The Effect of Hypothyroidism on Bone Loss at Dental Implants

Hypothyroidism (HT) is a condition characterized by abnormally reduced activity of the thyroid gland¹  and, in iodine-sufficient areas, is most commonly associated with autoimmune disease.^2,3  In addition to iodine deficiency, HT may be caused by surgical resection, radiation therapy, pregnancy,¹  and side effects of medications such as amiodarone, interferon, and lithium.⁴  Common manifestations of HT include weight gain, fatigue, alteration in cognition, and infertility.⁵  Analysis of 1999–2002 National Health and Nutrition Examination Survey (NHANES) data suggests that HT affects 3.7% of the population,⁶  although the prevalence can vary depending on patient demographics and how HT is defined.

Mosekilde and colleagues^7,8  found that there was a decrease in bone turnover in subjects with hypothyroidism. Other investigators have suggested that patients with HT experience a depression in the bone remodeling cycle with a corresponding decrease in bone turnover.^9–11  In 1 study, HT patients exhibited a mineralization lag time of 48 days vs only 25 days in the non-HT control group; HT patients also experienced a decrease in the amount of bone loss during episodes of bone resorption. Those processes resulted in a significant positive balance between resorption and formation per remodeling cycle of +16 μm,¹²  inferring that, overall, more bone was being deposited than being resorbed at a given time. In an orthodontic treatment study, it was found that thyroid supplementation, when prescribed to treat HT or provided as post-thyroidectomy therapy, was associated with a decrease in treatment time secondary to a corresponding increase in the rate of tooth movement through bone, which also was consistent with an environment associated with an elevated bone turnover rate.¹³  Collectively, those findings implied that patients with HT experience alterations in bone metabolism that differentially favor deposition, relative to resorption.

A dental implant is an artificial tooth root, commonly fabricated from titanium alloy, that is surgically placed into alveolar bone to support a replacement tooth or dental prosthesis, such as a fixed or removal denture.¹⁴  Initially, dental implants are mechanically stabilized into alveolar bone, until osseointegration occurs, beginning at about 4 weeks after placement.¹⁵  The process of osseointegration, or functional ankylosis, occurs when the implant surface forms a close approximation to bone, which involves osteoconduction, de novo bone formation, and bone remodeling.¹⁶  Once osseointegrated, dental implants have a high success rate, generally greater than 96% after 10 years.¹⁷  Dental implants are manufactured in known, standardized dimensions such as length, diameter, shape, and thread pitch, which, collectively, can facilitate measurements of adjacent bone, serve as reference markers, and allow correction for radiographic distortion, unlike natural teeth.¹⁸ 

Consequently, the purpose of this retrospective clinical study is to measure the clinical effect of altered alveolar bone metabolism in patients with HT compared with patients without HT in a dental implant model system. Because there is evidence of a depressed bone resorption phase during remodeling in HT patients, we hypothesized that patients with HT might exhibit less bone loss around implants.

This study was reviewed and approved by a Health Sciences Institutional Review Board (STUDY00002276), and the manuscript was prepared according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines as adapted for this retrospective clinical study.¹⁹  All patients 21 years or older receiving implant placement at a university postgraduate periodontics specialty clinic from 2000 to 2017 were considered for evaluation. Patients without radiographic documentation of both implant placement and subsequent implant examination at a later date were excluded. For measurement of millimeter bone loss, implants were excluded if the entire implant image could not be visualized or otherwise was not diagnostic. Radiographic analysis was accomplished using either direct measurement for film images or via MiPACS Dental Enterprise Viewer (DEV) software 3.0 (Medicor Imaging, Charlotte, NC) for digital images. Conventional periapical or panoramic radiographs were used for analysis. Standardization of bone loss measurements, and compensation for distortion or other confounding factors, was accomplished using ratios of apparent bone loss to implant length, corrected by actual (label) implant length. To minimize potential bias, all measurements of each implant were performed by the same examiner, who was blinded to the patients' thyroid status. Medication, medical, and dental histories were obtained following review of patient records.

A sequential time series pair of radiographs were reviewed for each implant: the initial assessment (T₁) was obtained at the time of implant placement, where apparent implant length was measured from the most crestal to the most apical aspects of the implant. A subsequent radiograph with the greatest available elapsed time between placement and implant re-evaluation (T₂) was used to again obtain implant length and bone loss measurements, in a manner identical to that described for T₁. Elapsed time between T₁ and T₂ was recorded for each implant. Bone loss at T₁ and T₂ was recorded in millimeters by measuring the distance from the most crestal aspect of the implant to the apical extent of the bone defect. Correction for radiographic distortion was performed by multiplying the measured bone loss by the ratio of the actual implant length to the radiographically measured apparent implant length.

To account for the effect of systemic conditions or disease on implant-associated bone loss in the patient population, the analysis was repeated after excluding implants from patients with a history of smoking and diabetes and repeated again after excluding implants from patients that smoked, had diabetes, were taking systemic steroids, hormone replacement therapy, or hormone medications, or those with autoimmune diseases other than HT.

To be classified as HT in this study, patients were required to have a clinical diagnosis of HT established by the patient's physician, as well as concurrent use of thyroid hormone supplementation (eg, levothyroxine, liothyronine). Patients having a diagnosis or history of thyroid cancer, thyroid surgery, hyperthyroid disease, or any other thyroid abnormality other than clinically evident HT requiring hormonal supplementation were excluded. Those criteria were used to be consistent with previous retrospective studies that have used identical surrogate diagnostic criteria and to allow comparison with the current study results.^20,21 

For all analyses, Levene's test for equality of variances was initially performed in conjunction with independent sample t tests (using equal or unequal variance as appropriate). Subsequently, significance based on unequal variances was set at 0.05 for all tests. To determine whether any differences in patient oral hygiene were present that might influence the outcome, patient plaque scores were obtained using either the Ramfjord index sampling technique²²  or via whole-mouth plaque detection, as available, and mean values (percent of teeth with plaque accumulation) were calculated. Statistical differences in oral hygiene among HT patients and control subjects, as well as any differences in implant evaluation time, were determined as described above. All analyses were performed using IBM SPSS Statistics v25, and the statistical methodology was reviewed by an independent statistician.

Patient record review resulted in initial identification of 1480 implants from 635 patients. After applying the exclusion criteria as noted above, a final pool of 1271 implants were available for analysis. As described in the Table, peri-implant bone loss from patients with HT was significantly lower than for those without HT (t₁₂₅₂ = −3.42; P < .001; M = 0.53 and 1.63 mm/yr, respectively). Among the subset of nonsmokers and nondiabetics, that relationship persisted (t₈₆₁ = −3.46; P < .001; M = 0.37 and 1.71 mm/yr, respectively), with HT patients experiencing less bone loss around dental implants compared with patients without HT.

When patients taking systemic steroids, hormone replacement therapy, hormone medications, or autoimmune diseases were additionally excluded from the analysis, that relationship continued (t₈₅₂ = −3.11; P < .002, M = 0.42 and 1.34, respectively), with HT patients having less bone loss around dental implants than patients without HT.

For all comparisons, no statistically significant differences in patient plaque control or mean radiographic evaluation time were found between HT and control patients (P > .05).

Our data suggest that patients with HT experienced significantly less bone loss around dental implants. Those findings persisted after eliminating diabetics and smokers, as well as those who used systemic steroids, hormone replacement therapy, hormone medications, or had autoimmune diseases. Those findings also were independent of the potentially confounding effects of patient oral hygiene and implant service time.²³  Although a limited number of reports exist describing implant placement in patients with HT, we believe this is the first study investigating the association of HT on crestal alveolar bone, as well as the first report quantifying the extent of alveolar bone loss at dental implants in patients with HT.

HT patients appear to have a lower bone turnover rate^7,8  because of decreased resorption during bone remodeling, which results in an increase in bone mass.^11,12  Thyroid hormone receptors are present in human bone²⁴  and may act on bone cells either directly via specific nuclear receptors or indirectly via increasing the secretion of growth hormone and insulin-like growth factor.²⁴  Consequently, patients with HT generally exhibit higher than normal bone density, whereas subjects with hyperthyroidism, characterized by lower levels of thyroid-stimulating hormone (TSH), appear to experience more bone loss and have a higher fracture incidence.^10,25 

Animal models of HT also have demonstrated alterations in bone metabolism,^26,27  possibly through a mechanism by which thyroid hormone has direct or indirect effects on osteoblasts and osteoclasts. Rats given T₃ and T₄ experience an influx of osteoclasts in long bones, which directly stimulate bone resorption.^28,29  Although the mechanism has not been completely elucidated, it has been proposed that T₃ might act either directly on osteoclasts or that the process of osteoclastogenesis—followed by bone resorption—is a secondary response to the direct actions of T₃ on osteoblasts, osteocytes, and other bone marrow cell types.³⁰  This potential feedback system may explain why HT patients taking hormone supplementation experience less bone loss around implants, as demonstrated in this study, compared with non-HT control subjects.

Most of the existing literature describing bone effects in HT have studied long bones. Although extrapolation to the oral cavity should be done with caution, bone matrix composition between the two have been shown to be quite similar with regard to cell type and lamellar structure.³¹  However, differences do exist in the rate of remodeling, which has been estimated to be up to 10-fold higher in alveolar bone compared long bones.^31–34  As a result, use of an alveolar bone model using implants of standardized dimensions might provide an alternative and accessible mechanism to further investigate the effects of HT on bone metabolism and remodeling.

Because our data are derived from a retrospective analysis of patient records, study limitations include variation in the amount of time patients were taking thyroid hormone supplementation, and the use of a convenience sample that included a 17-year period, over which time different dental implant systems were used and were surgically placed by multiple dental providers with varying expertise levels. It also is unknown whether there would be outcome differences if treatment with TH was initiated before implant placement compared with TH use after implant integration. Ideally, a prospective study that distinguished medication dose, duration, and thyroid hormone levels without other confounding factors is indicated to support a potential association between decreased bone loss at dental implants in hypothyroid patients and to elucidate the mechanism. Nevertheless, it is noteworthy that the results from this study were analogous to results from those human and animal studies discussed above.^{11,12,24,26–30} 

Although HT is associated with bone mineralization abnormalities,¹¹  the current study suggests that HT patients might not experience an increased risk of crestal alveolar bone loss after dental implant treatment, despite the presence of bone mineralization and metabolic abnormalities. As a result, dental implants do not appear to be contraindicated in this patient population.

Abbreviation

HT:

hypothyroidism

This study was supported in part by the William M. Feagans Endowed Chair Research Fund and the Department of Periodontics and Endodontics, University at Buffalo, School of Dental Medicine. The above funding sources had no role in the administrative or scientific conduct of the study. Consulting/independent statistician: Elaine L. Davis, PhD, University at Buffalo, School of Dental Medicine, Department of Oral Diagnostic Sciences, Buffalo NY14214.

The authors declare no conflicts of interest.