PERIODONTAL DISEASE IN PATIENTS WITH SYSTEMIC LUPUS ERYTHEMATOSUS: CARDIOVASCULAR RISK AND SIDE EFFECTS OF CORTICOTHERAPY INVOLVING ORAL HEALTH

Systemic lupus erythematosus (SLE) and periodontal disease (PD) are two chronic infl ammatory diseases that share common pathogenetic mechanisms. Therefore, it is thought that a correlation between the two pathologies might exist. PD itself, when not associated with SLE, does not cause major cardiovascular events, but it contributes to the development of atherosclerosis, which is the major cause of death among patients with SLE. This article aims to summarize the available literature concerning the association between these two disorders (PD prevalence in patients with SLE, correlations between periodontal damage and SLEDAI activity score), to present the main mechanisms by which the two pathologies affect the cardiovascular system and to evaluate the impact of SLE corticosteroid therapy on periodontal tissue.


INTRODUCTION
Systemic lupus erythematosus (SLE) is a multisystemic autoimmune disease, characterized by a chronic infl ammatory process. The impairment targets a range of tissues and organs, including the skin, joints, serous membranes and the nervous system. The etiology of the disease is complex and not completely understood, but it is admitted that genetic and environmental factors are involved. Disease prevalence varies between 15 and 50 cases in 100.000 inhabitants and it is more common in black people. Literature suggets a predilection for SLE in young women, the female:male ratio lying between 5:1 and 9:1. SLE is characterized by a B cell hyperactivity, resulting in the production of IgG and immune com-plexes. Certain antibodies (auto-antibodies) cause cellular damage by a type II hypersensitivity reaction or by participating in the formation of immune complexes. These will develop deposits in tissues and organs, causing tissue damage. The evolution of the disease alternates periods of fl ares and remission. (1) Periodontal disease (PD) (Fig. 1, 2) is a chronic infl ammatory oral disease that affects the support tissue around the teeth. PD's etiology is likely microbial, involving germs grouped in microbial complexes according to virulence. Contributing factors are tartar, morphological changes of teeth and dental arches, and systemic factors (systemic impairments, corticosteroid therapy, cytostatic medication, antiepileptic drugs etc.). PD's onset is infl uenced by the presence and direct action of pathogenic agents, but the evolution of PD is strictly dependent on the host immune response. The disease progresses gradually from mild reversible gingivitis to the following: damage of the deep marginal periodontal tissue, destruction of the alveolar bone and loss of contact between the tooth and the support tissue. PD prevalence in the general population lies between 10-15% for medium and severe forms. (2)(3)(4)(5) LINKS BETWEEN PD AND SLE: PATHOGENIC HYPOTHESES AND EPIDEMIOLOGICAL DATA PD and SLE are two disorders with multifactorial etiology, sharing pathogenic mechanisms such as high serum levels of proinfl ammatory cytokines and β2-glycoprotein-I dependent anticardiolipin antibodies. Due to pathogenic similarities, there may be a relationship between the two disorders. It is also known that patients with SLE are highly susceptible to infections due to the following: (a) alteration of cellular immune response (T cell lymphopenia, impaired LTc activity, reduced NK cells level and diminished NK cell function), (b) impairment of humoral immune response (anti-receptor Fcγ antibodies, antibodies against cytoplasmic components of neutrophils, hypo-γ-globulinemia), (c) defi ciencies of phagocytosis (failure to generate super oxide radicals), (d) cytokine defects and other immune abnormalities, (e) certain elements specifi c for the disease (immunosuppressive medication, corticosteroid therapy, biological therapy, transient or permanent splenic dysfunction). Therefore, predisposition to infection of patients with SLE may infl uence the body's defense mechanisms against periodontal microbial fl ora. (6,7) In a study on 15 patients with SLE, aged between 22 and 53 years, Ramos Sales et al attempted to verify the existence of correlations between the extent of periodontal damage and the SLE activity score (SLEDAI). Regarding PD prevalence in the study group, periodontal damage, measured by the depth of periodontal pockets (PPD) ≥ 3 mm, was found in 7 patients of 15, with a resulting prevalence of 46.7%. Small differences were identifi ed between the average SLEDAI score in patients that suffer from both SLE and PD (3.42) and those with healthy periodontal tissue (3.5). When analyzing individual components of the SLEDAI score, only C3 serum levels indicates a statistically signifi cant relationship with the frequency of visible plaque areas. Thus, the increase in the amount of plaque, is correlated with reduced C3 serum levels, which can be explained by the consumption of C3 in the course of the systemic infl ammatory process. (8) Ramos Sales et al state that their data regarding the prevalence of PD in patients with SLE are consistent with other studies. They found a prevalence of 46.7% in the study group (n = 15). Other studies also reported an increased prevalence of periodontal  Female patient with SLE, with total maxillary edentulism and partial mandibular edentulism (Kennedy class I mod 2) restored with removable acrylic dentures: maxillary complete denture and mandibular partial denture, engaged with dental wire clasps. The patient presents plaque deposits and untreated complicated multiple dental caries. Supragingival 8%. In the studies we analyzed, that are currently available in the literature, we observed a higher prevalence of PD in patients with SLE (between 46% and 94%) compared to the general population. PD is a pathological complex that brings together several types of changes in the marginal periodontal tissue. Due to the multitude of pathologic entities, there is still no consensus regarding the diagnostic criteria for PD, therefore its prevalence in the population varies widely, between 10-15% for medium and severe forms, while exceeding 50% for mild forms (5,(8)(9)(10)(11) In a recent study, Al-Mutairi et al evaluates the periodontal status in a group of 25 female patients with SLE, in comparison to a control group comprising 50 women not suffering from SLE. Average percentage of PPD ≥ 5 mm was higher among patients with SLE compared to the control group (11.3% versus 5.6%). Clinical attachment loss (CAL) is also revealed as being modifi ed with more than 2 mm in 75.8% of teeth evaluated in SLE patients and in 71.9% of the teeth in the control group. The results state that, even if the periodontal damage is more advanced in patients with SLE, the difference cannot be considered statistically signifi cant. Results with statistical signifi cance were reported in SLE patients with recent fl ares (< 1 year), compared to the stable ones, and also in patients with active disease (arthritis) compared to those without fl ares. Patients with recent fl ares (< 1 year) showed a smaller periodontal damage (measured by the percentage of periodontal pockets and lower CAL) compared to the stable group (> 1 year). Patients with lupus arthritis (active disease) showed greater periodontal damage compared to those without fl ares, regarding the percentage of PPD ≥ 5 mm. (7) Data from the available literature, concerning the mechanisms by which PD infl uences SLE are few and inconclusive. Furthermore, there is a single study regarding the relationship between SLE serum markers and severity of PD. PD prevalence in patients with SLE is higher compared to the general population, but there is no consensus regarding a correlation between periodontal damage and the SLEDAI score. In most studies, however, the lack of a correlation between periodontal parameters and SLEDAI score is suggested. SLE pharmacological therapy reduces the biological infl ammatory syndrome, a key element in periodontal destruction in PD. Also, increasing the dose in rebound periods explains the diminished periodontal damage, compared to the stable patients (> 1 year). (7)

CARDIOVASCULAR RISK IN PD AND SLE C-reactive protein
C-Reactive protein (CRP)is a non-specifi c infl ammatory marker, that is associated with increased risk for cardiovascular disease (CVD).
As a defense against the microbial attack towards the periodontal tissue, a non-specifi c systemic response appears. This is initially characterized by local cytokine production in the marginal periodontal tissue, and it later becomes systemic by stimulating hepatic synthesis of CRP and subsequently increasing its serum level. CRP works by forming deposits in the injured vessel, by binding to the affected cells and by binding the complement. Hence, it activates the phagocytes, that generate nitric oxide and contribute to the development of atheroma plaques. Also, the association between CRP and atherosclerotic plaques is explained by promoting the development of foam cells, by means of mediating the binding of LDL to macrophages. (12,13) Current studies show a direct correlation between CRP serum levels and PD, in patients with PD that do not associate other systemic infl ammatory diseases. CRP serum levels vary in direct proportion with the degree of periodontal damage and its levels are reduced 3 months after the correctly applied periodontal therapy. Studies focused on the PD-CRP-CVD correlation show an increased value of CRP in PD and CVD, when not associated in the same patient, and very high levels of CRP in patients associating these two pathologies (Genco et al reveals CRP levels of 8.7 mg/L in patients associating PD and CVD, in comparison to 1,14 mg/L in healthy individuals, and a decrease of 65% in its level 3 months after the beginning of treatment). (12)(13)(14) Serum levels of CRP are generally correlated with infl ammatory response during fl ares in illnesses, while literature indicates that SLE may be the exception. It was noted that most patients with SLE show only modestly increased levels of CRP, particularly in comparison to patients with rheumatoid arthritis (RA). (15) The less important CRP elevation could be explained by increased clearance or decreased production of CRP. Bell et al, cited by Ferreira, shows that the presence of auto-antibodies anti-CRP in patients with SLE stands for the increased clearance of this infl ammatory marker, while Vugushin et al states that the rate of serum clearance of CRP in patients with SLE is similar to that of healthy individuals. (6,15) Current studies use high-sensitivity CRP (HsCRP) to detect chronic micro-infl amation. Normal values of HsCRP lie between 0-3 mg/L, while high levels represent a risk factor for impaired cardiovascular function, being associated with the development and progression of atherosclerotic process, considered major cause of death in SLE. In a study conducted in 2012, Gheit et al compared a group of 45 women with SLE to 30 healthy women, none of the 75 patients having a history of cardiovascular events, clinical or electrocardiographic signs of cardiovascular impairment.The results indicated that HsCRP levelsin SLE patients were signifi cantly higher (4.84 ± 3.91 mg/L) compared to the ones of the healthy patients (1.74 ± 0.61 mg/L). (16) In the aforementioned study conducted on 15 patients with SLE, Sales Ramos et al have identifi ed a slightly increased average CRP serum level in patients associating SLE and PD, compared to those with no signs of PD (3.94 mg/L compared with 3.35 mg/L). Although not statistically signifi cant, it is mentioned that there is a positive correlation between serum levels of CRP and the number of periodontal pockets with PPD ≥ 4 mm. This observation highlights that periodontal damage advances with the increase of CRP serum levels. (7) Regarding the change in acute phase reactants during the independent evolution of PD and SLE, literature data provide various intervals for the CRP level. It is though unanimously accepted that its values are at least modestly higher than normal. There is only one study that assesses the relationship between the extent of periodontal damage and SLE markers (CRP). The authors of the study state that patients with both SLE and PD show higher levels of CRP compared with SLE patients without periodontal damage, but the results are not statistically signifi cant. Thus, we believe that further studies are needed to analyze correlations between periodontal damage and serum markers of SLE. However, taking into consideration the separate evolution of the two diseases, the role of CRP as active atherosclerotic agent in these patients indicates the presence of an increased CV risk.

Lipid profi le
Current studies analyzing mortality in SLE show that it is 3-5 times higher than in the general population and has as major cause CV impairment. A bimodal pattern of causes of death in SLE has been described: (a) an early peak attributed to those cases where death occurs within one year of diagnosis, being caused by active lupus and its consequences and (b) a late peak, regarding those cases where death occurs more than 5 years after diagnosis, mainly due to atherosclerosis. (6) Dyslipidemia is common in SLE and it may be a consequence of the disease itself or of the applied treatment. SLE directly affects the lipoprotein metabolism, resulting in elevated levels of triglycerides (TG) and VLDL and reduced levels of HDL and apolipoprotein A1 (apoA1). On the other hand, dyslipoproteinemia in patients with SLE undergoing treatment with glucocorticoids is characterized by elevated levels of VLDL, TG and total cholesterol. Calderon-Ticona et al show that a cause of dyslipedemia may be represented by the presence of auto-antibodies against proteins involved in lipoprotein metabolism. (6,17) Patients with SLE may have altered HDL function. Normally, HDL plays an anti-infl ammatory role by preventing the formation of oxidised LDL (ox-L-DL) and foam cells that would otherwise lead to the development of atheroma plaques in the vasculature. Paraoxonase-1 (PON1) is an anti-oxidant component of the HDL that prevents lipoprotein oxidation and removes ox-LDL. PON activity is inversely proportional to the serum level of anti-apoA1 antibodies. Thus, Batuca et al, cited by Abdulaziz et al, show high levels of anti-HDL antibodies, anti-apoA1 antibodies in SLE patients compared to healthy individuals. Tripi et al also show that in SLE, PON1 activity is altered, hence its reduced level is correlated with CVD and cerebrovascular events. (6,18,19) The atherogenic lipid panel consists of: (a) low HDL-cholesterol, (b) increased total cholesterol, (c) increased TG and LDL-cholesterol. Svenungsson et al determined the prevalence of CV risk factors in a study which compared three groups, each made up of 26 patients: with SLE and CVD, with SLE only and 26 healthy individuals. Therefore, it was shown that dyslipidemia, characterized by increased TG levels and low HDL levels, is common among SLE patients with CVD compared to the other two groups that both had similar lipid profi les, within normal li-mits. On the other hand, LDL values did not differ signifi cantly between the three groups. (20) The changes in the lipid profi le are not due to lifestyle. Biological infl ammatory syndrome with a periodontal starting point appears due to episodes of bacteremia and dissemination of endotoxins. In the available literature, PD is associated with the presence of endotoxins. The studies conducted so far indicate that patients with PD have a higher CV risk compared to individuals with periodontal integrity, and also reveal the presence of pathogenic periodontal agents in the early atheroma plaques. (21) Losche et al, cited by Thombre, demonstrates a relationship between periodontal changes and hyperlipidemia, by linking higher plasma TG levels and lower HDL-cholesterol in patients with periodontal damage, compared to healthy subjects. Thus, is it stood for the role of PD as a risk factor for hyperlipidemia. (21,22) Analyzing serum parameters before and after periodontal treatment in a group of 50 patients with PD, Thombre et al showed a positive correlation between periodontal damage and levels of total cholesterol, LDL, VLDL, TG and HDL and a direct correlation between periodontal health and HDL. (21) Cutler et al states that there is a tight correlation between the extent of periodontal damage, plasma lipid concentrations and the presence of anti-Porphyromonas gingivalis antibodies. This is explained by the fact that the high levels of TG may modulate the production of PMN IL-1b, as a consequence of the stimulation induced by P.gingivalis. Morrison et al noted that CRP, total cholesterol and fi brinogen are intermediary factors that correlate PD with a high CV risk. (23,24) Atherosclerosis represents the major cause of death among patiens with SLE and PD. Prospective studies are needed in order to clarify whether PD contributes to additional increase in the CV risk in patients with both SLE and PD.

PHARMACOLOGIC TREATMENT OF SLE: SIDE EFFECTS INVOLVING ORAL HEALTH
Treatment of patients with SLE varies depending on disease severity and on the presence or absence of complications. Classes of drugs used in less severe cases include NSAIDs, glucocorticoids and antimalarials (hydroxychloroquine). Glucocorticoids are administered initially in high-dose, followed by a progressive decrease to the minimum effective dose.
In patients with vital organ damage, in those who do not respond to glucocorticoids or in those cases where glucocorticoids can not be reduced below the maximum acceptable dose for chronic treatment, immunosuppressants such as methotrexate, azathioprine cyclophosphamide, or mycophenolate mofetil will be associated. (7) Impact of corticotherapy on the periodontal tissue Besides the already known general effects, systemic glucocorticoid therapy causes oral manifestations such as as oropharyngeal candidosis, changes in trabecular system of the upper and lower jaw and damage in the marginal periodontal tissue.
Beeraka et al have investigated the effects of long-term glucocorticoid therapy on oral health, assessing clinical and radiological changes. The evaluated periodontal parameters were plaque index (PI), gingival index (GI), CAL and PPD. The only statistically signifi cant differences were recorded for PPD and CAL, showing unfavorable effect of glucocorticoids, represented by apical migration of CAL and increased PPD. (25) Prolonged glucocorticoid therapy has plain consequences, such as an obvious decrease in calcium absorption, increased bone loss and high incidence of fractures. Bone loss is carried out in two phases: (a) the initial fast stage, during the fi rst months, representing 10-15% of the loss, (b) slow phase, representing 2-5%, annually. These losses are dependent on dose and duration of therapy, and occur faster in the trabecular bone, compared to the cortical bone. Doses of Prednisone ≥ 7.5 mg/day cause important bone loss and double the risk of fracture in treatments lasting longer than 6 months. (25) Loss of dental elements is dependent on many factors. Regarding corticotherapy, the major cause of tooth loss is the low bone mineral density (BMD), which, in association with occlusal masticatory forces, determines mobilization and avulsion. Klemetti et al states that individuals with increased BMD keep a larger number of teeth, with greater number of periodontal pockets, compared to the ones with low BMD, that is associated with horizontal bone atrophy, gingival retraction and tooth loss. Komerik et al have revealed that under corticotherapy, patients present a smaller number of teeth and decreased mandibular BMD compared to those not submitted to corticotherapy, but without statistical signifi cance. (25)(26)(27) CONCLUSIONS SLE and PD are two conditions that share pathogenetic similarities and that are characterized by altered infl ammatory response. Therefore, it is possible for these two diseases to infl uence each other regarding the clinical outcome. The susceptibility to infections, specifi c for SLE, may impair the antimicrobial oral defense against periodontal pathogenic agents and may therefore, determine the development of PD. This hypothesis explains the higher prevalence of PD among patients with SLE, compared to the general population. Regarding the relationship between SLEDAI activity score and the extent of periodontal damage, studies show an inverse correlation between the two, possibly due to the consumption of C3 component, as a result of the biological infl ammatory syndrome determined by PD. Patients who associate SLE and PD seem to have a more severe periodontal damage (higher percentage of tooth surfaces with periodontal pockets and loss of CAL) in comparison to those with PD, but without SLE.
Cardiovascular risk in patients with SLE is explained by the atherosclerotic process, developed due to chronic infl ammation. The dyslipidemia found in patients with SLE, characterized by low HDL levels, increased TG and LDL levels, is involved in the process of atherogenesis. The association of PD in patients with SLE may increase the CV risk, but further studies are needed in order to conclude this premise.
Glucocorticoids in high doses for short periods of time improve periodontal symptoms by means of diminishing the infl ammation. On the other hand, long-term glucocorticoids cause extended periodontal damage (by decreasing bone density and by determining the avulsion of teeth).
Regarding the management of SLE, rheumatologists should take into account the patients' high risk of developing PD, the potential effect of PD on the course of SLE and also the oral side effects of glucocorticoids.