk for the vertical discrepancy in adult patients.

| CONCLUSION
All studied implant-supported crowns showed infraocclusion occurring since a mean time of 12.5 years after implant restoration. The mean measured value of the vertical discrepancy was 0.62 mm. The amount of vertical discrepancy was not dependent on gender nor the location of the replaced anterior tooth (central or lateral incisor). Furthermore, the age of patients at implant placement and duration between initial and recall recordings were not associated with the severity of the infraocclusion.
Smile aesthetics were generally satisfactory despite involving an implant-supported crown in the anterior maxilla. A minority of patients had noticed the implant infraocclusion (35%) and 17% found the aesthetic defect severe enough to be willing to correct it.

AUTHOR CONTRIBUTIONS
All authors were involved in the conception, design, data acquisition, data analysis, interpretation, drafting of the manuscript, and its critical revision. All authors gave final approval of the version to be published. The study was supervised by Stavros Kiliaridis.
F I G U R E 2 Female patient with a single implant-supported crown replacing the 21 with an important vertical discrepancy (0.9 mm) after an 11-year follow-up who is fully satisfied with the aesthetic outcome of the treatment and did not perceive any changes during time.