The role of systemic diseases in the development and progression of peri-implantitis remains unclear and requires discussion from various perspectives. This retrospective cohort study aimed to evaluate whether the type of systemic disease affects the onset time of peri-implantitis. The cohort consisted of patients who underwent implant maintenance evaluations between January 1998 and June 2020. Information on age, sex, history of periodontal disease, smoking habits, body mass index, systemic diseases, implant placement sites, and diagnosis of peri-implantitis were obtained. The relationships between the time of onset of peri-implantitis and the medical history of systemic diseases were determined statistically by applying the Kaplan-Meier analysis method and log-rank test. A total of 216 implants in 89 patients (34 male and 55 female) were included in this study. The average patient age was 53.9 ± 11.8 years at the first visit, and the mean duration of maintenance was 7 years and 4 months. A total of 43 patients had medical histories and were assigned to the systemic disease group. In this group, the overall prevalence of peri-implantitis was 25.4% (29 of 114 implants). The medical history of systemic diseases (odds ratio [OR], 6.87; 95% confidence interval [CI], 2.37–19.9) and dental history of periodontitis (OR, 3.64; 95% CI, 1.25–10.6) were assessed as risk factors for peri-implantitis. A significant difference in the time of onset of peri-implantitis was confirmed between patients with systemic disease and healthy patients. Diabetes mellitus, osteoporosis, and hypertension had a significant impact on the earlier onset of peri-implantitis.
Introduction
Implant therapy is the most promising and effective rehabilitation method for treating patients who are partially or completely edentulous.1 Implant survival rates are above 90%, and implant therapy is recognized as a highly predictable restorative treatment.2 However, this treatment modality has limitations, such as being susceptible to peri-implantitis, a major biological complication that has a significant negative impact on patients' oral health, finances, and quality of life.3 Peri-implantitis is defined as a plaque-associated pathologic condition occurring in tissues around dental implants, characterized by inflammation in the peri-implant mucosa and subsequent progressive loss of supporting bone.4 Various treatment methods, mainly plaque removal from the implant surface, have been studied for their therapeutic effects.5
Researchers have conducted various studies on risk factors for the development of peri-implantitis, which affects both the patient and dental practitioner.6,7 Considering that oral bacteria cause progressive bone resorption, the causes and pathogenesis of peri-implantitis are similar to those of periodontitis.8,9 As with periodontitis, the peri-implant condition is dependent on the equilibrium between the microbiota and host immunoinflammatory reactions. Therefore, factors that alter either or both these components increase the risk of disease. Risk factors for exacerbating periodontitis include systemic factors, such as diabetes and smoking, that affect dental treatment.10 Researchers have reported that the prevalence of (1) cerebrovascular diseases, (2) malignant neoplasms, (3) hypertensive diseases, (4) cardiovascular diseases, and (5) spondylopathy is high in people older than 65 years of age.11 Additionally, these patients often present with possible multiple comorbidities. From a pathologic point of view, these diseases affect the health of implants in aging patients.12 The roles of systemic diseases in the development and progression of peri-implantitis remain unclear and require discussion from various perspectives.13 Therefore, this retrospective cohort study aimed to evaluate (1) whether a medical history of systemic diseases would be a risk indicator for the development of peri-implantitis and (2) do the types of systemic diseases affect the onset time of peri-implantitis?
Materials and Methods
This study was approved by the Nihon University School of Dentistry Ethics Committee (Permit EP16D013) and was conducted in accordance with the Helsinki Declaration of 1975, as revised in 2013. All patients provided written consent for inclusion before participating in the study. The study was performed following the Strengthening the Reporting of Observational Studies in Epidemiology guidelines for observational/descriptive studies.14 The cohort consisted of patients who visited the Department of Dental Implants at the Nihon University School of Dentistry Hospital for maintenance treatment between January 1998 and June 2020. The datasets used and analyzed during the current study will be available from the corresponding author upon reasonable request.
Inclusion criteria
Patients were diagnosed, an appropriate treatment plan was developed, and then the patients underwent periodontal treatment, implant surgery, prosthetic procedures, and maintenance by a single periodontal specialist (KS). The inclusion criteria obtained by reviewing medical records from 2003 to 2019 were as follows: (1) patients aged 21 years or older who had implants placed at the same hospital, (2) patients who had been functioning for a minimum of 6 months after superstructure placement, and (3) patients who received regular maintenance at intervals of 3–6 months.
Surgical protocol for implant placement
All implant surgical procedures were conducted under local anesthesia alone or with intravenous sedation as a 2-stage approach, following the manufacturer's recommended methods. In patients with insufficient bone quantity, simultaneous bone grafting was performed using autogenous or artificial bone during implant placement. As a postoperative medication, the patients used oral antibiotics (cefuroxime axetil, cefcapene pivoxil hydrochloride hydrate, and amoxicillin hydrate) 3 times daily for 3 days after either the primary or secondary surgery to prevent infection. Painkillers (diclofenac sodium and loxoprofen sodium hydrate) were used according to their symptoms. Following primary or secondary surgery, all sutures were removed after 1 week, and the patients were instructed on oral hygiene maintenance.
Diagnosis of peri-implantitis
After placing superstructures, maintenance treatments were provided at intervals of 3–6 months according to each patient's oral hygiene compliance. During follow-up visits, peri-implant diseases were diagnosed by probing and radiographic examination of the peri-implant tissues. The diagnosis of peri-implantitis was based on the World Workshop on the Classification of Periodontal and Peri-Implant Diseases and Conditions statement.4 This took into account inflammation of the peri-implant mucosa and progressive resorption of the supporting bone. The diagnostic criteria were as follows: (1) deepened implant probing depth and radiographic bone resorption compared to baseline, (2) probing depth ≥ 6 mm, (3) suppuration, (4) bleeding upon probing, and (5) bone resorption observed radiographically for ≥25% of the implant length.4 The follow-up period was calculated in months from the time of superstructure placement to the time of onset of peri-implantitis. For healthy implants that were not diagnosed with peri-implantitis, the observation period was calculated from the time of superstructure placement to the time of the patient's last visit.
Data collection
Information on age, sex, history of periodontal disease, smoking habits, and body mass index (BMI) were obtained from the records taken at the first visit. Information on systemic diseases (including the onset of the disease during implant maintenance), implant placement sites, and diagnosis of peri-implantitis were extracted from records collected during maintenance visits. This information included diseases that already existed during the initial examination and systemic diseases that developed after implant placement. In all patients, the duration from the date of superstructure placement to the patient's last visit was established as the period of observation, regardless of the presence of peri-implantitis. This study did not include measurements by individual examiners such as clinical parameters; therefore, intrarater reliability was not assessed. Patients were divided into 2 groups: (i) systemic disease group, those with a history of systemic diseases (even a single abnormal finding in their systemic condition); and (2) nonsystemic disease group, patients with no history of abnormal systemic disease findings. We did not examine the severity or management status of individual diseases because this study aimed to determine whether any systemic diseases affected the time of onset of peri-implantitis.
Statistical analysis
An independent statistician not involved in data acquisition performed all statistical analyses. Statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria, version 4.0.0)15 and were reviewed by a statistician (AH). It is customized with R Commander, a package of additional functions of R, for easy analysis. More precisely, EZR is a modified version of R commander (version 1.42) designed to add statistical functions used frequently used in biostatistics. To determine the test method for the statistical difference of each variable, normality was investigated for each data. The normality of the data distribution was examined using the Kolmogorov-Smirnov test. P ≥ .05 was assumed to be normal distribution, and because only BMI followed a normal distribution in both groups (systemic disease group, P = .3; nonsystemic disease group, P = .12), a t test for BMI and Mann-Whitney U test, a nonparametric method, for other continuous variables (age and mean maintenance duration; non-normal distribution) were used. A logistic regression model was used to measure the association between the predictor and outcome variables while controlling for confounders. To determine the relationships between the time of onset of peri-implantitis and the medical history of systemic diseases, the Kaplan-Meier analysis method and log-rank test were applied.
These tests were performed to describe the survival curves and to compare between the groups. In these tests, including the logistic analysis, the 95% confidence interval (CI) was used as an indicator of precision, and P < .05 was considered a significant difference.
All tests were performed on implant-based analysis to examine the incidence of peri-implantitis relative to the functional duration of a single implant, even in complex cases such as patients with multiple systemic diseases and/or multiple implants.
Results
Description of patients and implants
The collected personal information was highly reliable data that was strictly controlled by confidentiality. Furthermore, the information, including the medical history, consisted of the patient's declaration and the surgeon's confirmation and diagnosis; therefore, the validity of the data is high. This retrospective cohort study included 216 implants placed in 89 patients (34 men and 55 women) between 1998 and 2019. To verify the validity of the sample size, the effect size was set at 0.6, α = 0.05, and the power was 80%; the number of patients required was 84. Therefore, the present sample size (89 patients, 216 implants) could be statistically examined. The average patient age was 53.9 ± 11.8 years at the first visit, and the mean duration of maintenance was 7 years and 4 months. A total of 43 patients had medical histories and were assigned to the systemic disease group. The patient demographic statistics for both groups are summarized in Tables 1 and 2. Information on the number of implants and prosthetic devices for each group is presented in Table 3. Eighteen patients (20.2%) had a BMI ≥ 25, and 14 (15.7%) had a smoking habit or had smoked in the past. Fifty-two patients (58.4%) had a history of moderate to severe periodontitis (stage II-III, grade B), including 4 patients with generalized severe chronic periodontitis (stage III-IV, grade B) and 1 with aggressive periodontitis (stage IV, grade C). In the systemic disease group, 58.1% of patients had a single disease, and 41.9% had multiple diseases. Homogeneity was assumed using explanatory variables that can be expressed as means (age, BMI, mean maintenance duration). Only age showed a difference in expected values between the groups. Therefore, heterogeneity as an older age for the systemic disease group suggested the need to interpret the results carefully. Table 4 lists the systemic diseases presented by the patients in the systemic disease group. Among systemic diseases, hypertension, gynopathy, diabetes mellitus, malignancy tumor, and osteoporosis were common. The same number of patients had a comorbidity of hypertension and diabetes mellitus, hypertension and gynopathy, and malignant tumor and osteoporosis, respectively (2 patients, 4.7% each in the systemic disease group). There were 113 implant placements in the maxillae and 103 in the mandibular sites, with slightly more implants in the upper jaw than in the lower jaw. All implants had rough surfaces: Replace Select (Nobel Biocare, Zürich, Switzerland; n = 63), OsseoSpeed (Dentsply, Charlotte, NC; n = 49), Brånemark System Mk III (Nobel Biocare; n = 32), Nobel Replace Groovy (Nobel Biocare; n = 28), OSSEOTITE XP (Biomet, Warsaw, IN; n = 21), Straumann Bone Level (Straumann, Basel, Switzerland; n = 17), and Steri-Oss implant system (Nobel Biocare; n = 6). Those with TiUnite surfaces accounted for approximately 57% of the total.
Prevalence of peri-implantitis
The overall prevalence of peri-implantitis was 15.7% (34 of 216 implants) and 19.1% (17 of 89 patients) at the implant and patient levels, respectively. In the systemic disease group, the overall prevalence of peri-implantitis was 25.4% (29 of 114 implants) and 29.5% (13 of 44 patients), whereas in the nonsystemic disease group, it was 4.9% (5 of 102 implants) and 8.9% (4 of 45 patients) at the implant and patient levels, respectively. The prevalence of peri-implantitis was significantly higher in the systemic disease group at both levels. Logistic regression analysis revealed risk factors for peri-implantitis at the implant level. The explanatory variables were BMI (0 = <25, 1 = ≥25), smoking (nonsmoker or smoker), location (mandibular or maxillae), history of moderate or severe periodontitis, and the presence of systemic disease. For the development of peri-implantitis, statistically, significant differences were observed in systemic diseases (odds ratio [OR], 6.87) and history of moderate or greater periodontitis (OR, 3.64). However, there were no significant differences for the other independent variables (Table 5).
Analysis of time of onset of peri-implantitis
Kaplan-Meier tests were used to define the effect of systemic diseases on the onset of peri-implantitis. Figure 1 illustrates a significant difference in the time of onset of peri-implantitis at the implant level between patients with and without systemic diseases. Although both groups followed the same pattern until 50 months into the maintenance period, the prevalence of peri-implantitis in the systemic disease group progressively increased after 100 months in the maintenance period. Kaplan-Meier survival curves were divided into groups according to systemic disease (Figure 2). The target diseases were hypertension, gynopathy, diabetes mellitus, malignant tumors, and osteoporosis. In log-rank statistical tests, statistically significant differences were observed for hypertension, diabetes mellitus, and osteoporosis. The impact of diabetes mellitus was observed to be the greatest, followed by osteoporosis and hypertension, in the order illustrated in Figures 1 and 2.
Kaplan-Meier survival curves of cumulative implant success according to the presence of systemic disease.
Kaplan-Meier survival curves of cumulative implant success according to the presence of systemic disease.
(a–e) Kaplan-Meier survival curves are divided into groups according to systemic diseases.
(a–e) Kaplan-Meier survival curves are divided into groups according to systemic diseases.
Discussion
In the present study, logistic regression analysis revealed the medical history of systemic diseases and dental history of periodontitis as risk factors for peri-implantitis at the implant level. Furthermore, among systemic diseases, diabetes mellitus, hypertension, and osteoporosis had a significant impact on the onset of peri-implantitis.
A survey conducted by the Japanese government revealed that the most common age group for implant treatment was 65–69 years.16 The mean maintenance duration in the nonsystemic disease group was 7 years and 9 months, which was approximately 1 year longer than that of the systemic disease group. However, there was no significant difference. The observation period of interest ranged from 1 to 20 years. This long observation period permitted evaluation on any possible changes in the health status of the patients. The present cohort can be characterized as a group of patients with long-term maintenance at a university hospital setting. However, because this may be also considered as referral bias, caution should be exercised when applying the results to other clinics.
Our study found that the prevalence of peri-implantitis was 15.7% (34 of 216 implants) and 19.1% (17 of 89 patients) at the implant and patient levels, respectively, which is slightly higher than that reported in previous studies.17,18 These increased values may be attributed to the implant surface topography. Implants with TiUnite surfaces accounted for approximately 57% of the total implants. In a previous study, greater intrasurgical defect depths, defect widths, probing depths, and radiographic bone loss with TiUnite implants were observed.19
In addition to the history of periodontitis, smoking, poor oral hygiene, less keratinized tissue, and insertion in the maxillae have been identified as substantial risk indicators for peri-implant disease.17,20 In the present study, no significant differences were identified regarding smoking habits. In contrast, 25% of patients in the systemic disease group had a past or current smoking habit; the rate was about 6% in the nonsystemic disease group. These results suggest that smoking habits may be correlated with the development of systemic diseases. Previous studies have demonstrated that obesity contributes to complications of periodontal disease21 and maybe a risk factor for peri-implantitis.22 In the present study, the mean BMI indices were below 23 in both groups, and the difference was not significant. This suggests that the participants in both groups were within a healthy range. We did not find a significant association between implant position and the prevalence of peri-implantitis. The maxillae-to-mandibular ratio of the implant placement site was 4:3 in the systemic disease group and 1:1 in the nonsystemic disease group. In a previous review, a trauma in the maxillae and age-related inadequate blood supply in the mandible were risk factors for implant failure.23
Statistical analysis was performed on the number of implants rather than the number of affected individuals to examine the disease(s) (single or multiple) for individual implant functional periods. The reason for this approach was the small sample size of combined cases of multiple diseases, as shown in the results. However, nearly half of the patients in the systemic disease group were found to have multiple diseases; the association of multiple diseases in the same patient should be clarified in the future studies.
Although the time for the onset of peri-implantitis was reported to increase in the first 3–5 years of function,24,25 our results displayed onset at 6–10 years. This is due to the characteristics of the participants who were receiving continuous follow-up and maintained sufficient oral hygiene.
All but 1 of patients with diabetes included in the present study had type 2 diabetes; 1 patient had type 1 diabetes. The impact of diabetes was large among the systemic diseases considered in this study. Patients with diabetes are more likely to develop peri-implantitis within a relatively short period. The association between diabetes mellitus and peri-implantitis has been extensively explored in previous studies. A meta-analysis by Monje et al26 calculated that both the OR and risk ratio for peri-implantitis were statistically higher in hyperglycemia than in normoglycemia. Our results are inconsistent with those of previous studies.
A history of hypertension was also confirmed as a risk indicator for the onset of peri-implantitis. Hypertension is a common chronic disease in Japan. It has been reported that patients taking antihypertensive agents have worse clinical parameters around their implant(s) than those who do not take antihypertensive agents.27 Thus, the risk of hypertension and antihypertensives should be identified in future studies.
Osteoporosis demonstrated the third largest impact on peri-implantitis. A total of 15 implants demonstrated peri-implantitis in patients with osteoporosis, including 9 in oral bisphosphonate users and 6 in raloxifene hydrochloride users. Recently, peri-implant medication-related osteonecrosis of the jaw with critical symptoms has been reported.28 Considering that the number of patients with osteoporosis will further increase, maintaining good hygiene of the peri-implant tissue is critical for preventing serious implant-related diseases.
One of the limitations of the present study is the small sample size. To obtain detailed information on the influence of various diseases on the development of peri-implantitis in future studies, the number of individual cases for each disease must be increased to investigate the severity and progression of the disease, as well as the research on treatment and medication use. In addition, the number of medical visits and the social status should be investigated.
Because this study was performed on implant-based analysis, the effects of different placement sites and different prosthesis could be examined. However, because patients with multiple diseases accounted for about 40% of the systemic disease group, information on symptoms and medications, as well as multiple morbidities, need to be considered for a more accurately described health status. In future studies, patient-based risk analysis will be important to examine the interaction of systemic diseases within the model.
The mechanisms by which systemic diseases affect the pathophysiology of peri-implantitis have not yet been elucidated. However, the present study is significant in that it first examined whether there was a difference in the duration of disease onset by observing the long-term maintenance period. To support the hypothesis of this research, a variety of studies on the progression and severity of each disease, as well as physiologic and pathologic factors, will be needed in future.
Conclusion
A significant difference in the time to onset of peri-implantitis was confirmed between patients with systemic disease and healthy patients. Among systemic diseases, diabetes mellitus, osteoporosis, and hypertension had a significant impact on the onset of peri-implantitis.
Abbreviations
Acknowledgments
This work was supported by JSPS KAKENHI Grant JP19K19061 (Grant-in-Aid for Young Scientists) to A.H.
References
Note The authors declare no conflicts of interest.