Markers in blood and saliva for prediction of orthodontically induced inflammatory root resorption: a retrospective case controlled-study

Orthodontically induced inflammatory root resorption (OIIRR) is not an uncommon iatrogenic consequence of orthodontic treatment [1]. The inflammatory component of OIIRR comes from force application inducing local biological changes that are essential for tooth movement and also the cause of root resorption process [2]. While in most patients the degree of root resorption is minor and barely noticeable, moderate to severe resorption can be exhibited in several teeth, especially in the aesthetic zone [3]. In their 1988 paper, Lavender and Malmgren quantified root resorption with an index in which grade 1 is an irregular root contour; grade 2 is apical root resorption of less than 2 mm; grade 3 is apical root resorption greater than 2 mm to one third of the original root length; and grade 4 is root resorption exceeding one third of the original root length [4].

There are limitations in finding clinical markers (factors) of root resorption during active orthodontic treatment. Although the anatomical risk factors have been investigated extensively [5–7], they account for only 20–30% of the expected variation in severity of OIIRR, indicating that they are not definitive predictors [5]. Some studies indicated that progress radiographs, taken at 6 to 12 months into treatment, could detect early OIIRR [7]. However, there is difficulty in using conventional radiography in detecting early levels of root demineralisation [8] because it takes approximately 30–60% of mineral content loss in order to visualise changes radiographically [8, 9]. Hence, conventional radiography should be used as an aid for comparison of treatment outcome as opposed to clinical markers for monitoring early changes in OIIRR.

Orthodontic movement is a micro-trauma to the periodontal ligament associated with a cascade of local periodontal inflammatory cycle [10]. Secondary to orthodontic movement, several local anti-resorptive and pro-resorptive cytokines have been observed [11–13]. Pro-resorptive cytokines such as interleukin (IL)-1 family (i.e. IL-1β), IL-6, IL-7, IL-8 and TNF-α. IL-1β which directly induce osteoclastogenesis and promote osteoclast function [14, 15]. Similarly, IL-6 acts synergistically with IL-1 and TNF-α on osteoclastogenesis and promotes osteoclast function [14, 15]. IL-7 works indirectly through the induction of TNF-α, an important augmenter of receptor activator of nuclear factor kappa-B ligand (RANKL) mediated osteoclastogenesis [16] while IL-8 enhances RANKL expression [17], both increase osteoclast generation and activates osteoclasts. On the other hand, the anti-resorptive cytokine such as IL-4 and interferon-gamma (IFN-γ) suppresses osteoclastogenesis and T-cell of RANKL-induced osteoclastogenesis, respectively [18, 19]. Granulocyte macrophage colony-stimulating factor (GM-CSF) is another anti-resorptive cytokine that inhibits bone resorption along with IL-4, IL-10, IL-13, IL-18 and IFN-γ [20].

There are several systemic biomarkers that could be related to OIIRR. For instance, the relationship between OIIRR and thyroxine hormone is still controversial; however it was found that the level of alkaline phosphatase rises significantly in patients with OIIRR [21]. Calcitonin has been proven to reduce orthodontic movement and thus may also affect root resorption [22]. On the other hand, an animal study showed that administration of osteocalcin could accelerate movement and subsequently might induce severe OIIRR [23]. Parathyroid hormone (PTH) acts synchronisingly with calcitonin to control the level of calcium in bone. Increased level of PTH is associated with skeletal calcium deficiency, faster tooth movement and may be increased OIIRR [24, 25]. There is also some evidence indicating that T-lymphocytes are contributor to the inflammatory response secondary to orthodontic force, OIIRR and bone remodelling [26]. Brezniak et al. stated that consumption of alcohol during orthodontic treatment is related to high risk of OIIRR due to vitamin D hydroxylation in the liver [27]. Similarly, Collins and Sinclair showed that the number of mononuclear osteoclasts boosted following administration of vitamin D in the periodontal ligament of rats; this was associated with faster tooth movement and may be related to higher OIIRR [28]. The elevated level of IgE in patients with an allergy has association, but not statistically significant, with an increased risk of OIIRR [29].

The primary aim of this pilot study is to identify key local, in saliva, and systemic, in blood, biological markers that are associated with OIIRR in patients with moderate to severe root resorption. Secondly, this paper aims to use these findings to develop a predictive model for identifying patients susceptible to extensive OIIRR in order to allow for primary prevention of this serious complication. The null hypothesis of this project stated that there are no correlations between biological markers found in blood or saliva and the severity of OIIRR.

This pilot study identified that cytokine expression between different diagnostic groups is more sensitive using saliva than in blood samples. In comparison to gingival crevicular fluid (GCF), saliva as a source to identify cytokines is considered remote from the root inflammatory zone, which means least overlapping of cytokines expressions [31, 32]. Thus, sampling saliva, as a chair-side screening test, is both affordable and easy compared to collecting blood or GCF. However, this study also showed that taking saliva samples might provide some indication of systemic reaction to potential inflammation.

The concept that pro-inflammatory cytokines are resorptive factors and anti-inflammatory cytokines are anti-resorptive factors is an oversimplified view and one that is changing. In our study, although the significant increase in IL-7 and a decrease in IL-4 could be linked specifically to root inflammatory phase, the increase in anti-inflammatory cytokines IL-10, IL-12p70 and IFN-γ was also significant. The anti-inflammatory cytokines may be upregulated to stimulate a constant state of remodelling in the bone akin to the body’s defence mechanism. Cytokines mediate bone damage by driving the differentiation and activation of the bone-resorbing cell, the osteoclast [33], as well as playing an essential role in immune cell development and immune-regulation. More recent evidence suggested that the effect of cytokine levels’ alterations is not a singular effect, that is, cytokines that promote inflammation can also have anti-inflammatory and immunosuppressive actions [34]. O’Shea et al. suggested that the combination of cytokines exerts variable effects at different times during autoimmune disease processes. IFN-γ, for example, has been shown to have multiple roles in autoimmunity; one as a mediator in autoimmune disease and another as having a protective role [34]. Thus, based on the findings of our study, it was difficult to precisely define the link between cytokines in OIIRR based on the small sample size.

A lack of significant difference in cytokine expression between genders is similar to the findings of Linge and Linge [3]. They noted that there was no connection between sex and severity of root resorption; therefore, if elevated levels of certain cytokines are linked to root resorption, gender may not play a role in affecting cytokine expression.

Osteocalcin and P1NP seemed to be the only blood factors that showed a significant difference between CG and RRG. However, gender appeared to have a more significant effect on osteocalcin and P1NP levels than diagnosis. It is important to isolate the effects of age and gender on blood osteocalcin and P1NP levels since bone turnover can be upregulated as a natural process of human growth. Previous research has shown that the distribution of osteocalcin at different sites of remodelling may be age- and gender-related changes [35]. Another study has revealed pubertal increases in P1NP level in both sexes [36]. Therefore, blood osteocalcin and P1NP may not be useful clinical markers to predict the susceptibility of patients to OIIRR as often these patients are receiving treatment at stages of natural growth and development.

This study applied computations to estimate radiographical magnification differences (RMD) between pre- and post-treatment OPGs taken by different operators and machinery [30]. Nevertheless, it was limited in its capacity to account for tilting of teeth especially maxillary and mandibular anterior teeth which are the most commonly affected teeth with extensive OIIRR [4, 37]. Another factor that should be considered in this study is the radiographical judgemental errors (RGE) in equating roots’ length from the OPGs. RGE is the result of film-related factors and/or radiographical conspicuity. Film-related factors include variation in radiographical brightness and contrast between pre- and post-treatment radiographs [38]. While conspicuity is a radiographic term which describes the background ‘noise’ caused by adjacent anatomical structures causing difficulty in assessing the examination area [8, 38], conspicuity arose when root apices were outside the narrow focal trough [39], incompletely developed roots or when the cervical vertebrae and other anatomical structures obscured the root apices. To overcome RGE and IGE, digital subtraction radiography (DSR) is recommended in prospective researches. DSR is a technique that requires the subtraction of one radiographic image from another, provided these two images have identical and reproducible projections of the same anatomic region [8, 9, 38]. For reproducibility purpose of DSR, bite registration with polyvinylsiloxane silicon impression material, invested into the bite block of a beam-guiding device, is used [8, 38]. After a series of radiographs are obtained at different time intervals, stable horizontal and vertical landmarks can be selected and eliminated in order to ensure measurement accuracy of the periapical regions between radiographs [38].

It is worth mentioning that many eligible participants deterred from consenting to participate in this pilot study and a repeated measurement of the saliva and blood sample were not undertaken to decrease the risk of bias. The main reason was that this study required large volumes of saliva and blood sample to be sent for laboratory testing. Recently, the use of microfluidic biosensors (MFB) as part of point-of-care diagnostic portable devices (PoC) was recommended in the monitoring of cytokine expression in a clinical setting [40]. The detection times of MFB are relatively faster than conventional laboratory testing, taking only from a few seconds to tens of seconds to process the sample. Another advantage of MFB is the small sample volume that could be used, ranging from micro-litres to nano-litres [41]. Prospect adoption of PoC devices in orthodontic practices could be used for accurate detection of cytokine expression and level changes. Thus, it would be possible to ‘red-flag’ patients that are more at risk of extensive OIIRR earlier. Subsequently, patients at risk can receive adequate modification in their orthodontic treatment, perhaps even pausing treatment which would allow the regenerative properties of cementum to repair the root surface, thereby minimising irreversible loss of root structure [6, 7].

Finally, as the relationship between cytokines and OIIRR in our study was based on the small sample size, future prospective studies with larger samples may yield interesting results, in particular the pre- and post-orthodontic level of cytokine.

This study found that saliva might be more efficient in measuring changes in cytokine expression than blood. Osteocalcin and P1NP appeared to be the only blood factors that showed a significant difference between groups.

The weak association between treatment variables and patient characteristics with the degree of root resorption suggests that increased expression of pro-inflammatory, as well as anti-inflammatory cytokines, may be determinants in the development of moderate to severe OIIRR. However, the expression of cytokines may also be affected by any other potential inflammation in the body.

Further investigations could utilise digital subtraction radiography techniques and microfluidic biosensors, in a larger group of patients, to provide a more accurate and detailed insight into the mechanisms of OIIRR. Cytokine expression at different time intervals, during and after orthodontic treatment, may also give more information about the inflammatory nature of OIIRR.