On the Association between Dental Implants, Osteoporosis and Bone Modulating Therapy

Abstract

Background: Osteoporosis affects bone metabolism and may result in fragility fractures. Medications include bone modulating therapy (BMT), which come with the risk of osteonecrosis of the jaws (ONJ). The literature is contradictory about the impact of osteoporosis and/or BMT on dental implant success and the incidence of ONJ. Purpose: The aim of the present study was to assess the effect of osteoporosis and BMT on early implant failure (EIF) and ONJ incidence (primary outcome parameters) following dental implant placement. Materials and Methods: Retrospective, cohort study based on dental records. Implant-supported prostheses (ISP) were delivered by experienced oral and maxillofacial surgeons and prosthodontists. Inclusion criteria: diagnosis of osteoporosis, delivery of ISP, consecutive individuals, and available data. Confounding variables included—age, gender, smoking status, BMT, bone grafting, and implant length/diameter. Results: There were a total of 72 osteoporosis patients, 46 using and 26 not using BMT. There were a total of 279 inserted implants, 154 in those using and 124 in those not using BMT. Univariate analysis of factors that may affect EIF showed no statistically significant differences between non-osteoporotic, osteoporotic using BMT, and osteoporotic patients not using BMT regarding EIF and ONJ incidence following dental implant delivery. Multivariate model using logistic regression demonstrated one factor associated with increased risk for EIF—bone augmentation. No cases of ONJ were reported up to one year post ISP delivery in any group. Conclusions: Within the limitations of the present study, it can be concluded that installing dental implants in osteoporotic patients, treated or not with BMT, is a safe procedure with EIF comparable to non-osteoporotic patients. The short-term risk for ONJ following dental implant insertion in osteoporotic (even when using BMT) patients is negligible.

1. Introduction

The main current treatment alternative for partial/complete edentulism is implant-supported prostheses (ISP) [1]. Low bone volume is still a major issue challenging the option of implant dentistry in a substantial number of individuals [2,3]. In cases with compromised bone volume, the success rates are lower [4]. In such circumstances, guided bone regeneration, (GBR) using barrier membranes, is often used to increase bone volume [5,6,7]. Various studies suggested the application of stem cells both for therapeutic and preventive purposes in order to lower the percentage of failed osseointegrated implants. Stem cells also enhance GBR [8,9,10,11,12,13,14,15]. Still, even after the introduction of these methods, between 3 and 10% of implants fail (Moy et al., 2005, Alsaadi et al., 2007) [16,17]. Systemic factors, such as osteoporosis, may lead to early implant failures (EIF up to 12 months post-loading) [17].

Osteoporosis is a disease affecting the bone causing fragility fractures. The worldwide prevalence of osteoporosis is high affecting 21.2% of women older than 50 years and 6.3% of men [18,19]. Risk factors include postmenopausal state in women, age, smoking, drugs (mainly glucocorticoids) and a variety of systemic diseases.

Treatment of osteoporosis is long term and may last a lifetime. Treatment aims to reduce the risk of fragility fractures. Various treatment alternatives exist, including fall risk reduction and supplementary vitamin D and calcium, but the main and effective treatment are drugs developed for osteoporosisis. Antiresorptive drugs affect the osteoclast cells in various mechanisms leading to fracture rate reduction are among the most popular osteoporosis treatments. Two main classes are used: bisphosphonates—analogues of pyrophosphate characterized by a P-C-P bond and denosumab—antibody against receptor activator of nuclear factor ligand. Besides osteoporosis, these drugs are also used in other bone diseases, such as Paget and malignant lesion of the bone [20,21,22,23].

Osteonecrosis of the jaw (ONJ) related to osteoporotic treatment has been reported in post-marketing observation of bisphosphonates [5] and in the clinical trials of RANK ligand antagonists and also romosozumab. There have been no reports on osteonecrosis of the jaw related to treatments with SERM or teriparatide.

These drugs are among the bone modulating therapies (BMT) encountering the risk of osteonecrosis of the jaws (ONJ). An ONJ diagnosis consists of (1) exposed necrotic bone; (2) history > 8 weeks; and (3) no history of radiation therapy or metastatic disease to the jaw [18,19].

The aim of the present study was to assess the incidence of EIF and ONJ in osteoporotic vs. non-osteoporotic patients undergoing implant-supported oral rehabilitation. The secondary aim was to assess differences between osteoporotic patients using vs. not using BMT.

2. Materials and Methods

The present study is a retrospective cohort study. All treatments were performed by experienced oral and maxillofacial surgeons.

The study protocol was approved by the ethics committee of the Rabin Medical Center, Campus Beilinson, Israel (0674-19rmc). The present script complies with the STROBE guidelines [24]. Dental records of all individuals included were extracted and manually screened twice by 2 examiners (DM and LC).

2.1. Patient Population

2.1.1. Inclusion Criteria

Diagnosis of osteoporosis; consecutive patients who had received a dental implant between 01/2013 and 12/2018; available documentation; minimum follow-up—12 months following ISP delivery; and healthy periodontal status.

2.1.2. Exclusion Criteria

History of head and neck cancer; history of radiation therapy to the face; immunocompromised patients; immunosuppressant medications; heavy smokers (more than 10 cigarettes per day); and untreated periodontal disease.

2.2. Data Collection

(1)

Age;

(2)

Gender;

(3)

Physical status according to American Society for Anesthesiology (ASA) [23];

(4)

Implant dimensions (length, diameter (mm);

(5)

bone grafting (yes/no);

(6)

ONJ incidence (yes/no; primary outcome variable);

(7)

EIF incidence—lack of osseointegration up to 12 months after prosthesis delivery and occlusal loading (yes/no; primary outcome variable).

2.3. Data Analysis

The data were analyzed using SPSS software version 25. Descriptive statistics were performed using means and standard deviations for the continuous variables and frequencies for the discrete variables. Univariate correlations were performed using the Chi-square (${\chi }^2$) test. Tests between independent samples were done using the Mann–Whitney test. Significance was considered for a p-value lower than 5%.

3. Results

3.1. Study Population

The cohort (n = 792) comprised 37.2% males and 62.8% females. A total of 58% were aged ≤65, 33.4% were 66–79.9 years, and the remaining 8.6% were ≥80. Smokers comprised 5.1% of the cohort.

At patient level—40.8% implants were placed in pristine and 59.2% in augmented bone. At least one implant failed in 14.4% of the patients.

Most of cohort (90.6%) did not have osteoporosis (EIF 14.6%) while 9.4% had osteoporosis (5.2% received BMT (EIF 15.5%); 4.2% did not take medications (EIF 7.4%)).

No cases of ONJ were reported up to one year post-implant-supported prosthesis delivery.

All data were likewise measured at the implant level (n = 2971). The age distribution- was 57.7% ≤65, 34% 66–79.9 years and the remaining 8.3% were ≥80. Smokers’ prevalence was 6.6%. Augmented implants comprised 61.2% and 38.8% were placed in pristine bone. EIF accounted for 3.8% of the implants. Most implants (90.9%) were placed in patients without osteoporosis (EIF 3.9%), 5.8% in patients with osteoporosis (EIF 4.5%) and used BMT, while the remaining 3.3% (EIF 1.6%) were in those who did not take medication despite being diagnosed with osteoporosis. For a complete description see

Table 1. Demographic and clinical characteristics of the cohort at implant level (n = 2971).

	M	SD	N	%
Demographic characteristics
Age group (years)
≤65			1715	57.7
66–79.9			1009	34
≥80			247	8.3
Smoking			195	6.6
Clinical Characteristics
Augmentation
Pristine			1153	38.8
Augmented			1818	61.2
Implant length (mm)	11.38	1.60
Implant diameter (mm)	3.84	0.31
EIF			114	3.8
Osteoporosis			279	9.4
No osteoporosis			2692	90.6
Osteoporosis with BMT			155	5.2
Osteoporosis without BMT			124	4.2

for implant level descriptive statistics and

Table 2. Demographics and baseline clinical characteristics of the cohort at patient level (n = 792).

	N	%
Demographic characteristics
Gender
○ Male	294	37.2
○ Female	498	62.8
Age Group (years)
○ ≤65	459	58
○ 66–79.9	265	33.4
○ ≥80	68	8.6
Smoking	40	5.1
Clinical Characteristics
Augmentation
○ Pristine	323	40.8
○ Augmented	469	59.2
EIF	114	14.4
Osteoporosis	72	9.1
No osteoporosis	720	90.9
Osteoporosis with BMT	46	5.8
Osteoporosis without BMT	26	3.3

for patient level descriptive statistics.

3.1.1. Univariate Analysis

Univariate tests were conducted to assess the effect of each independent variable on EIF. At the implant level (

Table 3. Univariate analysis at implant level.

		Success		Failure
Variable	Group	N (%)	$\mathit{M}\pm \mathit{S}\mathit{D}$	N (%)	$\mathit{M}\pm \mathit{S}\mathit{D}$	p-Value
Age groups (Years)	≤65	1647 (57.6%)		68 (59.6%)		0.16
	66–79.9	967 (33.8%		42 (36.8%)
	≥80	243 (8.5%)		4 (3.5%)
Smoking	No	2667 (93.3%)		109 (95.6%)		0.34
	Yes	190 (6.7%)		5 (4.4%)
Diabetes mellitus	No	2426 (84.9%)		100 (87.7%)		0.416
	Yes	431 (15.1%)		14 (12.3%)
Augmentation	Pristine	1128 (39.5%)		25 (21.9%)		<0.001
	Augmented	1729 (60.5%)		89 (78.1%)
Implant length (mm)			$11.37\pm 1.60$		$11.57\pm 1.52$	0.41
Implant diameter (mm)			$3.87\pm 0.31$		$3.84\pm 0.34$	0.65
Osteoporosis	No	2587 (90.5%)		105 (92.1%)		0.39
	Osteoporosis with BMT	148 (5.2%)		7 (6.1%)
	Osteoporosis without BMT	122 (4.3%)		2 (1.8%)

), a significant relation was found between augmentation and failure (${\chi }^2\left(1\right)=14.22,p<0.001$). In pristine bone, the fraction of osseointegrated implants (39.5%) was significantly higher than the fraction of implants with EIF (21.9%). In augmented bone, the fraction of osseointegrated implants (60.5%) was lower than the fraction of implants with EIF (78.1%). Similar results were obtained at the patient level (

Table 4. Univariate tests at the patient level.

		Success		EIF
Variable	Group	N (%)	$\mathit{M}\pm \mathit{S}\mathit{D}$	N (%)	$\mathit{M}\pm \mathit{S}\mathit{D}$	p-Value
Gender	Male	256 (37.8%)		38 (33.3%)		0.36
	Female	422 (62.2%)		76 (66.7%)
Age groups	≤65	391 (57.7%)		68 (59.6%)		0.10
	66–79.9	223 (32.9%)		42 (36.8%)
	≥80	64 (9.4%)		4 (3.5%)
Smoking	No	643 (94.8%)		109 (95.6%)		0.73
	Yes	35 (5.2%)		5 (4.4%)
DM	No	574 (84.6%)		597 (88%)		0.371
	Yes	104 (15.4%)		81 (12%)
Augmentation	Pristine	298 (44%)		25 (21.9%)		<0.001
	Augmented	380 (56%)		89 (78.1%)
Osteoporosis	No	615 (90.7%)		105 (92.1%)		0.26
	Osteoporosis with BMT	38 (5.6%)		8 (7.0%)
	Osteoporosis without BMT	25 (3.7%)		1 (0.9%)

). For full model tests, see Table 3 and Table 4. All other tested parameters had no statistically significant differences.

3.1.2. Multivariate Analysis

A Logistic regression model at the implant level showed that the independent variables significantly predict failure (χ²(16) = 99.69, p < 0.001), while explaining about 12% of total variance in failure. The model is well-fit to the data (χ²(8) = 4.38, p = 0.82) while classifying about 96.1% of the total observations. It was found that the odds of failure for implants installed in an augmented bone was 2.14 times higher vs. implants in pristine bone (OR = 2.14, p = 0.002). All other tested parameters had no statistically significant differences.

Descriptive Tables about osteoporosis treatment (as shown in

Table 5. Type of medications used for treatment of osteoporotic patient.

Type of Medication	Number of Patients	Number of Implants
Alendronate	16	45
Zoledronic	13	49
Denasomab	10	31
PTH analog	6	28
Raloxifane	1	2
not take ARM	26	124

and

Table 6. Duration of medication intake of osteoporotic patients.

Duration of Medication Intake (Years)	Number of Patients
0	26
1	6
2	8
3	9
4	5
5	11
More than 5	7

).

4. Discussion

In the present study, the rate of EIF in three different groups were assessed: (1) non osteoporotic patients, (2) osteoporotic patients with BMT and (3) osteoporotic patients with no treatment. No statistically significant differences were found between the three groups. Bone augmentation was the only significant risk factor for EIF regardless of the study population.

EIF in osteoporotic patients is a contentious issue. Several contradicting studies exist. Alsaadi et al. [17] claimed that EIF in osteoporotic patients is higher than non-osteoporotic ones. Kasai et al. [25], emphasizing the impact of oral bisphosphonates on implant failure, found a survival rate of 86% vs. 95% in bisphosphonate users [25]. Yap et al., found that EIF odds for osteoporotic women with BMT is 2.69 (95% confidence interval, 1.49–4.86) [26].

Other studies are consistent with the present reported results. Jeffcoat et al., found no increase in the EIF odds ratio in osteoporotic patients with BMT [27]. Madrid and Sanz et al., in their systemic review found that dental implant placement in osteoporotic patients with BMT is safe and has no impact on dental implant survival in the short-term [28].

In the present study, no case of ONJ was reported up to one year post-implant-supported prosthesis delivery, demonstrating the safety of placing implants in osteoporotic patients even when using BMT. Many patients receiving medications associated with ONJ have other comorbidities or conditions, which are likely to increase or may contribute to higher failure rates and the incidence of ONJ, for example, diabetes mellitus and smoking. In a recent review and meta-analysis on the impact of smoking on implant failure, they found a significant increase in the relative risk (RR) of implant failure in patients that smoked >20 cigarettes per day compared with non-smokers; RR = 4 at patient level and RR = 2.45 at implant level [29].

In the present study, we excluded heavy smokers, and among smokers no correlation was found between smokers and EIF with similar rates of smokers among non-failure and failure both at the patient and implant level. Moraschini et al., found in a systematic review that the number of implant failures does not differ between diabetic and non-diabetic subjects [30].

In the present study, at patient level, no correlation was found between DM and EIF with similar rates of DM among osseointegrated (15.4%) and EIF groups (12%).χ² = 0.800, p = 0.371. Similarly, at the implant level (15.1%) vs. (12.3) χ² = 0.662, p = 0.416 respectively.

A recent review [31] demonstrated that infection seems to have a major role in the pathogenesis of ONJ. Although there is no conclusive evidence for the infection hypothesis yet, available data have shown a robust association between local infection and ONJ development [31].

“As evidence that systemic inflammation can promote ONJ, rheumatoid arthritis patients suffer more from ONJ with more serious and intense clinical and radiological appearance of ONJ”. As an additional support for the infection hypothesis, removal of infected ligature form mice induced periodontitis and ONJ demonstrating reduced inflammation and cessation of ONJ progression.

Consequently, implant placement was always performed in clean conditions in healed sites. Implants were not inserted in fresh extraction sites. This may explain the reason for zero ONJ cases following implant placement in the present study.

To the best of our knowledge, this is the first study comparing EIF in osteoporotic patients treated with BMT to osteoporotic patients without BMT and to non-osteoporotic patients. The results of the present study are in the spirit of recent studies claiming the safety of dental implant installation in osteoporotic patients in general and in those receiving BMT in particular [32,33,34,35], with success rates of osseointegrated dental implants comparable with rest of the study cohort.

Several new approaches for BMT and osseointegration were suggested. One study examined the effects of raloxifene during bone formation around the dental implant in the ovariectomy-induced osteoporotic rat maxilla. Female Sprague-Dawley rats were divided into three groups (n = 18 each): sham-operated (control), ovariectomized (OVX), and ovariectomized and raloxifene-administered (RAL). Eight weeks after ovariectomy, both upper first molars were extracted, and implants were placed 4 weeks post-extraction. It was concluded that Raloxifene administration enhanced the osteogenic genes and protein expression in the bone around the implant [36].

Another study aimed to determine the influence of hydrophilic titanium surfaces on gene expression and bone formation during the osseointegration process in an osteoporotic model. Their results supported that hydrophilic surfaces in situations of osteoporosis may provide additional benefits in the early stages of osseointegration [37].

A recent review [38] suggests that topographical modifications, improvement in hydrophilicity and the development of controlled-release drug-loading systems to improve cellular adhesion, proliferation and differentiation. Surface modifications, along with drug coating, undoubtedly demonstrate better osseointegration, especially in challenged degenerative conditions, such as osteoporosis, osteoarthritis and osteogenesis imperfecta. Loading substances on modified titanium surfaces may be achieved by mechanisms, such as direct coating, adsorption and incorporating in biodegradable polymers.

5. Conclusions

It can be concluded that within the limitations of the present study, installing dental implants in osteoporotic patients, treated or not with BMT, is a safe procedure with EIF comparable to non-osteoporotic patients. The incidence of ONJ is negligible and should not discourage practitioners from placing dental implants in osteoporotic patients.