Marginal Bone Loss Around Dental Implants and Its Association with Chronic Corticosteroid Use: A Prospective Cohort Study

Abstract

Background

Prolonged corticosteroid use leads to glucocorticoid-induced osteoporosis.

Objective

This study aimed to assess marginal bone loss around dental implants in patients with chronic corticosteroid use.

Methodology

This prospective cohort study included patients who required dental implant therapy on the posterior maxilla and mandible with a history of chronic corticosteroid use (12 months or longer). A control group was considered without corticosteroids to compare marginal bone loss (MBL). The implants were loaded with cemented restorative crowns 4 months after placement. Radiographs were obtained immediately after loading and 12 months later. The vertical distance between the shoulder of the implant and the highest point of the surrounding alveolar bone was defined as the initial marginal bone surface. The use of chronic corticosteroids (12 months or more) is the primary predictive factor. Marginal bone loss (MBL) was the study’s primary outcome, defined as the difference between the reference point and the new marginal bone level after 12 months. The study covariates included age, gender, implant sites, corticosteroid dose, and gingival type (thick or thin).

Results

In total, 154 subjects (75 cases and 79 controls) were included in our study. The mean age was 45.33 ± 10.44 and 43.37 ± 10.07 among the cases and controls, respectively (p-value = 0.237), and the mean corticosteroid dosage was 6.43 ± 2.50. Among the cases, 68% (n = 51) had bone type III, and 32% (n = 24) had bone type IV. Among the control group, 40.50% (n = 32), 58.22% (n = 46), and 1.26% (n = 1) had bone types IV, III, and II, respectively. We found a significant correlation between MBL and corticosteroid use in both the mandible (Pearson’s correlation = 0.921, p-value < 0.001) and maxilla (Pearson’s correlation = 0.871, p-value < 0.001).

Conclusion

The study results revealed a dose–response relationship between marginal bone loss and corticosteroid use, suggesting that prolonged corticosteroid use increases the risk of implant failure.

Introduction

Over the past 30 years, dental implants have become a routine and reliable treatment for complete and partial edentulism. The clinical success of dental implants depends on primary osseointegration. Osseointegration is defined as the direct structural and functional connection between the bone and the intraosseous implant surface. The successful osseointegration of dental implants relies on the proper functioning of the biological activities inherent in bone remodeling, including the dynamic processes of bone resorption by osteoclasts and new bone formation by osteoblasts. Insufficient osseointegration may result in micromobility of the implant, which is attributable to bone loss around the implant. [1–3]. Radiological assessment of bone quality around the implant is essential for evaluating long-term success in follow-up studies. Several criteria have been suggested to gauge the success of implants. A commonly employed criterion stipulates that bone loss exceeding 1 mm in the first year and more than 0.2 mm in subsequent years can indicate the initiation of implant treatment failure. Another criterion for assessing implant marginal bone resorption (MBL) suggests that a maximum bone resorption of 2 mm during 5 years of functional activity is acceptable. Consequently, any factor influencing bone regeneration can compromise successful osseointegration, leading to premature implant loss and soft tissue dehiscence. [4–6].

Glucocorticoids are potent immunosuppressive medications used to treat a wide range of systemic conditions. Glucocorticoids are crucial therapeutic agents for managing the symptoms of chronic inflammatory diseases. Approximately 1% of the global adult population has undergone chronic glucocorticoid treatment, defined as use for at least three consecutive months [7–9]. One of the most prevalent and debilitating side effects associated with prolonged and high-dose glucocorticoid use is its effect on the skeletal system, leading to bone loss that can adversely affect implant osseointegration. Glucocorticoid-induced osteoporosis (GIOP) is a significant contributor to secondary osteoporosis, affecting 30–50% of chronic glucocorticoid users. Glucocorticoids induce a rapid decline in bone density, leading to elevated bone resorption within 3–6 months of use [10–12]. Subsequently, there is a gradual reduction in bone formation and turnover due to a decrease in osteoblasts, accompanied by the apoptosis of both osteoblasts and osteocytes. This condition is associated with decreased bone density and poses a significant risk of implant instability. Furthermore, chronic use of glucocorticoids has been cited as a contraindication for facial implant placement [13–15]. Due to the crucial role of osseointegration in the long-term success of implant treatment and the limited studies on bone loss rates in individuals using corticosteroids, our study aimed to compare the rate of bone loss and assess marginal bone resorption around implants in individuals who use corticosteroids.

This study was designed to answer the following question: Does the use of glucocorticoids increase MBL around dental implants? It was assumed that glucocorticoids did not enhance the MBL around dental implants. Therefore, this study investigated the amount of MBL in patients using glucocorticoids and compared it to those not using glucocorticoids.

Methods and Materials

Study Design/Sample

This prospective cohort study was designed by the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) statement.

This prospective cohort study included patients who required dental implant therapy with a history of chronic corticosteroid use. The sample was derived from a population of patients who were introduced to the Oral and Maxillofacial Surgery Department of Shahid Beheshti University of Medical Sciences between September 1, 2020, and December 30, 2023. The Medical Ethics Committee of Shahid Beheshti University of Medical Sciences approved this study.) IR.SBMU.DRC.REC.1401.084(

The eligibility criteria were as follows: (1) age between 18 and 65 years, (2) use of corticosteroids for 12 months or more, and (3) being a candidate for dental implant therapy in the posterior maxilla and/or mandible. The exclusion criteria were as follows: (1) use of antiresorptive drugs or immunosuppressants, (2) smoking, (3) need for bone augmentation (graft or sinus augmentation), (4) chronic diseases that affect bone remodeling, and (5) candidates for fresh socket implant placement. (6) Patients with diabetes (n = 7) failed to return for follow-up or refused study enrollment.

Data Collection Methods

After proper anesthesia (lidocaine 2% with epinephrine 1/80000, Daroopakhsh Co., Iran), a full-thickness flap was prepared to initiate implant placement. Placement was performed according to the standard protocol for the Straumann Standard Plus (SP) Implant system. Oral prophylactic antibiotic therapy (amoxicillin, 2 g) was administered to all patients. The implants were loaded 4 months after placement. All restorative crowns were cemented. Two examiners measured MBL from radiographs obtained immediately after loading and 12 months later using a periapical long-cone digital X-ray machine (iRay D3; Dexcowin, Seoul, South Korea) at 60 kV and 7 mA DC for 0.20 s. The reproducibility of the radiographs was confirmed using individual bite blocks attached to a beam-guiding device (XCP; Rinn, Elgin, IL, USA). The bone surface was measured on the mesial and distal surfaces of the implants. The shoulder of the implant was considered the reference point. The vertical distance between the shoulder of the implant and the highest point of the surrounding alveolar bone was defined as the initial marginal bone surface. Marginal bone loss was defined as the difference between the reference point and the new marginal bone level after 12 months. If the amount of bone loss was unequal on the mesial and distal sides of the dental implant, the mean MBL was recorded. (Fig. 1).

Fig. 1.

The measurements of MBL on the mesial and distal sides of the implants

The gingival biotype (GB) was categorized as thin and thick based on the transparency of a double-ended periodontal probe (Rolle, Switzerland; Ø 0.5 and 0.75 mm; DBS12 prototype, Deppeler SA). The periodontal probe prototype has two heads of different thicknesses. Transparency was assessed by probing the sulcus at the mid-buccal aspect of the implants, with the gingival margin in view. The gingival biotype was thick if the thick probe ending was not obvious through the tissue.[16].

Variables

The use of corticosteroids was the study’s predictive factor. MBL was the primary outcome. Age, gender, implant sites, dose of corticosteroids, and gingival type (thick or thin) were the study covariates.

Bias: To reduce the risk of bias, two examiners measured MBL.

Sample Size Calculation

The equation. N = (r + 1)(Z_α/2 + Z_1-β)² s²/rd² was used to calculate the sample size^.Z_α was considered 1.96 at a 95% significance level for type I errors; for type II errors, Z_1-β was 0.84. According to a previous study, the accepted hypothesis for the clinical difference was 1.2 standard deviations. [17] The calculation involved 90 patients, with a 95% confidence level and a 80% study power. [17].

Statistical Analysis

Statistical analyses were performed using IBM SPSS V.21.0. (IBM Corp., Armonk, NY, USA). An independent t-test was used to compare MBL and age between the osteoporotic and non-osteoporotic groups. Pearson’s correlation test was used to evaluate any correlation between MBL and T-scores. A general linear regression model was used to predict MBL with corticosteroid dose. Statistical significance was set at a p-value of less than 0.05. Inter-examiner reliability(Kappa test) was used to determine the similarity between examiners.

Results

One hundred and seventy-five subjects were studied. Twenty-one subjects were excluded from the study because they failed to attend follow-up visits. Finally, 154 patients (75 patients and 79 controls) were included in this study. All patients were followed up for 1 year after implant loading. Among the cases, 80% (n=60) were female, compared with 30% (n=30) in the control group (p-value<0.001). The mean age was 45.33±10.44 and 41.65±9.89 among the cases and controls, respectively (p-value = 0.026). Further information regarding the cases and controls is presented in Table 1.

Table 1.

Comparing covariates between the case and control groups

Variables	Case group	Control group	p-value
Age (years)	45.33 ± 10.44	43.37 ± 10.07	0.24*
Site	35 maxilla, 40 mandible	40 maxilla,39 mandible	0.63 **
Gingival thickness	59Thick, 16 thin	63 thick,16 thin	0.51**
Bone type	47 D3,28 D4	1 D2,46 D3,32 D4	0.46**
Plaque index score	38.88 ± 3.66	38.24 ± 4.30	0.32*

*Independent T test **Chi-square test

The mean of MBL was 1.15±0.43 mm in the case group and 0.91±0.18 mm in the control group. The data analysis revealed a significant difference in the mean of MBL between the case and control groups. (p<0.001) ( Table 2)

Table 2.

The assessment of MBL between the case and control groups

Outcome	Case group	Control group	Independent T test
MBL (mm)	1.15 ± 0.43	0.91 ± 0.18	p < 0.001

The mean corticosteroid dosage was 6.43 ± 2.50 (Table 3). Sixty-eight percent of subjects (n = 51) had bone type III, and 32 (n = 24) had bone type IV. However, among the control group, 43% (n = 34), 14% (n = 11), and 43% (n = 34) had bone types IV, III, and II, respectively. There was no significant difference in the frequency of bone types between the two groups.(p = 0.67).

Table 3.

The comparison of corticosteroid dose between the maxilla and mandible

		Maxilla	Mandible	Both	p-value
GCS (cont.)		6.35 with 2.52 SD	6.5 with 2.51 SD	6.43 with 2.50 SD	0.807
GCS	2.5 mg	5 (14.29%)	4 (10%)	9 (12%)	0.729
	5.0 mg	14 (40%)	19 (47.50%)	33 (44%)
	7.5 mg	8 (22.86%)	6 (15%)	14 (18.67%)
	10 mg	8 (22.86%)	11 (27.50%)	19 (25.33%)

Based on the results of Pearson’s correlation test (Table 4), we found a significant dose–response association between MBL and corticosteroid use (Pearson’s correlation = 0.882, p-value < 0.001). This is also evident in the second model, which has a group-wise classification of GCS use. The linear model coefficients were 0.05 (p < 0.001), 0.13 (p = 0.001), 0.37 (p < 0.001), and 0.71 (p < 0.001) for those using 2.5 mg, 5.0 mg, 7.5 mg, and 10 mg of GCS, respectively. The General linear model indicated a 1 mg increase in the corticosteroid dose and a 0.15 mm increase in MBL in 1 year after loadings, which could account for 77.8% of the population receiving corticosteroids. (p < 0.001, R² = 0.778) The kappa agreement coefficient was 0.577, indicating a moderate agreement between the examiners. (Figs. 2 and 3).

Table 4.

The correlation between corticosteroid dosage and MBL based on the implant site

		Maxilla (n = 35)		Mandible (n = 40)		Both (n = 75)
		Coefficient	p-value	coefficient	p-value	coefficient	p-value
GCS (cont)	PC*	0.871	< 0.001	0.921	< 0.001	0.882	< 0.001
	LR**	0.161	< 0.001	0.138	< 0.001	0.149	< 0.001
Dosage	0	Ref	–	Ref	–	Ref	–
	2.5 mg	0.023	0.012	0.028	0.008	0.05	< 0.001
	5.0 mg	0.16	0.007	0.12	0.031	0.13	0.001
	7.5 mg	0.39	< 0.001	0.35	< 0.001	0.37	< 0.001
	10 mg	0.81	< 0.001	0.63	< 0.001	0.71	< 0.001

*Pearson’s correlation; **Linear regression

Fig. 2.

A The dental implant was loaded at site #30. B Marginal bone loss was observed at the mesial and distal sides in the patient from the corticosteroid group at the 12-month follow-up time

Fig. 3.

A The dental implant was loaded at site #30. B The bone was stable at both the mesial and distal sides in the patient from the control group at the 12-month follow-up time

Discussion

Corticosteroids can affect bone metabolism and increase blood glucose levels. [18] Patients who received chronic corticosteroids were studied to determine the effect of GCS on marginal bone loss around dental implants during the first year of follow-up after loading. The results of our study indicate a dose–response relationship between MBL and corticosteroid use. Therefore, long-term corticosteroid use can increase the risk of implant failure.

To the best of our knowledge, no prior study has been conducted to evaluate the association between marginal bone loss and steroid medication. However, two studies investigated the relationship between implant failure and steroid use.

Our findings are based on a recent study that investigated the role of systemic steroids in MBL.[19]. Their findings indicated a correlation between steroid medications and heightened implant failure risk, irrespective of whether steroids were categorized as systemic steroid medications, hormone replacement therapy, or oral contraceptives or solely as hormone replacement therapy and oral contraceptives. Furthermore, the results were independent of factors such as smoking, diabetes, use of other systemic medications, or systemic immunological conditions.

Conversely, a 2019 study by Carr et al. [20] investigated the association between implant failure and medication use, reporting that corticosteroid use was associated with a decreased risk of implant failure and had a protective effect (hazard ratio = 0.82). Such results should be interpreted cautiously, as both the univariate and adjusted multivariate analyses indicate a significant difference. Based on their univariate logistic regression model, the hazard ratio for corticosteroid use and implant failure was 0.72, with a p-value of less than 0.001. However, after adjustment, the hazard ratio changed to 0.82, and the p-value increased significantly to 0.04. This suggests that several confounding factors have not been considered, such as the duration of corticosteroid use and the type of condition being treated (chronic versus acute). This is of great importance because acute corticosteroid use is far more prevalent with chronic GCS medication. GCSs are prescribed to lower local inflammation in numerous implant therapy candidates, even during the initial days after implant placement.

Petsinis et al. assessed the impact of glucocorticosteroids on the osseointegration and survival of dental implants placed without bone grafting. Thirty-one patients were included in this study. The authors concluded that the use of glucocorticosteroids, which are typically indicated for systemic diseases, does not significantly impact the osseointegration or 3-year survival of dental implants. Implants were placed using a traditional two-stage surgical protocol without bone grafting. Therefore, it should not be considered a contraindication for dental implants.

Previous research has demonstrated a link between steroids, bone resorption, and periodontitis. Corticosteroids stimulate the differentiation of osteoclasts, resulting in increased bone resorption and the activation of inflammatory mediators in the periodontium. [2, 21, 22]. Long-term administration of glucocorticoids has a significant effect on bone metabolism. Medication decreases the production of new osteoblast precursors, causes early apoptosis of mature osteoblasts, and increases the lifespan of osteoclasts. This results in systemic loss of bone mass, secondary osteoporosis, and impaired bone healing.[23].

Numerous clinical studies have noted associations between corticosteroids and changes in the development and progression of periodontal disease, including impaired bone metabolism and an increased prevalence of caries, plaque, calculus, and fungal infections [6–8, 24]. Additional clinical investigations have elaborated on the mechanisms by which glucocorticoids may affect bone physiology [11, 25, 26]. One study revealed that short-term glucocorticoid treatment significantly reduced the levels of type I procollagen N-terminal peptide and osteocalcin, which are markers of bone formation. Moreover, there was a decline in IGF2, a regulator of osteoblast function, accompanied by the upregulation of RANK-L expression and a decrease in osteoprotegerin expression in osteoblastic and stromal cells. Osteoprotegerin acts as a decoy receptor, counteracting the RANK ligand —a cytokine receptor that activates osteoclasts and promotes bone resorption. These alterations imply that glucocorticoids play a role in modulating osteocyte differentiation [27–30]. Discontinuation of glucocorticoid intake during a 2-week recovery period reversed these changes, suggesting that elevated glucocorticoid concentrations may reduce bone formation rates, osteoblast and osteocyte numbers, and activity while extending osteoclast lifespan [30–32]. An animal study demonstrated that prolonged glucocorticoid use is associated with significantly reduced alveolar bone thickness and volume, diminished bone formation, and lower mineral apposition rate. This finding aligns with a previous study, which demonstrated that a 21 day glucocorticoid treatment in rats resulted in decreased bone mineral density in the femur. These data indicate that glucocorticoids influence the inflammatory pathways that regulate bone formation and resorption [27–29, 31].

Our study indicated that the MBL in patients who received corticosteroids was higher than that in the control group 1 year after loadings (1.15 ± 0.43, 0.91 ± 0.18 retrospectively). The difference in MBL between the two groups was 0.24 mm, which may not have had a clinical impact on the prognosis of dental implant treatments. A long follow-up is necessary to achieve meaningful clinical outcomes.

Conclusion

The results of our prospective cohort study indicate that patients taking systemic steroids are at a significantly greater risk of marginal bone loss. This finding suggests that steroidal medications may affect long-term implant success and increase the risk of implant failure. However, further randomized studies, such as clinical trials, should be conducted to determine whether there is a causal or associative relationship between implant failure and marginal bone loss with long-term corticosteroid use.

Author Contributions

RT and SC designed the study and assisted in writing the paper. ShM and HK helped with patient recruitment, collected, and compiled the raw data. RT and ECH reviewed the entire work, generated the results, and drafted the manuscript.

Funding

No funding source.

Declarations

Conflict of interests

The authors declare that the research was conducted without any commercial or financial relationships that could be construed as a potential conflict of interest.

Ethical Approval

The medical ethics committee of Shahid Beheshti University of Medical Sciences approved this study. (IR.SBMU.DRC.REC.1401.084).

Consent to Participate

All participants were provided with a participant information sheet, and written consent was obtained from each one.

Consent for Publication

Authorship consent form has been signed by all the authors.