The Correlation of Clinical Outcomes (Marginal Bone Loss, Probing Depth, and Patient Satisfaction) With Different Prosthetic Aspects of Implant Overdentures: A Five-Year Retrospective Cohort Study

Given the effect of prosthodontic treatment of edentulous patients on their quality of life, improvement of clinical or patient-centered outcomes should be considered.¹  For most cases, the role of dental implants has been accepted regarding increasing retention, stability, support, and subsequently comfort and patient satisfaction with removable dentures.^1–3  Until now, many researchers have considered different aspects of implant overdentures to enhance the success of treatment.^4–7 

The long-term efficiency of implants is desired for both the clinicians and patients.^8,9  Some factors, such as bone quality and quantity and an adequate band of keratinized mucosa, are related to the host condition and difficult to improve.¹⁰  Therefore, the selection of factors such as wearing behavior, maintenance conditions, and oral health care can be improved to decrease the marginal bone loss (MBL) around implants.^11–17 

Some studies have reported better clinical outcomes by increasing the number of abutments,^12,18  while some others have found no significant effect on the outcomes by increasing the number of implants.^19–22  Also, a meta-analysis showed less MBL and higher success of implants for single implant overdentures compared to 2-implant overdentures.⁵  A study reported a positive association between increment of implant diameter and higher MBL.²³  Another study reported a positive relationship between the narrow diameter of implant and higher MBL compared with regular implants,⁴  while one study reported the same findings for conventional or mini-implants.²⁴ 

Some studies have reported no difference in the outcomes among different types of attachment systems,^13,21  while another study reported opposing results.²⁵  Both trauma and microbiological factors have been considered etiologies in failures and MBL of implants.^26–30  These issues have caused many controversies among researchers.

Presented controversies persuaded us to design this study: There is controversy about ideal prosthetic aspects and the importance of bio-mechanic or biologic factors in successful implant therapy and some unapproved claims about the treatment protocol as the type of attachment. This study aimed to present descriptive statistics of the clinical status of implant overdentures and investigate the relationship between different treatment plan details of mandibular implant overdentures with patient satisfaction as well as clinical outcomes as marginal bone loss and success rate.

The null hypothesis was that there would be no differences in patient satisfaction, marginal bone loss, and probing depth among implant overdentures with different prosthetic aspects and varied oral health condition.

This clinical retrospective cohort study was performed at two specialized dental implant clinics as two sampled clusters in Isfahan. The study population was selected among complete edentulous patients treated by root form implant overdentures between June 2012 and July 2016. All participants had been treated by a special team of surgeons and prosthodontists using the same fabrication protocol.¹⁰  The overdenture wearers reported the following conditions that disqualified them from entering the study: chemotherapy/radiotherapy, taking bisphosphonates or corticosteroid drugs, pregnancy, or uncontrolled systematic disease. The minimum age of the studied overdenture had to be 12 months. Seventy-four participants were included in the study after meeting the aforementioned criteria and providing written informed consent. The study was performed in accordance with World Medical Association Declaration of Helsinki. Ethical principles for medical research involving human subjects and ethical approval of the protocol was granted by Isfahan Regional Bioethics Committee.

The whole study population was examined and evaluated by an operator. Long-cone parallel vertical bite-wing photostimulable phosphor (PSP) plate-based radiographs (Dürr Dental AG, Bietigheim-Bissingen, Germany) were taken for all the studied implants by a radiologist. All examiners, recorders, and radiologists were completely blind to the clinicians who fabricated the overdentures.

A fulfilled checklist used for all samples presented different records, including age, sex, overdenture age, number and characteristics of implants at both jaws separately, type of attachment system (ball, bar, and bar-ball), opposite dentition (natural dentition, implant overdenture, and conventional denture), presence or absence (P/A) of annual recalls, P/A of the symmetry of position of implants, P/A of systemic disease, P/A of smoking, P/A of the signs of tissue irritation, P/A of the need for occlusal adjustment, P/A of the need for relining overdentures, P/A of nocturnal exclusion, P/A of the need for matrix/patrix adjustment, patient satisfaction (100 mm visual analogue scale),^25,31  marginal bone loss, probing depth (PD), P/A of bleeding on probing,^10,24  and P/A of mobility. Also, the condition of attached gingival width¹⁰  by 1–3 for “no, less than 1 mm, equal or more than 1 mm,” respectively, and gingival index and plaque index adapted from Loe and Silness^10,19,24  with 0–3 (“0″ presents “no”) was recorded.

In entering data into the software, all the items that had yes/no states (P/A) scored “1″ were allocated to “yes” or “presence” and scored “2” to “no” or “absence.” Taking into account the variation in the baseline radiography of the participants, MBL was measured using a digital ruler in the PSP software from the first implant thread to the crest of bone.^17,32  Negative value was allocated to the apical crest rather than the first thread, and positive value was given to the coronal level of bone. The mean of the bone level at both mesial and distal sites of each implant was calculated as implant MBL. Also, for each participant, the mean MBL of all implants in one jaw was calculated to perform the comparisons.⁴ 

To record the PD of each implant using millimeter-graded color-coded periodontal probe (PCP 15, Hu-Friedy, Chicago, Ill) until pain was felt (0.24 N), a single operator who had instructed for evaluating PD examined all the population. The probing depth was measured at 4 circular points, including mesiobuccal, midbuccal, distobuccal, and midlingual. The mean PD was calculated first for each implant and then for each jaw.^4,8,33 

Early survival rate refers to considering the failed implants during the time interval between implant insertion and prosthetic rehabilitation initiation. This item was evaluated based on the patients' files. Late survival rate demonstrates the number of implants remained in place after overdenture delivery to follow-up visits for the present study.^9,31  Successful dental implant is defined as implants with no mobility, no radiographic radiolucency along with fixture, no infection, and being functional.^27,31  Considering these criteria, the success rate was calculated.

The normality of data distribution was analyzed by the Kolmogorov-Smirnov test. Mean ± standard deviation and median (interquartile range) were reported to present the scale of data with normal distribution and without normality, respectively. Categorical data were presented by frequency and percentage. The relationship between MBL and PD with independent variables was measured by Pearson correlation and Spearman's rho correlation. To evaluate tissue irritation differences among independent variables, the chi-square test was used. Analysis of covariance (ANCOVA) included both analysis of variance and general linear regression, simultaneously. ANCOVA was used to compare MBL, PD, and variation of patient satisfaction among independent variables. To distinguish the effect of selected independent variable and manage the confounding factors, data about all other variables with possible confounding effects (called covariates) include age, sex, overdenture age, implant brand, implant symmetry, opposite dentition, annual recalls, controlled systemic disease, smoking, tissue irritation, need for occlusal adjustment, need for relining, nocturnal exclusion, need for matrix/patrix adjustment, bleeding on probing, mobility, attached gingival width, gingival index, and plaque index. These data were collected and adjusted for each comparison. The final analysis included Tukey and Tamhane post hoc tests for binary comparisons of significant items resulting from ANCOVA. Data analysis was performed by a statistician blind to the data using IBM SPSS version 24 statistical software. The alpha level was considered 0.05 for all tests.

As shown in Table 1, 74 participants and 186 implants were investigated. The mean ages of participants and overdentures were 61.62 ± 9.28 (year) and 38.45 ± 16.64 (month), respectively. Descriptive data about the studied implants and overdentures were reported in Tables 2 through 4. From 186 studied implants, 17 implants were in the maxilla of 5 participants and 169 in the mandible of all 74 samples. The mean MBL and mean PD of 17 maxillary implants and 12 mandibular implants in 5 participants who wore implant overdenture at both jaws were reported in Table 5. The cumulative survival rate after prosthetic delivery was 100%, and the cumulative success rate was calculated to be 97.8% based on the criteria.

Table 6 shows the relationship of two major dependent variables—MBL and PD—with overdenture age and four factors affecting and presenting the health of implants surrounding tissues. The values for coefficient of correlation including 0 to 0.3, 0.3 to 0.7, and 0.7 to 1 (0 to −0.3, −0.3 to −0.7, and −0.7 to −1) indicate weak, moderate, and strong positive (negative) linear relationship, in that order.³⁴  Higher PD was moderately associated with a higher level of MBL. Increase in overdenture age had a weak significant correlation with higher MBL but not with higher PD. The higher width of attached gingiva had a weak correlation with lower PD significantly. PD was positively correlated with gingival index (moderate) and plaque index (weak) while lack of bleeding on probing was moderately correlated with PD negatively.

Tissue irritation was another dependent variable whose correlation with independent variables was studied using chi-square test (Table 7). Participants who had regular annual recalls and nocturnal exclusion of dentures showed significantly less tissue irritation. The need for occlusal adjustment provided significantly more tissue irritation.

Three dependent variables, including patient satisfaction, PD, and MBL were compared among all independent variables (Table 8). These comparisons were carried out by ANCOVA, with adjustment of all other independent variables for each comparison. Both MBL and PD did not differ significantly among three types of attachment systems and among the different number of mandibular implants placed for mandibular overdenture.

Finally, the Mean and Standard deviation and results from binary comparisons for significant items from Table 8 were presented in Table 9. The mean PD for participants who had annual recall was 2.85 ± 1.5, and for the group lacking, PD was 2.65 ± 1.01. The mean MBL for those who needed matrix/patrix adjustment was significantly more than those who did not. Implants placed in areas lacking attached gingiva had higher PD and mean MBL. Implants manifesting higher gingival indices were surrounded by higher PD sulci.

The null hypothesis that there would be no differences in patient satisfaction, marginal bone loss, and probing depth among implant overdentures with different prosthetic aspects and varied oral health condition was partially rejected. While the type of attachment and number of implants supporting mandibular overdenture did not influence the clinical outcomes of dental implants, cases that needed matrix/patrix adjustment showed more MBL.

This study found no difference in patient satisfaction among other independent variables. This finding is in accordance with the results of a study that reported more satisfaction with implant overdenture than with conventional dentures and concluded that factors such as age, sex, treatment age, attachment type, rehabilitated jaw, number of implants, and splinting did not affect the satisfaction level.³ 

The first standard choice of care in the treatment of edentulous patients is 2-implant mandibular overdenture and maxillary conventional denture.¹²  However, 1-implant mandibular overdentures are accepted in some studies.^5,19  The multi-factorial nature of MBL around dental implants has been accepted. Implant design, host response, bone quality and quantity, and biomechanical and microbiological factors were considered effective in the longevity of implants.^15,17,26,27 

One meta-analysis reported no significant effect for the type of attachment in implant overdenture on implant around outcomes.¹¹  Two systematic reviews concluded that splint or no-splint systems show the same patient/clinical outcomes.^13,22  The results of this study in that respect confirm the findings of these two reviews.

The participants who had regular annual recalls and nocturnal exclusion of dentures showed significantly less tissue irritation. Another item, need for occlusal adjustment, showed significantly higher tissue irritation. These findings are in accordance with the knowledge available on the maintenance phase of overdentures, need for regular follow-ups, and the role of occlusal loads on mucosal health.^14,26 

Considering the small sample size and lack of statistical comparison, maxillary implants manifested more MBL and PD than mandibular implants. The mean 5-year MBL around maxillary implants supporting overdentures was 0.51 mm in one study.²⁰  The result of a systematic review presented that more complications and less success correlated with maxillary implants rather than mandibular implants.⁷  The findings of the present study confirmed these results; however, the small sample size of maxillary implants may cause non-comparable and non-deductive but higher MBL.

Based on the criteria of the present study, the survival rate of post-prosthetic delivery was 100%, and the cumulative success rate was calculated to be 97.8%. A study with 5 years follow-up reported a 97% survival rate and a 90% success rate.³¹  In another study with a mean follow-up of 14 years, 92% survival rate and 85% success rate were reported.²⁶  The results of the present study are in agreement with previous studies with regard to the duration of follow-up. Overall, 5%–7% failure rate has been accepted for implant therapy.²⁷  Hence, the treatment level of the studied population is acceptable. The higher failure rate in immuno-suppressed people receiving dental implants suggests involvement of systemic health in successful treatment and effect of microorganisms in implant failures.²⁷ 

As expected, a higher PD was associated with a higher level of MBL. Also, an increase in overdenture age was related to higher MBL, not higher PD. The results of the present study show that 0.19 mm bone loss from the first thread in a mean age of about 3 years. That is to say, our study indicated approximately 0.1 mm bone loss per year after the first year. This result is in line with the universal criteria. Based on the current criteria, about 1 mm at the first post-insertion year and 0.1 mm annual bone resorption after the first year are acceptable conditions for dental implants.^5,11 

One controversial point is the difference between the bone loss in the first year and later years (1 mm vs 0.1 mm). Plaque accumulation is the same over the years, while trauma and occlusal loading and healing amounts changed over the years. This can point to the role of biomechanical factors in MBL. In our results, better conditions for the three factors involved in presenting the healthcare and gingival inflammation—that is, gingival index, plaque index, and bleeding on probing—were significantly correlated with lower PD. However, these three factors were shown to have no significant relationship with MBL. The findings of the present study indicate the higher role of biomechanics than of microbiology. The mean MBL for those who needed matrix/patrix adjustment was −0.32 ± 1.37 and for those who did not need it was 0.05 ± 0.6. This finding confirms the role of biomechanics and also verifies the importance of denture follow-up appointments. Because an unstable overdenture will promote more biological/mechanical complications, the clinicians should reevaluate the denture occlusion and denture-tissue adaptation and improve these items to optimize denture stability and provide best support from primary stress-bearing areas for the overdentures. However, some studies have reported that occlusal overload and unadjusted occlusion can influence long-term implant stability and increase MBL.^16,26  Another study has highlighted the role of microorganisms and gingival inflammation.²⁷ 

The important role of an adequate band of keratinized tissue in the health of peri-implant mucosa and plaque control and preventing gingival inflammation has been accepted,¹⁰  which is verified by the present study. Nonetheless, higher attached gingival width was correlated with lower PD, it was not correlated with MBL. Based on these results, the role of biomechanics in the MBL around dental implants is more significant than that of micro-organisms during 1–5 years follow-up. Hence, the authors propose the factors for MBL around dental implants as follows:

In the first year of post-prosthetic delivery, the following factors seem to be more important and in this order: uncompleted healed osseo-integration after surgical trauma as a major factor; creating biological width, biomechanics and occlusal overload; peri-implantitis caused by microorganisms; and implant design factors as micro-gap and implant crest module.

After the first year with the completion of the healing, the order of factors' importance seemed to change to: biomechanics and occlusal overload as the major factor, followed by microorganisms. However, with a rise in MBL and subsequently PD over time, plaque accumulation increases; therefore, the role of biomechanics decreases and also the important role of microorganisms increases.

The importance of sufficient soft tissue thickness and attached gingiva as a biological seal has been well accepted. Previous studies found direct connection between preserving bone around dental implants and pre-implant mucosa of edentulous alveolar ridge.^29,30  While most of studied implants in this study were tissue level, the results of this study are in agreement with these studies regard to necessity of sufficient attached gingiva around the dental implants. It seems the quantity and quality of peri-implant soft tissue is more important than tissue level-bone level. This assumption should be tested in future studies.

One of the limitations of this study is a result of the type of study. Correlation studies can investigate relationships; however, based on a problem-oriented approach, causes need to be found in order to control undesired effects. This item will not be realized using cross-sectional studies. Nevertheless, the importance of these studies regarding ideas for new research should be emphasized. Accordingly, considering the findings of the present study about the width of attached gingiva, it is proposed for future research; the success rate and bone loss around dental implants received gingival graft compared to without gingival graft.

Lack of control on the number of participants in each group (such as type of attachment, brands of implants, and type of opposite dentition) was another limitation in performing this study and data extraction. This item is also true in cross-sectional studies. Furthermore, for evaluating the success of dental implants, studies with longer duration of follow-up are preferred. Hence, short duration of follow-up of this study is another limitation. Therefore, clinical trials with more than 10 years follow-ups are recommending.

Other limitations of this study include small sample size and lack of baseline radiography as same as follow-up radiography. These items were considered as much possible in statistical analysis and measurement of MBL from the first thread, respectively. Since the shear forces exerted on the bone are converted into compressive forces at the first thread, the MBL slows down at this area.^17,32  Therefore, the MBL was measured from this point of implants. The follow-up radiography was long-cone parallel vertical bite-wing PSP plate-based radiographs including digital ruler in the PSP software. This standardized radiograph as one of the most precise 2-D dental radiographs, which increases the accuracy of measuring MBL.

Other limitations are an unequal number of participants in different groups for some independent variables such as the brand of implant and type of attachment as well as the presence of confounding factors. These problems usually are observed in retrospective uncontrolled studies. However, an attempt was made to neutralize these limitations by using ANCOVA. While some factors (age, sex, overdenture age, implant brand, opposite dentition, implant symmetry, and controlled systemic disease) did not show significant influence on clinical outcomes, adjusting these factors using ANCOVA reduced the possible confounding effect of these variables.

The authors recommend that future investigations embark on long-term longitudinal studies on measuring MBL and other factors involved in implant health and evaluating all etiologies for implant failure in 6-month periods to calculate and improve the factors listed in the above-mentioned model.

With the caution of our study limitations, the results of this study present the following:

• Healthy peri-implant mucosa is significantly correlated with the presence of attached gingiva and appropriate oral health care.

• Clinicians and patients receiving implant overdentures should be informed about the importance of regular annual recalls. Also, patients should be trained about denture maintenance and nocturnal exclusion of dentures.

• In each recall visit, the condition of occlusion, denture base adaptation, and matrix/patrix of attachment should be rechecked and adjusted if necessary to provide optimum overdenture retention, stability, and support, thereby decreasing complications.

• The type of attachment and number of implants supporting mandibular overdenture do not influence the success of an implant and MBL around dental implants.

Abbreviations

ANCOVA:

analysis of covariance

MBL:

marginal bone loss

P/A:

presence or absence

PD:

probing depth

PSP:

photostimulable phosphor plate-based radiographs

We gratefully acknowledge the Isfahan University of Medical Sciences (#1396.324) for ethical approval and financial support.

The authors have stated explicitly that there are no conflicts of interests in connection with this article.