REVIEW S51

Thomas E. Van Dyke The impact of genotypes and immune reactivity on peri-implant inflammation: identification and therapeutic use of anti-inflammatory drugs and immunomodulators Thomas E. Van Dyke, DDS, PhD Department of Periodontol- ogy, The Forsyth Institute, 245 First Street Cambridge, MA 02142 USA Tel: 617-892-8503 Fax: 617-892-8504 Email: [email protected]

Key words genetic markers, inflammation, lipoxins, peri-implantitis, resolvins

Purpose: Peri-implantitis is a growing problem because of the increasing number of patients rehabili- tated with oral implants. Predictors of susceptibility to peri-implantitis and pharmacological agents for the treatment of inflammatory peri-implantitis are of significant interest to the practicing community. The goal of this paper is to review the current literature relating to the impact of genotypes and immune reactivity on peri-implant inflammation and to identify the potential use of anti-inflamma- tory and immunomodulatory drugs. Materials and methods: A MEDLINE search of the relevant literature was performed and all related articles were evaluated. Results: There was little support in the literature for a specific genotype or phenotype of immune reactivity that could be reliably used as an indicator of susceptibility to peri-implant disease. Further- more, no currently available anti-inflammatory drugs were identified that could be used long term for the treatment of inflammatory peri-implantitis. New pathways for the control of inflammation have been identified that have the potential for new pharmacologic therapies. The basis for these new discoveries is reviewed in some detail. Conclusions: While there are varying reports of associations of specific genotypes with peri-implan- titis, the studies are inconsistent and generally underpowered precluding any interpretable pattern. Inflammatory peri-implantitis is associated with increased local inflammation, but no consistent sys- temic inflammatory markers have been identified. To date, no safe and effective anti-inflammatory therapy is known for the treatment of peri-implantitis. However, a new class of molecules in devel- opment for the active regulation and resolution of inflammation shows theoretical promise for the treatment of inflammatory lesions.

Conflict-of-interest statement: Boston University is assigned patents on lipoxins and resolvins that are licensed for clinical development and are subject to consultant agreements for Dr Thomas E. Van Dyke.

Eur J Oral Implantol 2012;5(Suppl):S51–S60 S52 Van Dyke Host response in peri-implant disease

Introduction still have teeth in the mouth with periodontitis. As is common with studies of peri-implant inflammatory Peri-implant diseases (peri-implant mucositis and disease, it is difficult to draw firm conclusions because peri-implantitis) are mostly associated with peri- of the reported criteria for implant success, that is, implant biofilm accumulation. These diseases are lack of mobility and retention of the implant. Where inflammatory; either limited to the surrounding bone loss and pocketing were measured, it seems mucosa (peri-implant mucositis) or involving the clear that peri-implant disease was more preva lent supporting bone (peri-implantitis)1. The diagnosis in former and current periodontitis patients. Other of peri-implant inflammatory disease is based upon confounders include supportive therapy regimens clinical criteria including (BoP), and smoking. suppuration, increased probing depths relative to The reporting of genetic markers associated with baseline, mucosal recession, fistula formation and peri-implant disease has been limited to polymorph- peri-implant swelling and hyperplasia1. However, isms in interleukin (IL)-1a, IL-1b and the natural the peri-implant lesion is an inflammatory lesion specific inhibitor IL-1 receptor antagonist (IL-1ra). that histologically does not differ substantially from There have been a number of small clinical stud- periodontitis; compared to periodontitis, the peri- ies that report conflicting results. The most recent implantitis lesion appears more acute with a greater reports14,15 document peri-implantitis and implant predominance of neutrophils and macrophages2. loss, respectively. Hamdy and Ebrahem14 reported The purpose of this review is to assess the cur- 50 implant cases, 25 with peri-implantitis and 25 rent understanding of susceptibility to peri-implant without. Dirschnabel et al15 reported 277 cases, 92 disease and to evaluate available evidence for strat- with implant loss. The combination of IL-1 allele 2 egies to control peri-implant inflammation. A litera- at IL-1A-889 and IL-1B+3954 has been reported to be ture search of the MEDLINE database was carried associated with increased periodontal tissue destruc- out through July 2011. Since there were few studies tion16 and was the association of interest for these related to genotype and immune reactivity in peri- studies of peri-implantitis progression, treatment implant diseases, parallels to and periodon- outcomes and implant loss. Genotype-positive indi- titis were made where appropriate. viduals were found to cluster in the peri-implantitis and implant-loss groups and demonstrated a poor response to treatment compared to genotype-nega- Genotype and immune reactivity tive subjects, but associations were marginal and not significant. All subjects were non-smokers by history, The search for diagnostic markers and predictors but this was not verified biochemically. Similar find- of susceptibility to peri-implant diseases has mainly ings were reported by Feloutzis et al17 in a popula- focused on biochemical markers in peri-implant cre- tion that comprised both smokers and non-smokers. vicular fluid (PICF). A variety of pro-inflammatory This study found a synergistic effect of smoking with indicators have been assessed measuring PICF vol- the carriage of the IL-1 genotype, demonstrating the ume3, proteases4-6 and an array of cytokines and impact of the polymorphism only in smokers; the lipid mediators of inflammation7-9. It is clear that same finding was reported for periodontitis16. These inflammation is associated with peri-implantitis, findings were confirmed later in studies by Jansson et however, virtually all of the studies to date are cross- al18 and Laine et al19. However, Wilson and Nunn20 sectional in design. Prospective longitudinal studies and Lachmann et al21 were unable to demonstrate are required to determine the role of these markers any association between implant failure or bone loss in disease progression. and the IL-1 genotype22. Perhaps the most reliable predictor of susceptibil- The study of IL-1 polymorphisms in periodontitis ity to peri-implantitis is susceptibility to periodontitis. has yielded similar inconsistent results23. There are A series of systematic reviews9-13 point to a strong major differences between populations from differ- association between peri-implantitis and patients ent geographic locations and with different ethnic who lost their teeth to periodontitis or patients who backgrounds in the distribution of the alleles and

Eur J Oral Implantol 2012;5(Suppl):S51–S60 Van Dyke Host response in peri-implant disease S53

their association with disease. Furthermore, numer- and sometimes the availability, of compounds that ous studies have attempted to correlate other poly- inhibit cyclooxygenases, precluding their use in peri- morphisms with disease, with limited results24. In odontal therapy. general, the sample size of all reports is too small and the same problems of ethnicity and geographic loca- tion rapidly surface. Sample populations of >10,000 What have we learned from are needed for these studies and gene-wide scans periodontitis? for polymorphisms should be performed. To date, reports are limited to samples of hundreds of subjects Infection by bacteria is well-documented as the pri- using candidate genes. As a consequence, reports mary aetiologic agent for the development of peri- are inconclusive and inconsistent. Considerably more odontal disease27. However, the specific bacterium or data is needed to assess whether this approach is bacteria that cause periodontitis is not established28. viable for periodontitis and peri-implantitis25. Although cross-sectional studies have identified cer- tain bacteria, such as (P.g.), (T.f.) and Immune reactivity and (T.d.), the ‘’ bacteria29, the cause-and- peri-implantitis effect relationship of specific microbes remains to be determined in longitudinal studies28. It is now Peri-implant mucositis and peri-implantitis, like peri- known that the host-mediated response to the bio- odontitis, are inflammatory diseases initiated by oral film in the susceptible individual destroys tissue and microbial biofilm. Current understanding of the bone of the in the pathogenesis of pathogenesis of these diseases indicates that the the disease27,30. It is important to remember that host response to the biofilm destroys the supporting physiologic inflammation is a good thing; the acute structures of the implant or tooth; the pathogenesis inflammatory response is protective. However, there of the disease is mediated by an overly robust inflam- are several possible outcomes that follow the acute matory response in susceptible individuals. Whether response; elimination of the pathogen and return to the excessive host response in under genetic or epi- homeostasis is the optimum response. If this fails, genetic control or mediated by environmental fac- abscess formation, repair with scarring or chronic tors, or both, is a matter of conjecture. Nonetheless, inflammation are less desirable endpoints27. therapeutic strategies for peri-implant disease have, The current therapeutic approach to control to date, solely focused on the bacterial aetiology inflammation is to remove the aetiology. Since the rather than interruption of the pathogenesis1. In this infection is probably related to overgrowth of com- review, the potential of host modulation as a thera- mensal organisms and the oral cavity cannot be peutic strategy will be explored. Since there is virtu- sterilised, the control of inflammation in susceptible ally nothing published with regard to host-modula- people by this method is difficult and recurrence of tory therapy for peri-implant disease, learning from disease is high31. This probably is the same reason studies of host-modulatory therapy in periodontitis that people who lose their teeth to periodontitis are will serve as the scientific basis. at greater risk for peri-implantitis9. Pharmacologic The traditional focus of therapies for inflamma- approaches to control inflammation in periodontitis tory disease has been proinflammatory pathways. are limited by available pharmacological agents such Anti-inflammatory pharmacologic agents (non- as NSAIDs that antagonise pro-inflammatory path- steroidal anti-inflammatory drugs, NSAIDs) have ways and/or signalling32. The use of these drugs in been shown to prevent and slow the progression periodontics is limited by their side-effect profiles26. of periodontitis in animals and humans26. However, Salvi et al33 suggested that the appropriate approach in recent years, new data uncovering serious side to control peri-implant disease was to target inflam- effects resulting from chronic use of NSAIDs such as mation. However, the available data supporting the ibuprofen, flurbiprophen and the celecoxib class of concept used drugs that are no longer considered drugs, has raised safety concerns that limit the use, desirable.

Eur J Oral Implantol 2012;5(Suppl):S51–S60 S54 Van Dyke Host response in peri-implant disease

The uncovering of the cardiovascular com- ciated with , including inflam- plications of NSAIDs led to new concepts in matory cell recruitment, oedema, pain, collagen inflammation research. Rather than inhibition of destruction and bone resorption38. pro-inflammatory signals, work was focused on The switch from pro-inflammation to pro- reso- regulation of inflammation. New pathways and lution is temporal and is linked to the expression processes underlying the resolution of inflamma- of new genes. As the acute inflammatory lesion tion have been discovered, stimulating increased matures, there is an accumulation of cells contain- interest in pro-resolving lipid mediators of inflam- ing lipoxygenases and corresponding pro-inflam- mation and the pathways that these molecules matory products such as prostaglandin E2 (PGE2) regulate. The realisation that resolution of inflam- and the HETEs. In the normal resolution process, mation is an active process (rather than a passive a ‘class switch’ occurs within neutrophils28,39 giv- decay of pro-inflammatory signals) mediated by a ing rise to the synthesis of pro-resolving molecules new class of pro-resolving lipid mediator receptor through spatially and temporally distinct pathways. agonists28,30,34,35 has sparked interest in the poten- Pro-resolving lipid mediators are produced from AA tial use of pro-resolving molecules to neutralise and through cell-to-cell interactions where 15-S-H(p) eliminate inflammatory leukocytes and prevent peri- ETE is produced by oxidation of AA by 15-LO, fol- odontal pathology27,30,32. lowed by 5-LO oxidation of the same molecule. The products include lipoxins such as lipoxin A4 (LXA4) 27,40 and lipoxin B4 (LXB4) . The lipoxins (LXs) pro- Physiological processes involved duced bind to FPRL135, also known as the FMLP in the restoration of tissue receptor, to stimulate the resolution of inflammation. The actions of LXs include limitation of polymorpho- homeostasis nuclear neutrophil (PMN) migration into sites of Resolution pathways are initiated following an acute inflammation with promotion of PMN apoptosis, inflammatory response. Release of arachidonic acid and activation of monocytes to a non-phlogistic (AA) from cell membranes generates lipid mediators phenotype (e.g. non-inflammatory). The resulting of inflammation including the classic eicosanoids, non-phlogistic macrophage clears apoptotic PMN prostanoids (e.g. prostaglandins, thromboxanes) by phagocytosis and exhibits enhanced clearance of and prostacyclins, as well as leukotrienes (LTs). The bacteria at mucosal surfaces41-44. response is initiated through numerous pathways; in The actions of aspirin are quite distinct from the case of periodontal tissues, there is a large body other NSAIDs and aspirin plays an important role in of evidence implicating pattern recognition recep- the biology of resolution of inflammation. Aspirin tors and pathogen-associated molecular patterns has a unique property amongst the NSAIDs: it (PAMPs)36. Agonist stimulation of G-protein recep- inactivates COX-2 by acetylation of the enzyme tors in the cell membrane triggers the release of AA creating a new, active enzyme. Aspirin-modified from phosphatidylcholine by the enzyme phospholi- COX-2 is a 15R lipoxygenase. The product of this 37 pase A2 (PLA2) (reviewed by Kantarci et al ). The so called ‘aspirin-triggered’ pathway is 15R-H(p) free AA is metabolised by cyclooxygenase-1- (COX- ETE, which is further metabolised by PMN 5-LO to 45 1) and COX-2-dependent pathways that result in yield 15R-LXA4, or aspirin-triggered lipoxin (ATL) . the generation of prostanoids, or a lipoxygenase ATLs are more bioactive and possess more power- (LO)-dependent pathway. The three LOs are cell ful pro-resolving properties27,46,47. Interestingly, specific; 5-LO is expressed by myeloid cells, 12-LO statins also stimulate the production of 15R-LXA4 is expressed by platelets and 15-LO is expressed by through a nitrosylation reaction involving COX-2, endothelium and epithelium. The end products of perhaps explaining in part the anti-inflammatory the LOs are 5-, 12- or 15-hydroxyeicosatetraenoic actions of statins48. acid (HETE); the 5-HETE is further metabolised to While aspirin has been shown to facilitate resolu- 37 LTs, specifically LTB4 . Prostanoids and LTs produce tion pathways, other NSAIDs may actually be toxic a wide range of pathophysiological responses asso- to the resolution of inflammation. The stimulus for

Eur J Oral Implantol 2012;5(Suppl):S51–S60 Van Dyke Host response in peri-implant disease S55 induction of generation of lipoxins or ATLs triggered Resolution agonists derived by pro-inflammatory lipid mediators such as PGE2 from dietary PUFA may explain the potentially serious cardiovascular consequences of chronic use of selective COX-2 While lipoxins are produced from the endogenous antagonists. Inflammation is a multifactor process, -6 fatty acid substrate arachidonic acid, another and a single ‘pan anti-inflammatory’ agent does not family of pro-resolution agonists is metabolised by exist that antagonises deleterious pathways while the same LO enzyme systems from dietary -3 preserving resolution pathways30,32. As a result, polyunsaturated fatty acids (PUFAs): the resolvins the unfortunate consequence of antagonism of a and protectins52. The resolvins and protectins and single enzymatic pathway (COX-2) by NSAIDs (e.g. are derived from eicosapentaenoic acid (EPA) and rofecoxib, valdecoxib and celecoxib) is the inhibi- docosahexaenoic acid (DHA), the major -3 PUFAs tion of resolution of inflammation. While inhibition found in fish oil53. EPA is the substrate for resolvins of of COX-2 by NSAIDs may attenuate signs of acute the E series and DHA the resolvin D series as well as 28,52 inflammation (pain, swelling), the reduction in PGE2 protectins (or neuroprotectins in neural tissues) . would fail to generate pro-resolving lipoxins that Resolvins and protectins bind to distinct receptors to are required for restoring homeostasis in tissues39,49. induce anti-inflammatory and pro-resolving path- The resulting chronic low-grade vascular inflamma- ways with a unique mechanism of action that dif- tion may explain the increased risk of cardiovascular fers from the lipoxins28,52,54. The actions of aspirin- disease in individuals receiving long-term treatment modified COX-2 produces a class of molecule with with selective COX-2 inhibitors49. the same increased stability and duration of action as ATLa55. Resolution of inflammation induced by resolvins The potential of lipoxins as and protectins is characterised by stopping neu- therapeutics trophil infiltration and driving neutrophil apoptosis, thereby attracting non-phlogistic monocytes that Lipoxins drive resolution of inflammation and restore differentiate into resolution macrophages. Resolu- tissue homeostasis. However, these short-lived auta- tion macrophages exhibit enhanced phagocytosis coids are rapidly metabolised and unstable, making of apoptotic neutrophils and enhanced clearance them poor pharmacologic candidates in their native of bacteria at mucosal surfaces, promoting wound form. To circumvent this problem, stable analogues clearance and return to homeostasis53,56. The of lipoxins have been synthesised by total organic rapid resolution of acute inflammation induced by synthesis. Aspirin-triggered lipoxin analogues (ATLa) resolvins and protectins are, at least in part, the basis selectively interact with lipoxin receptors on neu- of the beneficial actions of dietary -3 PUFAs (EPA trophils in vitro. The potential of ATLa as pharmaco- and DHA) in inflammatory diseases such as cardio- logic anti-inflammatory agents was examined in ani- vascular disease57-59. mal models of PMN-mediated tissue injury50,51. In the mouse, intravenous injection of the pro-inflam- matory mediator LTB4 induces peripheral vasodilata- Management of inflammation tion and leakage of blood from capillaries, resulting in periodontitis in stasis of blood in tissues. The resulting cyanosis can be enhanced by intravenous blood dye that also The goal of treatment of an inflammatory disease leaks into the tissues rendering all exposed skin on such as periodontitis is the return of tissue homeo- the mouse (e.g. the ear) a distinctive blue. ATLa was stasis, which implies complete healing, with regen- topically applied to one ear and vehicle alone to the eration of lost or damaged tissue without scarring. other immediately before injection of the LTB4. The The rapid and complete elimination of invading ATLa-treated ear did not turn blue, illustrating the leukocytes from the lesion is, therefore, the ideal potential of topically applied ATLa for preventing outcome following an inflammatory event28. For inflammation. periodontal disease, typical intervention procedures

Eur J Oral Implantol 2012;5(Suppl):S51–S60 S56 Van Dyke Host response in peri-implant disease

target removal of the aetiologic agents associated plete absence of inflammation and complete pre- with inflammation to arrest further progression of vention of gingivitis and periodontitis. There were periodontal disease. However, to date, therapeutic no inflammatory changes, osteoclast formation or procedures do not restore tissue homeostasis; the bone loss observed. Radiographic assessment also outcome of periodontal therapy is repair with fibrosis demonstrated reduced bone loss with RvE1 treat- and scarring resulting in a reduced periodontium. ment (<5%) compared with vehicle controls (>35%) Periodontitis and the inability to reach homeostasis or ligature alone (~12%) (P < 0.05)61,63. as defined above can be attributed to a failure of resolution of inflammation pathways60. Treatment of experimental periodontitis Pro-resolving therapy in an experimental model of The potential of RvE1 for the prevention of peri- odontitis was established in the rabbit model61. periodontitis Experiments were then undertaken to determine To explore the potential of resolution agonists as the potential for RvE1 treatment of established peri- therapeutics in periodontitis, an experimental model odontitis63. In this series of experiments, experimen- of P.g. and ligature-induced periodontitis was estab- tal periodontitis was induced for 6 weeks in three lished. Silk ligatures were tied bilaterally around the groups of rabbits as above with ligatures and topical second premolars of both mandibular quadrants and application of P.g. The first group of animals was P.g. was applied to ligatured sites three times a week sacrificed at 6 weeks to determine the extent of peri- (Monday, Wednesday and Friday) for 6 weeks61. odontitis (baseline disease). In the other two groups, Inflammation and bone loss was induced with appli- P.g. application ceased and animals received either cation of P.g., which was inhibited with systemic RvE1 or vehicle treatment following the same regi- administration of metronidazole; ligature alone men for an additional 6 weeks prior to sacrifice at induced little inflammation and disease61. Histo- week 12. logical, morphometric and radiological evaluations Compared to baseline disease animals, vehicle- revealed severe inflammation with profound loss of treated rabbits exhibited progression of disease and collagen and bone with infrabony (vertical bone loss) bone loss over the second 6 weeks of the experiment. pockets. P.g. + ligature induced 43.0% (± 4.6) bone Osteoclasts and osteoclast-like cells were clearly loss over the 6-week period whereas ligature alone observed in the baseline disease and vehicle control resulted in less than 10% bone loss. groups. Treatment with RvE1 resulted in complete An evaluation of resolvin E1 (RvE1) for the pre- resolution of inflammation and restoration of both vention of inflammation and bone loss (periodon- soft and hard tissues. There was no inflammation titis) in the model was explored61. RvE1, prepared clinically or histologically with an absence of osteo- by total organic synthesis34,62, was topically applied clasts associated with bone. The clinical appearance directly to study teeth (4 μg/tooth) three times of the tissues resembled pre-disease, naive animals. weekly over the 6-week study period immediately There was a significant reduction in soft tissue pocket after P.g. application. The control group received depth, bone loss and infrabony defect depth com- topical application of the vehicle (ethanol); a third pared to baseline disease and vehicle control groups group received ligature alone (no P.g., no therapy). (P < 0.01). Histological assessment of undecalcified Ligature alone did not produce much disease. How- ground sections revealed complete regeneration of ever, animals receiving P.g. + ligature with the etha- the periodontium with new formation, nol as therapy exhibited marked inflammation and connective tissue attachment and new bone, sug- significant bone loss. Histology revealed a profound gesting regeneration of the natural architecture of inflammatory infiltrate, loss of collagen and loss of the periodontal organ63. bone. Osteoclasts were visible in resorption lacunae. Using RvE1 as a mono-therapy resulted in a com-

Eur J Oral Implantol 2012;5(Suppl):S51–S60 Van Dyke Host response in peri-implant disease S57

Inflammation and the biofilm has no direct antiseptic properties. A combination of known resolution agonist properties probably comes The observations relating control of inflammation and into play: resolvin molecules promote the release progression of periodontitis, albeit in an animal model, of antimicrobial peptides, such as defensins and challenge existing paradigms regarding the tempo- bacteri cidal/permeability-increasing protein (BPI)66; ral association of biofilm composition and disease. In P.g. is an asaccharolytic organism that requires essen- retrospect, the observation that pharmacologic con- tial amino acids as a food source derived from deg- trol of inflammation can control progression of peri- radation of collagen fragments, such as those pro- odontitis is not new; only the agents are new. Experi- duced during an inflammatory response28; increased ments in the 1980s with flurbiprophen and the 1990’s bacterial clearance by resolution macrophages53,67. with tetracyclines clearly demonstrate the proof of These observations suggest that an inflammatory principle that control of inflammation independent of site is an environment that favours bacterial growth. plaque (biofilm) control can have a beneficial impact The implication is that inflammation precedes the on the progression of periodontitis64. In the rabbit emergence of pathogens in the periodontal pocket. experiment, RvE1 treatment stimulates the resolution This hypothesis is supported by the observations of of inflammation and the restoration of bone loss; the Tanner et al that demonstrate in a longitudinal study temporal composition shifts in the microflora in the that gingivitis predicts future attachment loss, but biofilm were also investigated63. The dynamic nature specific bacteria or groups of bacteria do not68. of biofilm composition is well studied and significant changes in the balance of microbial species have been described65. A significant reduction in oral microflora Resolution of inflammation due to the direct actions of RvE1 was not anticipated; in human periodontitis previous studies failed to demonstrate significant anti- bacterial actions for RvE128. None of the resolution agonists has, to date, been DNA-DNA checkerboard hybridisation analyses approved for clinical use, although several are were performed using probes to human isolates. The in development. However, the biologic impact of microbial species present in the rabbit periodontal dietary -3 PUFA on inflammation and inflamma- flora prior to ligature placement were relatively few tory disease has been described69. The question of and largely comprised Actinomyces viscosus and whether there is a clinical impact on periodontitis Peptostreptococcus micros-like organisms63. Intro- is beginning to be investigated. In a pilot trial, 80 duction of P.g. to the ligatures resulted in high num- subjects with moderate to severe periodontitis were bers of this organism in 90% of the rabbit plaque treated with either followed samples that was apparent at 6 weeks and persisted by a regimen of 900 mg of EPA/DHA + 81 mg aspirin to 12 weeks. A corresponding shift in the composi- daily for 6 months, or the same mechanical regimen tion of the anaerobic flora occurred during this time and placebo tablets. Addition of the resolvin-gen- with overgrowth of resident organisms (e.g. A. vis- erating dietary supplement to standard peri odontal cosus, P. micros, , Streptococ- therapy provided an added benefit, reducing pocket cus intermedius, and Eikenella corrodens-like organ- depth and increasing clinical attachment, and pro- isms) and the emergence of other pathogens such viding measurable reductions in inflammatory medi- as Aggregatibacter actinomycetemcomitans-like ators in saliva. and -like organisms, and an overall increase in bacterial load. Treatment with RvE1 reversed the changes in the microflora induced Can this be applied to peri-implant by development of periodontitis. P.g. disappeared inflammatory diseases? as did A. actinomycetemcomitans and the levels of other organisms returned to pre-disease levels. Until recently, there was insufficient data to answer The reason for the changes in biofilm composi- the question of whether periodontitis and peri- tion with RvE1 treatment is unknown, since RvE1 implantitis lesions have common characteristics.

Eur J Oral Implantol 2012;5(Suppl):S51–S60 S58 Van Dyke Host response in peri-implant disease

While the number of studies is limited, evaluation of lipid mediators that drive resolution of inflamma- the available evidence suggests that the lesions are tion enhance microbial clearance at mucosal surfaces not substantially different and that, if anything, peri- and improve bacterial clearance in infection73. The implant disease is more inflammatory in nature with a potential for treating peri-implant disease with one larger percentage of neutrophils and macrophages 2. or more of these specialised lipid mediators awaits These observations suggest that host modulatory thorough investigation. approaches that are successful for periodontitis will likely be applicable to peri-implantitis. This, of course, does not provide much in the way References of therapies that can be applied at the clinical level today. There are reports suggesting that bisphos- 1. Zitzmann NU and Berglundh T. Definition and prevalence of peri-implant diseases. J Clin Periodontol 2008;35:286-291. phonates might be useful in preventing peri-implant 2. Berglundh T, Zitzmann NU and Donati M. Are peri-implan- bone loss70, but these animal studies do not have titis lesions different from periodontitis lesions? J Clin Peri- human follow-up so far. Considering the side effect odontol 2011;38(Suppl 11):188-202. 3. Niimi A and Ueda M. Crevicular fluid in the osseointegrated profile of bisphosphonates, their use for peri-implant implant sulcus: a pilot study. Int J Oral Maxillofac Implants disease is unlikely, as is the use of NSAIDs. There is a 1995;10:434-436. 4. Eley B, Cox S, and Watson R. Protease activities in peri- single report of the use of the H2 (histamine receptor) implant sulcus fluid from patients with permucosal osseointe- antagonist, cimetidine, for possible use as a therapy grated dental implants: correlation with clinical parameters. Clin Oral Implants Res 1991;2:62-70. 71 for periodontitis . In addition, in a review of mast 5. Boutros SM, Michalowicz BS, Smith QT, Aeppli DM. Cre- cell function in oral inflammation, Zizzi et al72 suggest vicular fluid enzymes from endosseous dental implants and natural teeth. Int J Oral Maxillofac Implants 1996;11: that histamine inhibitors are a logical host modula- 322-330. tory therapy for peri-implantitis. However, with the 6. Teronen O, Konttinen YT, Lindqvist C, Salo T, Ingman T, Lauhio A, Ding Y, Santavirta S, Sorsa T. Human neutrophil possible exception of dietary fatty acid supplements collagenase MMP-8 in peri-implant sulcus fluid and its inhi- with aspirin, which have never been evaluated for bition by clodronate. J Dent Res 1997;76:1529-1537. 7. Salcetti J, Moriarity J, Cooper L, Smith F, Collins J, Socran- peri-implant disease, it would appear that treatment sky S, Offenbacher S. The clinical, microbial and host is limited to therapeutic strategies that target the bio- response characteristics of the failing implant. Int J Oral film until new compounds or drugs that regulate the Maxillofac Implants 1997;12:32-42. 8. Ataoglu H, Alptekin NO, Haliloglu S, Gursel M, Ataoglu T, inflammatory response become available. Serpek B, Durmus E. Interleukin-1beta, tumor necrosis factor- alpha levels and neutrophil elastase activity in peri-implant crevicular fluid. Clin Oral Implants Res 2002;13:470-476. 9. Heitz-Mayfield LJA. Peri-implant diseases: diagnosis and risk indicators. J Clin Periodontol 2008;35(Suppl 8):292-304. Summary 10. Van der Weijden G, van Bemmel K, Revert S. Implant therapy in partially edentulous, periodontally compromised The primary aetiology of peri-implant inflamma- patients: a review. J Clin Periodontol 2005;32:506-511. 11. Schou S, Holmstrup P, Worthington HV, Esposito M. Out- tory diseases is bacterial, but it appears that the host come of implant therapy in patients with previous tooth loss inflammatory response may be a critical determinant due to periodontitis. Clin Oral Implants Res 2006;17(Suppl 2):104-123. in the pathogenesis of peri-implantitis. New data 12. Karoussis IK, Kotsovilis S, Fourmousis I. A comprehensive suggest that failure of resolution of inflammation and critical review of prognosis in periodon- pathways may play as important a role in disease tally compromised partially edentulous patients. Clin Oral Implants Res 2007;18:669-679. as overproduction of pro-inflammatory mediators, 13. Quirynen M, Abarca M, Van Assche N, Nevins M, van extrapolating from periodontitis studies. It is clear Steenberghe D. Impact of supportive periodontal therapy and implant surface roughness on implant outcome in that the current pharmacopoeia has limited potential patients with a history of periodontitis. J Clin Periodontol for the treatment of peri-implantitis. The efficacy 2007;34:805-815. 14. Hamdy AA, Ebrahem MA. The effect of interkeukin-1 allele potential of the drugs has been demonstrated in 2 genotype (IL-1a(-889) and IL-1b(+3954)) on the indi- human periodontitis, but the side effect profile of vidual’s susceptibility to peri-implantitis: case-control study. J Oral Implantol 2011;37:325-334. NSAIDs and bisphosphonates precludes their chronic 15. Dirschnabel AJ, Alvim-Pereira F, Alvim-Pereira CC, use. Drugs in development, such as the resolution Bernadino JF, Rosa EA, Trevilatto PC. Analysis of the associ- ation of IL1B(C-511T) polymorphism with dental implant loss agonists, are physiologic molecules, not inhibitors and the clusterization phenomenon. Clin Oral Implants Res that will compromise host defence. Also, the same 2011;22:1235-1241.

Eur J Oral Implantol 2012;5(Suppl):S51–S60 Van Dyke Host response in peri-implant disease S59

16. Kornman KS, Crane A, Wang HY, di Giovine FS, Newman 35. Chiang N, Serhan CN, Dahlen SE, Drazen JM, Hay DW, MG, Pirk FW. The interleukin-1 genotype as a severity fac- Rovati GE, Shimizu T, Yokomizo T, Brink C. The lipoxin tor in adult periodontal disease. J Clin Periodontol 1997;24: receptor ALX: potent ligand-specific and stereoselective 72-77. actions in vivo. Pharmacol Rev 2006;58:463-487. 17. Feloutzis A, Lang NP, Tonetti MS, Bürgin W, Brägger U, 36. Darveau RP. Periodontitis: a polymicrobial disruption of host Buser D, Duff GW, Kornman KS. Il-1 gene polymorphism homeostasis. Nat Rev Microbiol 2010;8:481-490 and smoking as risk factors for peri-implant bone loss 37. Kantarci A, Oyaizu K, Van Dyke TE. Neutrophil-mediated in a well-maintained population. Clin Oral Implants Res tissue injury in periodontal disease pathogenesis: find- 2003;14:10-17. ings from localized . J Periodontol 18. Jansson H, Hamberg K, De Bruyn H, Brattha G. Clinical 2003;74:66-75. consequences of IL-1 genotype on early implant failures in 38. Pouliot M, Clish CB, Petasis NA, Van Dyke TE, Serhan CN. patients under periodontal maintenance. Clin Implant Dent Lipoxin A(4) analogues inhibit leukocyte recruitment to Relat Res 2005;1:51-59. Porphyromonas gingivalis: a role for cyclooxygenase-2 19. Laine ML, Leonhardt A, Roos-Jansåker AM, Peña AS, van and lipoxins in periodontal disease. Biochemistry 2000;39: Winkelhoff AJ, Winkel EG, Renvert S. IL-1RN gene polymor- 4761-4768. phism is associated with peri-implantitis. Clin Oral Implants 39. Levy BD, Clish CB, Schmidt B, Gronert K, Serhan CN. Lipid Res 2006;17:380-385. mediator class switching during acute inflammation: signals 20. Wilson T, Nunn M. The relationship between the inter- in resolution. Nat Immunol 2001;2:612-619. leukin-1 periodontal genotype and implant loss: initial data. 40. Serhan CN. Lipoxin biosynthesis and its impact in inflam- J Periodontol 1999;70:724-729. matory and vascular events. Biochim Biophys Acta 1994; 21. Lachmann S, Kimmerle-Müller E, Axmann D, Schneider L, 1212:1-25. Weber H, Haas R. Association between peri-implant crev- 41. Serhan CN, Fiore S, Brezinski DA, Lynch S. Lipoxin A4 icular fluid volume, concentrations of crevicular inflamma- metabolism by differentiated HL-60 cells and human mono- tory mediators, and composite IL-1A -889 and IL-1B +3954 cytes: conversion to novel 15-oxo and dihydro products. genotype. A cross-sectional study on implant recall patients Biochemistry 1993;32:6313-6319. with and without clinical signs of peri-implantitis. Clin Oral 42. Maddox JF, Serhan CN. Lipoxin A4 and B4 are potent Implants Res 2007;18:212-223. stimuli for human monocyte migration and adhesion: selec- 22. Bormann KH, Stühmer C, Z’Graggen M, Kokemöller H, tive inactivation by dehydrogenation and reduction. J Exp Rücker M, Gellrich NC. IL-1 polymorphism and periim- Med 1996;183:137-146. plantitis. A literature review. Schweiz Monatsschr Zahnmed 43. Maddox JF, Hachicha M, Takano T, Petasis NA, Fokin VV, 2010;120:510-520. Serhan CN. Lipoxin A4 stable analogs are potent mimet- 23. Greenstein G, Hart TC. A critical assessment of interleukin-1 ics that stimulate human monocytes and THP-1 cells via (IL-1) genotyping when used in a genetic susceptibility test a G-protein-linked lipoxin A4 receptor. J Biol Chem 1997; for severe . J Periodontol 2002;73: 272:6972-6978. 231-247. 44. Mitchell S, Thomas G, Harvey K, Cottell D, Reville K, Ber- 24. Zhang J, Sun X, Xiao L, Xie C, Xuan D, Luo G. Gene poly- lasconi G, Petasis NA, Erwig L, Rees AJ, Savill J, Brady HR, morphisms and periodontitis. Periodontol 2000 2011;56: Godson C. Lipoxins, aspirin-triggered epi-lipoxins, lipoxin 102-124. stable analogues, and the resolution of inflammation: stimu- 25. Huynh-Ba G, Lang NP, Tonetti MS, Zwahlen M, Salvi GE. lation of macrophage phagocytosis of apoptotic neutrophils Association of the composite IL-1genotype with peri- in vivo. J Am Soc Nephrol 2002;13:2497-2507. implantitis: a systematic review. Clin Oral Implants Res 45. Claria J, Serhan CN. Aspirin triggers previously unde- 2008;19:1154-1162. scribed bioactive eicosanoids by human endothelial cell- 26. Williams RC, Jeffcoat MK, Howell TH, Reddy MS, Johnson leukocyte interactions. Proc Natl Acad Sci U S A 1995;92: HG, Hall CM, Goldhaber P. Topical flurbiprofen treatment of 9475-9479. periodontitis in beagles. J Periodontal Res 1988;23:166-169. 46. Claria J, Lee MH, Serhan CN. Aspirin-triggered lipoxins 27. Van Dyke TE, Serhan CN. Resolution of inflammation: a (15-epi-LX) are generated by the human lung adenocarci- new paradigm for the pathogenesis of periodontal diseases. noma cell line (A549)-neutrophil interactions and are potent J Dent Res 2003;82:82-90. inhibitors of cell proliferation. Mol Med 1996;2:583-596. 28. Van Dyke TE. Control of inflammation and periodontitis. 47. Serhan CN, Maddox JF, Petasis NA, Akritopoulou-Zanze I, Periodontol 2000 2007;45:158-166. Papayianni A, Brady HR, Colgan SP, Madara JL. Design 29. Socransky SS, Haffajee AD. Evidence of bacterial etiology: a of lipoxin A4 stable analogs that block transmigration historical perspective. Periodontol 2000 1994;5:7-25. and adhesion of human neutrophils. Biochemistry 1995; 30. Serhan CN. Novel omega -- 3-derived local mediators in 34:14609-14615. anti-inflammation and resolution. Pharmacol Ther 2005; 48. Birnbaum Y, Ye Y, Lin Y, Freeberg SY, Nishi SP, Martinez JD, 105:7-21. Huang MH, Uretsky BF, Perez-Polo JR. Augmentation of 31. Socransky SS, Haffajee AD. Dental biofilms: difficult thera- myocardial production of 15-epi-lipoxin-a4 by pioglitazone peutic targets. Periodontol 2000 2002;28:12-55. and atorvastatin in the rat. Circulation 2006;114:929-935. 32. Serhan CN, Brain SD, Buckley CD, Gilroy DW, Haslett C, 49. FitzGerald GA. COX-2 in play at the AHA and the FDA. O’Neill LA, Perretti M, Rossi AG, Wallace JL. Resolution of Trends Pharmacol Sci 2007;28:303-307. inflammation: state of the art, definitions and terms. Faseb J 50. Takano T, Fiore S, Maddox JF, Brady HR, Petasis NA, 2007;21:325-332. Serhan CN. Aspirin-triggered 15-epi-lipoxin A4 (LXA4) and 33. Salvi GE, Williams RC, Offenbacher S. Nonsteroidal anti- LXA4 stable analogues are potent inhibitors of acute inflam- inflammatory drugs as adjuncts in the management of peri- mation: evidence for anti-inflammatory receptors. J Exp odontal diseases and peri-implantitis. Curr Opin Periodontol Med 1997;185:1693-1704. 1997;4:51-58. 51. Takano T, Clish CB, Gronert K, Petasis N, Serhan CN. 34. Arita M, Bianchini F, Aliberti J, Sher A, Chiang N, Hong S, Neutrophil-mediated changes in vascular permeability are Yang R, Petasis NA, Serhan CN. Stereochemical assignment, inhibited by topical application of aspirin-triggered 15-epi- antiinflammatory properties, and receptor for the omega-3 lipoxin A4 and novel lipoxin B4 stable analogues. J Clin lipid mediator resolvin E1. J Exp Med 2005;201:713-722. Invest 1998;101:819-826.

Eur J Oral Implantol 2012;5(Suppl):S51–S60 S60 Van Dyke Host response in peri-implant disease

52. Serhan CN, Chiang N, Van Dyke TE. Resolving inflamma- pentaenoic acid, protects against 2,4,6-trinitrobenzene sul- tion: dual anti-inflammatory and pro-resolution lipid media- fonic acid-induced colitis. Proc Natl Acad Sci U S A 2005; tors. Nat Rev Immunol 2008;8:349-361. 102:7671-7676. 53. Schwab JM, Chiang N, Arita M, Serhan CN. Resolvin E1 and 63. Hasturk H, Kantarci A, Goguet-Surmenian E, Blackwood A, protectin D1 activate inflammation-resolution programmes. Andry C, Serhan CN, Van Dyke TE. Resolvin E1 regulates Nature 2007;447:869-874. inflammation at the cellular and tissue level and restores tis- 54. Serhan CN, Gotlinger K, Hong S, Arita M. Resolvins, sue homeostasis in vivo. J Immunol 2007;179:7021-7029. docosatrienes, and neuroprotectins, novel omega-3-derived 64. Howell TH, Williams RC. Nonsteroidal antiinflammatory mediators, and their aspirin-triggered endogenous epimers: drugs as inhibitors of periodontal disease progression. Crit an overview of their protective roles in catabasis. Prostaglan- Rev Oral Biol Med 1993;4:177-196. dins Other Lipid Mediat 2004;73:155-172. 65. Marsh PD, Bradshaw DJ. The effect of fluoride on the stabil- 55. Serhan CN, Hong S, Gronert K, Colgan SP, Devchand PR, ity of oral bacterial communities in vitro. J Dent Res 1990;69 Mirick G, Moussignac RL. Resolvins: a family of bioactive Spec No:668-671; discussion 682-663. products of omega-3 fatty acid transformation circuits ini- 66. Levy O, Canny G, Serhan CN, Colgan SP. Expression of tiated by aspirin treatment that counter proinflammation BPI (bactericidal/permeability-increasing protein) in human signals. J Exp Med 2002;196:1025-1037. mucosal epithelia. Biochem Soc Trans 2003;31:795-800. 56. Ariel A, Fredman G, Sun YP, Kantarci A, Van Dyke TE, 67. Bannenberg GL, Chiang N, Ariel A, Arita M, Tjonahen E, Luster AD, Serhan CN. Apoptotic neutrophils and T cells Gotlinger KH, Hong S, Serhan CN. Molecular circuits of sequester chemokines during immune response resolution resolution: formation and actions of resolvins and protectins. through modulation of CCR5 expression. Nat Immunol J Immunol 2005;174:4345-4355. 2006;7:1209-1216. 68. Tanner AC, Kent R, Jr., Kanasi E, Lu SC, Paster BJ, Sonis ST, 57. DeCaterina R, Giannessi D, Mazzone A, Bernini W, Murray LA, Van Dyke TE. Clinical characteristics and micro- Lazzerini G, Maffei S, Cerri M, Salvatore L, Weksler B. Vas- biota of progressing slight chronic periodontitis in adults. cular prostacyclin is increased in patients ingesting omega-3 J Clin Periodontol 2007;34:917-930. polyunsaturated fatty acids before coronary artery bypass 69. El-Sharkawy H, Aboelsaad N, Eliwa M, Darweesh M, graft surgery. Circulation 1990;82:428-438. Alshahat M, Kantarci A, Hasturk H, Van Dyke TE. Adjunctive 58. Albert CM, Campos H, Stampfer MJ, Ridker PM, Manson treatment of chronic periodontitis with daily dietary supple- JE, Willett WC, Ma J. Blood levels of long-chain n-3 fatty mentation with omega-3 fatty acids and low-dose aspirin. acids and the risk of sudden death. N Engl J Med 2002;346: J Periodontol 2010;81:1635-1643. 1113-1118. 70. Shibutani T, Inuduka A, Horiki I, Luan Q, Iwayama Y. 59. GISSI-Prevenzione Investigators. Dietary supplementa- Bisphosphonate inhibits alveolar bone resorption in experi- tion with n-3 polyunsaturated fatty acids and vitamin E mentally-induced peri-implantitis in dogs. Clin Oral Implants after myocardial infarction: results of the GISSI-Prevenzione Res 2001;12:109-114. trial. Gruppo Italiano per lo Studio della Sopravvivenza 71. Hasturk H, Kantarci A, Ebrahimi N, Andry C, Holick M, Jones nell’Infarto miocardico. Lancet 1999;354:447-455. VL, Van Dyke TE. Topical H2 antagonist prevents periodon- 60. Van Dyke TE. The management of inflammation in peri- titis in a rabbit model. Infec Immun 2006;4:2402-2414. odontal disease. J Periodontol 2008;79:1601-1608. 72. Zizzi A, Aspriello SD, Rubini C, Goteri G. Peri-implant 61. Hasturk H, Kantarci A, Ohira T, Arita M, Ebrahimi N, Chiang diseases and host inflammatory response involving mast N, Petasis NA, Levy BD, Serhan CN, Van Dyke TE. RvE1 cells: a review. Int J Immunopathol Pharmacol 2011;24: protects from local inflammation and osteoclast-mediated 557-566. bone destruction in periodontitis. Faseb J 2006;20:401-403. 73. Spite M, Norling LV, Summers L, Yang R, Cooper D, Petasis 62. Arita M, Yoshida M, Hong S, Tjonahen E, Glickman JN, NA, Flower RJ, Perretti M, Serhan CN. Resolvin D2 is a Petasis NA, Blumberg RS, Serhan CN. Resolvin E1, an potent regulator of leukocytes and controls microbial sepsis. endogenous lipid mediator derived from omega-3 eicosa- Nature 2009;461:1287-1291.

Eur J Oral Implantol 2012;5(Suppl):S51–S60