Correlations between clinical parameters in implant maintenance patients: analysis among healthy and history-of-periodontitis groups

This study enrolled 55 patients (25 men, 30 women; mean age, 63.53 ± 10.51 years) who visited Nihon University School of Dentistry Dental Hospital for implant maintenance from April to September 2016. The inclusion criteria were treatment at the same hospital with two-stage implant placement before 2015, and ongoing maintenance at 3–6-month intervals after superstructure placement. Patients whose superstructure had been in place for less than 3 months were excluded. The dental implants used were Replace Select™/Steri-Oss® implant system (Novelbiocare, Switzerland), Novel Replace® (Novelbiocare, Switzerland), OsseoSpeed™ (ASTRA TECH Implant System, Switzerland), OSSEOTITE® XP (BIOMET 3i, Indiana, USA), and Brånemark system® Mk III (Novelbiocare, Switzerland); all implants had a rough surface (Table 1). This study was conducted with the approval of the Nihon University School of Dentistry Ethics Committee (Permit Number: 2013-15).

Patients with chronic periodontitis were diagnosed based on the latest AAP Classification of Periodontal Diseases and Conditions at the first visit. Clinical attachment level (CAL) was examined for each individual tooth. A CAL of 1–2 mm was defined as mild periodontitis, 3–4 mm was moderate periodontitis, and 5 mm or more was severe periodontitis. The range of the disease was considered to be localized when 30% or less of all teeth were affected, and generalized when more than 30% of teeth were affected [16]. Cases of aggressive periodontitis were not included in this study. Implant therapy was performed after active periodontal treatment with confirmed improvement of the periodontal tissue. We classified patients into two groups based on their medical records: a healthy group without a history-of-periodontitis (H) and a history-of-periodontitis group (HP).

We recorded the following information from the medical chart: patient sex and age, number of natural teeth at first visit, final number of natural teeth at the beginning of implant treatment, total number of extracted teeth and installed implants, and average maintenance period.

At the time of maintenance, the status of the peri-implant tissue was evaluated. Peri-implantitis is defined as a chronic infection of peri-implant tissue, with resorption of the supporting bone observable radiographically. Peri-implant mucositis is a reversible inflammatory reaction around the implant [17]. We diagnosed peri-implantitis when the probing pocket depth (PPD) around the implant was 6 mm or more, suppuration and bleeding were observed at the time of probing, and bone resorption was present radiographically for 25% or more of the implant length [18].

At the time of optional maintenance, the following parameters were measured for each implant. Data collection and analysis were conducted by two operators who were blind to the research purpose and methods. One operator statistically analyzed the final data set.

Data analysis was performed with statistical software (GraphPad Prism 5.0; GraphPad Software Inc., San Diego, CA, USA). Quantitative analysis of differences between the H and HP groups was performed with the Mann–Whitney U test. Clinical parameters were analyzed with Spearman’s rank correlation coefficient (R _s ). A significance level of 95% (p < 0.05) was considered statistically significant.

Of the 55 subjects, 25 were male and 30 were female (H group: 8/10, HP group: 17/20, respectively). A total of 130 implants (H group 39, HP group 91) were included in the analysis (Table 2). The mean maintenance period was 6 years and 6 months. The average number of natural teeth at the first visit and at the beginning of implant treatment were 25.3 ± 2.9 and 22.4 ± 4.5, respectively. Comparison between the H and HP groups revealed a significant difference in patient age at the time of maintenance, number of natural teeth at the beginning of implant treatment, total number of extracted teeth, and time since implant placement.

Among all the subjects, 14 implants (10.8%) had peri-implantitis, all belonging to the HP group (6 patients). Details of periodontitis diagnosis in the HP group were generalized moderate periodontitis (22 patients, 53 implants), generalized moderate periodontitis + localized severe periodontitis (11 patients, 23 implants), and generalized severe periodontitis (4 patients, 15 implants) (Table 3). The proportion of peri-implantitis for each diagnosis were 5.7% (3/53) for generalized moderate periodontitis, 13.0% (3/23) for generalized moderate periodontitis + localized severe periodontitis, and 53.3% (8/15) for generalized severe periodontitis at the implant level. The prevalence of peri-implantitis in the HP group was 15.4% (14/91 at the implant level) and 16.2% (3/37 at the patient level).

The data from each subgroup are shown in Table 4.

The mPI score for the total study population was 0.21 ± 0.30 (mean ± SD). The mean scores of the H and HP groups were 0.18 (median, 0.25) and 0.23 (median, 0.25), respectively.

The PPD of the total study population was 3.19 ± 0.96 mm (mean ± SD). The mean PPDs of the H and HP groups were 2.87 mm (median, 2.83 mm) and 3.33 mm (median, 3.00 mm), respectively.

The BoP score of the total study population was 0.28 ± 0.29 (mean ± SD). The mean scores of the H and HP groups were 0.30 (median, 0.17) and 0.28 (median, 0.17), respectively.

Because normality and homoscedasticity were not obtained for each clinical parameter between the two groups, a statistical study was conducted using the nonparametric test (Mann–Whitney U test) for the difference test between the two groups. PPD was recognized to be significantly greater (p < 0.01) in the HP group, but there were no significant differences in mPI and BoP between the two groups. Therefore, we focused on the PPD parameter, in which a significant difference was found between the two groups, and analyzed correlations between PPD and patient characteristics (age, number of teeth, number of extracted teeth, maintenance period). Significant correlation coefficients were observed for the number of extracted teeth in the H group and for all characteristics in the HP group (Table 5).

We analyzed the correlation between PPD and other parameters in each group using Spearman’s rank correlation coefficient. We found a weak correlation between PPD and BoP for the H group (r _s  = 0.401, p < 0.05). In the HP group, weak correlations were found between PPD and mPI (r _s  = 0.228, p < 0.05) and between PPD and BoP (r _s  = 0.241, p < 0.05) (Table 6).

In this study, we reviewed the clinical parameters of 130 implants in 55 patients during ongoing long-term maintenance. We focused on the history of periodontitis and compared the parameters in two groups classified according to a history of periodontitis. A previous study on the prevalence of periodontitis revealed that moderate or severe periodontitis was observed in 64% of people over 65 years of age [19], and that the number of remaining teeth also decreased after the age of 60 [20]. Consistent with these findings, our study suggests that periodontal treatment was started at an older age in the HP group. Previous reports recommended that the follow-up period evaluating the peri-implant tissue should be 5 years or longer. The mean maintenance period of this survey was 6 years and 6 months in total. Therefore, we consider that the observation period of this study was sufficiently long-term and reasonable, resulting in a potentially predictive result [3]. The implant maintenance period of the HP group was longer than that of H group, possibly because the HP group consisted of patients with a history of good compliance who had visited the university hospital for long-term periodontal treatment and understood the importance of periodontal treatment and were likely to continue with maintenance visits after the implant treatment. However, it was expected that patients in the H group are often only partially treated and they had less experienced to receive comprehensive or long-term treatment. We hypothesized that the patients of H group might have felt less necessity of receiving maintenance because they were healthy.

A history of periodontitis has been recognized as an important risk indicator for peri-implantitis [2, 3, 21] and is known to lower the success rate during maintenance. Thus, for long-term implant stability, it is important to perform appropriate periodontal treatment [22‐24].

The difference between the number of natural teeth at the first visit and at the beginning of implant treatment indicates that many teeth were extracted in the HP group compared with the H group during the active treatment period. One limitation of this study is that we did not investigate the reasons for tooth extraction. We inferred that in the HP group, many teeth were removed because of periodontitis, whereas a smaller number were extracted in the H group because of root fracture or apical periodontitis. There was no significant difference in the number of implants between the groups. However, the number of natural teeth before implant treatment was smaller in the HP group than in the H group. It is possible that the implants in these patients were used as bridge abutments for wide defects and that the size of the defect was not necessarily reflected in the number of implants.

All the implants with peri-implantitis came from the HP group, not the H group, thus supporting the view that a history of periodontitis is a risk indicator for peri-implantitis. In the diagnosis of periodontitis within the HP group, it is notable that peri-implantitis increased as periodontitis become more severe. This supports the findings of a previous study that reported that the survival rate of implants is inversely proportional to the severity of past periodontitis [25]. In our study of long-term maintenance patients, the findings that the prevalence of peri-implantitis was 10.8% at the implant level and 10.9% at the patient level were similar to previous findings. It was suggested that the criteria for peri-implantitis that we adopted were reasonable. The low prevalence of peri-implantitis in this study (approximately 10%) may be related to the fact that all the subjects received maintenance therapy [26].

The average mPI was about 0.2 for both groups, indicating little plaque adhesion. One characteristic of patients in this study is that they understood the importance of maintenance and they received oral hygiene instruction. Although poor oral hygiene has been reported to be a risk indicator for peri-implant disease [3], our results suggest that the oral hygiene was maintained in all patients including peri-implantitis patients.

Among all the subjects, 11.5% of the total implant sites had an average PPD of more than 4 mm. Our results were comparable with previous studies, which have reported peri-implant PPDs ranging from 2.52 to 3.8 mm [27‐29]. However, we believe that this study was more thorough because of its long-term follow-up. PPD in the HP group was 3.33 ± 1.07 mm, which was significantly greater than that in the H group (2.87 ± 0.48 mm). Interestingly, we found that PPD was greater in patients with a history of periodontitis than in those without this history. It has been reported that worsening periodontal disease of natural teeth affects the pocket depth of adjacent implants [30]. However, although in natural teeth there is a criterion of critical probing depth (4 mm or more) [31], there are no such reference values for implants because the site and placement depth differs and the biological width of the implant is not constant [32].

BoP has been used to evaluate inflammatory conditions of periodontal tissue [33] and can also be an important evaluation item for peri-implant tissues. To avoid diagnosing bleeding resulting from strong probing as a false positive, we set the probing pressure to 0.15 N [34]. BoP values were low, with no significant difference between the groups (H group, 0.30; HP group, 0.28). This result was similar to that of a previous study that reported a low bleeding index in the implant group with a low plaque index [35].

There was no significant difference between mPI and BoP when clinical parameters were compared, and there was a significant difference for PPD only in the HP group. For this reason, we analyzed the correlation of the characteristics of patients with significant differences only for PPD. In the H group, a significant correlation with PPD was found only for the number of extracted teeth, suggesting that PPD was correlated with the number of teeth extracted due to reason that it was not periodontitis.

However, in the HP group, age, number of teeth, extracted teeth, and maintenance period all correlated with PPD. This finding suggests that the implant PPD reflects the period required for periodontitis treatment and complexity of treatment before implant placement. Although the number of teeth at the beginning of implant treatment showed only a negative correlation, the small number of teeth present indicates that a large number of teeth were extracted as a result of periodontitis, suggesting that the implant PPD is affected by this. From these results, it can be speculated that the severity of periodontitis before the implant treatment is also reflected in the PPD.

Correlations between PPD and the two other clinical parameters, mPI and BoP, were examined in each group. Because this study targeted all patients who received maintenance, it was predicted that the hygiene around the implants in both groups would be good. However, there was a significant difference between PPD and mPI only in the HP group. It was assumed that the probing depth and oral hygiene around implant appeared to be affected each other in the HP group. In contrast with the target group of our study, there may have been a significant correlation between PPD and mPI in a patient group that had not received maintenance and that had poor oral hygiene.

A correlation between the PPD of the peri-implant tissue and lesion progression has been reported, similar to that seen with periodontitis [27]. Probing around the implant is useful for evaluating soft tissue inflammation and is considered reproducible [36]. Examination of peri-implant bleeding is reported to have higher diagnostic accuracy than probing around natural teeth [37]. However, because measurement results differ depending on the shape of the superstructure and because bleeding can occur even if peri-implant tissue is healthy, implant probing is not completely established as a method to distinguish between healthy and inflamed sites [37, 38]. As in natural teeth, we consider that peri-implant BoP should be assessed after checking for sulcus bleeding. The results of this study, especially the correlation between PPD and BoP, suggest that implant probing is useful for evaluating peri-implant tissue.

In this study, we measured clinical parameters at only one point during the maintenance period. Further investigation is needed with attention not only to the amount of parameter change over time but also to bone resorption on radiological images.