Peri-implant Indices of Remodeling as a Response to Mechanical Loading

THESIS

Presented in Partial Fulfillment of the Requirements for the Degree Master of Science in the Graduate School of The Ohio State University

By

Michael Lynn Gurney

Graduate Program in Dentistry

The Ohio State University

2012

Dissertation Committee:

Dr. Binnaz Leblebicioglu D.D.S. PhD “Advisor”

Dr. Edwin McGlumphy D.D.S. M.S.

Dr. John Walters D.D.S. M.Med.Sci.

i

Copyright by Michael Lynn Gurney 2012

i

Abstract

Background: The current study aims to detect peri-implant tissue changes that may occur due to bone and soft tissue remodeling as a response to mechanical loading.

Material and methods: Systemically healthy, non-smoking adults with a single posterior edentulous space with intact proximal natural teeth were recruited. Inclusion criteria were: adult patient, systemic health, periodontal health, single tooth posterior edentulous space, non- smoking, no grafting needed, and ability to consent to treatment. Exclusion criteria included presence of parafunctional habits and implants placed with significant angle precluding screw retained implant crowns. Peri-implant tissue changes were recorded prior to crown delivery and 1, 3, 6 and 12 weeks in function. Clinical parameters were: plaque and gingival indices (PI and GI), Resonance Frequency Analysis (RFA), and crevicular fluid (CF) samples of the implant and adjacent teeth. CF volume was determined using a calibrated electronic volume quantification unit. Samples were stored at -80C. Multiplex assay was performed for cytokine detection. Polymerase Chain Reaction was used to detect the presence of eight putative microflora. Data was analyzed using multiple model regression analysis corrected with

Bonferroni adjustment. Correlation analysis was conducted by calculating Spearman correlation coefficient, statistical significance was accepted at p<0.05.

ii

Results: Sixteen patients were recruited and fifteen completed the study. Mean healing time prior to loading was 6.1±0.3 months. Implant Stability Quotient (ISQ) increased from

76.5±1.7 to 83.7±1.7 units (P<0.001). Peri-implant plaque accumulation was statistically lower than adjacent periodontal plaque accumulation at baseline (P=0.02) and this difference between groups stayed significant throughout the observation weeks (P<0.004). Peri-implant gingival index was statistically lower compared to adjacent periodontal gingival index before loading and at 1 week following loading (P<0.05). No difference between implant and adjacent tooth sites for GI at follow-up appointments (P>0.05). Peri-implant GI increased observed by 12 week was statistically significant compared to baseline (P=0). Gingival crevicular fluid (GCF) was statistically higher than peri-implant crevicular fluid (PICF) prior to loading (P=0.03). PICF increased following loading and was significantly greater than GCF at 1, 3, 6 and 12 weeks

(P<0.04). Post-loading peri-implant crevicular fluid was statistically higher than pre-loading crevicular fluid when compared to baseline value (P=0.004). No significant differences in periodontal and peri-implant probing depths or the keratinized tissue width throughout the study (P>0.05).

Cytokines: PDGF, IL-4, IL-6, IL-7, IL-10, IL-12, IL-17, Eotaxin, FGFb, IFN and TNFα were expressed at detectable but low levels at both PICF and GCF samples with no significant differences throughout the observation period. IL-1, IL-8, G-CSF, IP-10 and VEGF presented higher expression levels in gingival crevicular fluid samples compared to PICF samples prior to loading with no statistically significant difference. Following one week of loading, there was approximately a two fold increase in peri-implant crevicular fluid IL-1, IL-8 and MIP-1b levels although the differences were not statistically significant. VEGF and MCP-1 levels showed a iii trend towards increased expression at one week (P>0.05). Other cytokines, G-CSF and IP-10, were expressed at higher levels at both PICF and GCF samples following 3 weeks of loading compared to before loading values. The differences were not statistically significant (P>0.05).

Microbiology data showed the presence of certain putative periodontal pathogens throughout observation period in a transient fashion.

Conclusion: Within the limits of this study, it appears that mechanical loading following a traditional osseointegration period has significant effect on implant stability and peri-implant clinical indices used to evaluate soft tissue healing and health. However, the expression of pro- inflammatory cytokines and the presence of certain putative periodontal pathogens show significant variations among individuals.

iv

Dedication

This document is dedicated to my family. They had to sacrifice a lot to allow me to complete this project.

v

Acknowledgments

The greatest of thanks and appreciation for her guidance and mentorship needs to be given to my wonderful advisor Dr. Binnaz Leblebicioglu D.D.S. PhD of The Ohio State

University College of Dentistry Graduate Periodontics Program. Thank you!

I appreciate and offer my sincere thanks the help from my thesis committee, Dr. Edwin

McGlumphy D.D.S M.S. of The Ohio State University College of Dentistry Implant

Clinic and Dr. John Walters D.D.S. M.Med.Sci. of The Ohio State University College of

Dentistry Periodontics Department.

Special thanks are in need for the help of Dr. Julie Holloway D.D.S. M.S., now at the

University Of Iowa College Of Dentistry, for input with the study protocols.

Thanks to Dr. Tim Eubank PhD of the Ohio State University College of Medicine for helping with the Multiplex assays.

Thanks to Dr. Purnima Kumar D.D.S. PhD of The Ohio State University College of

Dentistry Periodontics Department for her help with the bacterial PCR detection procedures.

vi

Thanks to Mr. Vedat Yildiz M.S. from the Ohio State University Biostatistics

Department for the statistical analysis.

A special thanks needs to be extended to the graduate residents of the Graduate

Periodontics Residency program for the recruitment of patients and their help with surgical placement of dental implant.

Thanks are in need for the faculty and staff of the Graduate Prosthodontics Residency

Clinic for their help and support throughout the treatment period of the patients in the study.

This research project was funded by a grant through NIDCR: NIDCR R21DE018718 to

Dr. Leblebicioglu

vii

Vita

June 1998 ...... Emmett Senior High School

April 2005 ...... B.S. Biology, BYU – Idaho

May 2009 ...... D.D.S. Creighton University

2009 to Present ...... Graduate Prosthodontics Resident –

...... The Ohio State University College of Dentistry

Fields of Study

Major Field: Dentistry

Publications

AADR Abstract – 2012

Gurney M; Eubank T, Holloway JA, McGlumphy E, Walters J, Leblebicioglu B. Peri- implant indecies of Remodeling as a Response to Mechanical Loading. Abstract #168,

AADR 2012

viii

Table of Contents

Peri-implant Indices of Remodeling as a Response to Mechanical Loading ...... i

THESIS ...... i

Abstract ...... ii

Dedication ...... v

Acknowledgments ...... vi

Vita ...... viii

Fields of Study ...... viii

Publications ...... viii

Table of Contents ...... ix

List of Figures ...... xiii

CHAPTER 1: INTRODUCTION ...... 1

Wound healing during early phases of osseointegration ...... 2

Bone Response to Load ...... 4

Clinical parameters used to evaluate peri-implant tissue health ...... 6

CHAPTER 2: MATERIAL AND METHODS ...... 15

Study Population Study Design ...... 15

ix

Initial Comprehensive Oral Exam and Details on Surgical /Restorative Phases of Treatment

...... 16

Diagnostic Planning and Surgical Guide Fabrication ...... 17

Surgical Implant Placement ...... 18

Restoration of Implant ...... 19

Peri-Implant Data Collection ...... 22

Plaque and Gingival Indices ...... 22

Probing Depth and Keratinized Tissue ...... 23

Implant Stability Quotient Readings ...... 25

Laboratory Analysis of the Crevicular Fluid ...... 26

Statistical Analysis ...... 28

CHAPTER III: RESULTS ...... 30

Study population...... 30

Clinical measurements ...... 30

Gingival Crevicular Fluid (GCF) ...... 32

Resonance Frequency Analysis ...... 33

Crevicular Fluid Content Prior to and Post-loading ...... 33

Microbiological Findings ...... 36

Correlations Between Clinical Parameters and Biomolecules ...... 37

CHAPTER IV: DISCUSSION ...... 61 x

References ...... 77

xi

List of Tables

Table 1: Study Timeline/Outline ...... 38

Table 2: Description Demographics ...... 40

Table 3: Pro-inflammatory Biomarkers Distribution in PICF and GCF...... 41

Table 4: Spearman Correlation for Clinical Parameters at Implant and Tooth Sites ...... 42

Table 5: Spearman Correlation Analysis for Cytokine Data at Implant Sites ...... 42

Table 6: Spearman Correlation between Clinical Parameters and Cytokines at Implant

Sites ...... 43

Table 7: Pro-inflammatory Biomolecules ...... 44

xii

List of Figures

Figure 1: Surgical Guide on Diagnostic Cast ...... 45

Figure 2: Plaque Accumulation Prior To and Post Crown Delivery ...... 46

Figure 3: Soft Tissue Characteristics Prior to and Post Crown Delivery ...... 47

Figure 4: The amount of Keratinized Tissue prior to and post loading ...... 48

Figure 5: Changes in peri-implant and adjacent periodontal probing depths prior to and following loading ...... 49

Figure 6: Peri-implant and Periodontal Crevicular Fluid Volume Changes prior to and following loading ...... 50

Figure 7: Resonance Frequency Analysis prior to and following loading ...... 51

Figure 8: PICF and GCF IL-1β total amount prior to and following loading ...... 52

Figure 9: PICF and GCF IL-1ra Total Amount Prior to and Following Loading...... 53

Figure 10: PICF and GCF IL-8 Total Amount Prior to and Following Loading ...... 54

Figure 11: PICF and GCF G-CSF Total Amount Prior to and Following Loading ...... 55

Figure 12: PICF and GCF IP-10 Total Amount Prior to and Following Loading ...... 56

Figure 13: PICF and GCF MCP-1 Total Amount Prior to and Following Loading ...... 57

Figure 14: PICF and GCF MIP-1b Total Amount Prior to and Following Loading ...... 58

Figure 15: PICF and GCF VEGF Total Amount Prior to and Following Loading ...... 59

Figure 16: Microflora Formation – Implant and Tooth ...... 60

xiii

CHAPTER 1: INTRODUCTION

Successful outcome of implant placement is dependent on deposition of bone on the implant surface during the healing stage. Although not well documented, early implant failures have been reported to occur at rates ranging from 3 to 20 percent. [1] [2]

[3] Failed implants have detrimental effects on the patient’s comfort, esthetics, and oral health. Dental implant therapy is a time consuming, and financially demanding process.

Therefore, it is imperative that the initial implant placement outcome is a success.

Osseointegrated implants have been reported to have positive long-term clinical results.

[2] [3] Implant stability is a key characteristic of successfully osseointegrated implants.

[1] [4] [5]

Implant placement requires surgical manipulation of the jaw bone; accomplished with an osteoectomy procedure resulting in trauma to the bone. Osseointegration then begins with an acute inflammatory response to the surgical trauma. A blood clot forms between newly sectioned bone and newly introduced implant device. The implant surface functions as a scaffold controlling cell migration. Inflammation is a necessary part of the early phase of wound healing. Previous work (Emecen-Huja MS thesis 2011) demonstrates a fast response to injury and physiologically resolves within one to two weeks of post-operative healing in humans. The current study aimed to determine whether similar peri-implant inflammatory responses are detectable following loading.

1

Changes in implant stability following occlusal loading were also investigated by using resonance frequency analysis (RFA).

Wound healing during early phases of osseointegration

In order to achieve osseointegration [6], bone and soft tissue surrounding a newly placed dental implant device must heal properly. Peri-implant bone healing follows the general bone healing cascade of events which includes: hemorrhage, clot formation, an immunologic response, immature or woven bone formation, and remodeling [7]. This process results in the removal and repair of surgically damaged bone and subsequent osseointegration of a dental implant. Similarly, soft tissue healing follows the general sequence of: trauma, hemorrhage, clot formation, cellular activation, granulation tissue formation, scar formation, and final connective tissue remodeling [7]. The final product is a rigidly bound bone-to-implant surface and a mucosal seal. The process of osseointegration of dental implants is divided into early and late phases. The early phase includes healing around an implant immediately following its surgical placement until loading. The late phase includes healing or remodeling that occurs after the implant is loaded.

Davies [8] divided early phases of wound healing at bone level into three stages: osteoconduction, de novo bone formation, and bone remodeling. The de novo bone formation around a newly placed implant can be divided into two categories: contact osteogenesis (implant surface apposition of bone) and distant osteogenesis (bony surface apposition of bone). Oral mucosal epithelium healing follows the same basic wound healing mechanism as epidermal tissue. As described by Wikesjo, [9] injury to the 2 mucosal epithelium results in bleeding. Within seconds, plasma proteins and tissue proteins are released into the wound, initiating the coagulation cascade. Minutes later a fibrin clot is formed providing wound hemostasis. Phagocytic cells (neutrophils dominate early and then macrophage in the later stages) begin to debride the area of damaged tissue and foreign material. The presence of phagocytic cells stimulates cell activation and migration to the wound area. Epithelial cells begin to migrate across the top of the fibrin network and within three to four days they seal the wound from the oral environment. Within the first two to three days following tissue injury, fibroblasts begin producing a disorganized collagen network and endothelial cells begin capillary formation. As a result, granulation tissue is formed. This granulation tissue undergoes a maturation and remodeling process culminating with the formation of a scar and eventually normal oral mucosa. Dental implants placed following a one-stage (non- submerged) surgical protocol allow the investigation of early phases of peri-implant wound healing during osseointegration by creating a clinical study model. [9]

Soft tissue around an unloaded one stage implant, when fully healed, exhibits similar characteristics as a natural tooth periodontal mucosa. Healthy peri-implant soft tissue characteristics include: keratinized gingiva, 0.23 to 0.65 mm of non-keratinized sulcular epithelium, 1.16 to 1.71 mm of junctional epithelium, and 0.79 to 1.39 mm of collagen rich (scar like) connective tissue. Together these components comprise the peri- implant biologic width, about 2.84 to 3.01 mm [10] [11] [12] [13]. It’s important to note that the peri-implant junctional epithelium is attached to the implant surface via hemidesmoses, but not to the extent as seen with natural teeth [14] [15] [16] and the

3 collagen fibers of the connective tissue layer run parallel to the implant surface with no apparent connection to the implant surface [17].

Bone Response to Load

Bone is a dynamic tissue, meaning bone is continually changing or remodeling in response to the forces it receives. In the early 1800’s Bougery, and Bell separately documented the unique internal structure of long bones. [18] [19] Their concept of bone architecture were confirmed and reported independently by von Meyer. [19] [18] [20]

Later in the 1800’s an engineer named Culmann related the boney architecture of a femoral head, to the trajectories used to create the “Culmann Crane.” [19] [21] Julius

Wolff added to his predecessors work and presented the laws of bone remodeling which is currently known as “Wolff’s Law” or “The law of transformation of the bone.” [18]

[19] [22] [23] Later Roux added the principle of functional adaptation. [18] [19] [24] In summary, bone responds to stress and strain by modeling and remodeling its boney architecture.

Bone adjacent to a dental implant’s surface may experience bone remodeling in response to forces generated during function [18]. Stanford and Brand [18] review the concepts of bone remodeling around dental implants. The bone remodeling process appears to exhibit an “on” and “off” response, known as the mechanostat theory. [25]

Frost presented the mechanostat theory as part of bone remodeling mechanism. [25] His theory states that activation or deactivation of the bone remodeling process is dependent on the amount and frequency of force applied to the bone. This theory in turn suggests that there is a minimal strain level required in order to initiate bone remodeling. Frost

4 claims that strain levels of 50 - 200 μ-strains result in disuse atrophy of bone. [25] Also,

2000 μ-strains is the “minimum effective strain” to induce bone remodeling. [26] It is thought that mechanostat mechanism is controlled in part by the process of mechanotransduction. [18]

Mechanotransduction is the process by which bone translates mechanical stress into a biologic response of remodeling or modeling [18]. Stress on bone results in an increase in intra-boney fluid pressure and movement. The resultant fluid changes are recognized by osteocytes. In turn, osteocytes respond by producing bone remodeling mediators which regulate activation and coupling of osteoclastic and osteoblastic activity, also known as the basic multicellular units (BMUs) of bone. [27] [28] [29] [30] [31] [32]

[33] The induction of newly forming bone is closely coupled to osteoclast activity, which allows bone to remodel as an adaptation to the forces which it receives. [28] [29]

Occlusal forces applied to an implant restoration are transferred to the bone via the implant-bone interface and affect bone in varying degrees of strain. [18] [34] [35]

[36] [37] An implant and its restorative components act as a single entity if the screw preload and passivity of the restoration are adequate. [38] In vitro studies, using finite elemental analysis and photoelastic studies, predict an even distribution of force into the adjacent peri-bone when an axial load is applied to a vertically positioned implant. The resultant peri-bone strains are greatest in the coronal bone adjacent to an implant followed by apical peri-implant bone. Bone strain distribution between the implant-bone interface is affected by a variety of factors. Some of these factors include: quantity of bone, quality of the bone, bone height, implants’ spatial location within bone, the implant

5 size, and the implant angulation. [34] [35] [36] [37] [39] [40] [41] Peri-implant bone strain as a result of occlusal load induces peri-implant bone remodeling. Animal studies indicate that loaded implants result in greater peri-implant bone density. [42] Static load studies report rearrangement of osteocytes adjacent to loaded implants [43], greater remodeling in peri-implant bone with loaded implants [43], cancellous bone trabeculae remodeling based on strain patterns [44], and greater bone density during first weeks of loading especially in the coronal part of the implant site. [45] [46] [47] Also, Turner, using a rat model, demonstrated that bone remodeling response increases by increasing the amount and the frequency of mechanical stress. [27]

Peri-implant soft tissue characteristics after loading remain similar to natural periodontal tissues and unloaded peri-implant structures. [12] [48] [49] Others report slight changes in the junctional and sulcular epithelial dimensions, suggesting an increasing in length of the junctional epithelium [12].

Clinical parameters used to evaluate peri-implant tissue health

Implant osseointegration success has always been the primary goal of dental implant therapy. Success and failure criteria have been significantly discussed and debated. [1] [4] [5] At present, many schools of thought have accepted the success and failure criteria developed by Albektsson and Zarb. [4] [5] However, in specific cases, implants may clinically present with characteristics somewhere in between the

“successful” and “failed” criteria. These have been described as failing or ailing

6 implants. [1] Monitoring and determining peri-implant disease progression is an active area of research.

Conventional methods used for evaluation of peri-implant health include peri- implant probe depths (PD), some form of recording for gingival inflammation, plaque accumulation, radiographic evaluation, peri-implant crevicular fluid (PICF) evaluation, and recording implant stability. [1] [5] All methods are helpful but possess inherent limitations especially regarding implant health prognostication. [1] However, PICF characterization and monitoring implant stability with resonance frequency analysis

(RFA) have shown some promise. The following describes peri-implant probe depths, gingival bleeding, radiographic evaluation, implant stability using periotest and RFA, and crevicular fluid as possible diagnostic tools.

Peri-implant Probe Depths

Peri-implant probe depths when accurately recorded over time are helpful in monitoring peri-implant inflammation and heath. [1] However, due to inherent differences in peri-implant gingival tissues [50] [51], difficulty in controlling probing angulation and forces [52] [53], and varying soft tissue factors [1] limits its use in accurately determining peri-implant health and prognostication.

Peri-implant Gingival Bleeding

Gingival bleeding or inflammation indices are used to monitor implant health.

Bleeding has been correlated with peri-implant disease. [1] [51] Conversely, peri-

7 implant bleeding on probing has been reported to occur with healthy and diseased peri- implant gingival tissues. [50] [51] [54]

Radiographic Evaluation

Radiographic evaluation is the most widely used method of clinically monitoring peri-implant bone support [55] [56] [57] Limited resolution and lack of three dimensional analyses prevent the panoramic radiograph from being used for fine measurements. [57] Peri-apical radiographs obtained with proper paralleling technique in conjunction with opaque spheres (BBs) of know dimension for measurement correction provide a minimally distorted fine detail analysis of peri-implant bone. They are often used to measure crestal bone height changes over time. [55] [57] Some limitations exist with peri-apical radiography to evaluate peri-implant bone. These factors include: human error in reading the radiograph, angulation repeatability, two dimensional view of the implant, and limited resolution. [1] Cone beam CT is the most accurate method of evaluating integration of dental implants. It provides a better one to one scale and a three dimensional representation of the implant site. However, the higher exposure to radiation, compared to conventional dental radiography, and increased cost limit their routine usage as a diagnostic tools [58] [59].

Implant Stability

8

Non-invasive techniques for evaluating implant stability include Periotest®a and resonance frequency analysis. Periotest® utilizes mechanical tapping of a tooth or implant. It is designed to quantify the dampening effect of a periodontal ligament [60].

When placed over an implant surface and activated, it provides 16 consecutive taps at a controlled force to the implant surface. Stability determination is based upon the duration of each tap on the implant surface. This recording is given a numeric value, the manufacture reports a range of -8 to 50. A smaller number signifies greater stability. It is reported that implant stability recording produce numeric values in the range of -5 to 5.

[61] [62]. If an implant exhibits any movement within the bone, the contact duration of each Periotest® tap is increased and results in a larger numeric value. Due to relatively poor sensitivity, poor resolution, and its susceptibility to operator error [63] [62] new technology has been developed.

Resonance frequency analysis (RFA) was introduced as a possible method to determine implant stability by Meredith during the 1990’s. [64] The concept is based on recording the resultant vibration/movement of a transducer attached to an implant when stimulated. Originally, an L-shaped transducer was attached to an implant and excited with a known frequency. The L-shaped transducer was attached to a frequency analyzer which recorded the transducers vibration response to the stimulation. Recordings were in

Hz; the more rigid the implant-bone interface was the larger the Hz recording. [63] This was interpreted as an increase in implant stability. This technology was mainly used for research. More recently, RFA was developed into a clinical device known as the

a Periotest Medizintechnik Gulden e. K.. Eschenweg 3, 64397 Modautal, Germany 9

Osstell®b. The Osstell® system utilizes a stimulator, a Smart Peg®c, and recording device. The Smart Peg® consists of an aluminum peg with a screw at one end and a cylindrical magnet on the other end. To use, one first attaches the Smart Peg to an implant, then places the Osstell® wand within 2 mm of the magnet to activate the system.

The wand stimulates the magnet on the Smart Peg with a magnetic pulse. The magnet vibrates and the Osstell® wand records the degree of vibration in Hz. The Osstell® unit then converts the Hz recording into a numeric value of 1-100 which is referred to as the implant stability quotient (ISQ). The larger the ISQ value recorded indicates the more stable the implant-bone interface. [65]

RFA has been used extensively to evaluate implant stability during post implant surgery healing. [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] A proposed acceptable

ISQ reading at the time of implant placement is generally higher than 50 ISQ. [68] [69]

[74] A decrease in ISQ recordings is to be expected during the third and fourth weeks following implant surgery. By week six, original ISQ readings are normally re- established. [69] [71] [73] [74] [75] Early implant failure ISQ values appear to decrease during the first two weeks post implant surgery and continue to decrease in the subsequent weeks. ISQ readings in the 40s at weeks 2-4 have been reported on early failure implants. [69] [70] [72] [73] [74] [75] However, some implants that exhibit lower ISQ recordings which are left untouched for a period of time may recover and become successful. [75] Factors that affect initial ISQ recordings include: amount of cortical bone present, method of site preparation, amount of implant above crestal bone,

b Ostell Mentor® and Smart Peg® Ostell AB, Gamlestadsvägen 3B, SE 415 02 Göteborg, Sweden c Ostell Mentor® and Smart Peg® Ostell AB, Gamlestadsvägen 3B, SE 415 02 Göteborg, Sweden 10 and implant shape. [62] [68] [70] [72] [74] ISQ is not affected by implant length. [62]

[74] Implants placed in higher quality bone, rich in cortical bone, present less variation in ISQ readings during healing when compared to implants placed in lesser quality bone.

[62] [70] [72] ISQ changes as a result of mechanical loading of the implant have been investigated in limited number of studies. [72] [73] [74] [75] Studies indicate that there might be slight increases in ISQ readings once an implant is loaded. Further research is needed to better correlate ISQ and loaded implants.

Peri-implant Crevicular Fluid (PICF)

Gingival crevicular fluid (GCF) arises, in healthy gingival tissues, from the junctional-epithelium cells of the gingival attachment apparatus. Alfano described crevicular fluid as: osmotically-modulated, pre-inflammatory exudates or transudate pulled from the cells via an osmotic gradient created by the bacterial by-products within the gingival sulcus. [76] GCF flow rate in healthy gingiva, as described by Goodson, is three to eight µl per hour. [77] GCF possesses similar characteristics as plasma, interstitial fluid, and serum [78] [79] [80] with minor modifications; such as greater concentrations of sodium and potassium. [78] [81]

When bacterial presence increases, resultant by-products will penetrate the junctional epithelium and underlying basement membrane via inter-cellular channels.

These by-products result in an inflammatory response. A resultant gingival fluid exudate is generated from the capillary plexus subadjacent to the gingival basement membrane.

The exudate passes through the gingival basement membrane and through enlarged inter-

11 cellular channels of the junctional epithelium and into the gingival crevice.

Subsequently, GCF volume and flow rate increase exponentially, 10 to 70 times, compared to that of healthy non-inflamed gingival tissue. [76] [82] [77]

Healthy GCF contains a variety of molecular components and cells. Contents include host derived cells, biologic molecules, bacteria, and bacterial by-products.

Sixty-five to 100 different components of GCF have been evaluated as markers to monitor inflammation and disease of the periodontal apparatus. [83] [84] These markers can be classified as host-derived enzymes, inflammatory mediators, and tissue- breakdown products. [83] Many markers have been shown to increase in the presence of inflammation and periodontal disease with a subsequent decrease after receiving periodontal therapy. Their presence may be affected by genetics, pocket depth, extent of inflammation or periodontal disease. [85] [86] [87] [84] Notable pro-inflammatory markers include: prostaglandin E2 (PGE2), elastase, lysosomal β-glucoronidase, alkaline phosphatase, cathsepsin β, collagenases or metal metalloproteinases (MMPs), interleukin-

8 (IL-8), interleukin-6 (IL-6), tumor necrosis factor-alpha (TNF-α), osteocalcin, proteoglycans, fibronectin, epidermal growth factor (EGF), transforming growth factor- alpha (TGF-α), immunoglobulin-gamma (IgG), immunoglobulin-alpha (IgA) complement components, and others. [82] [88] [89] [90] [91] However, it is difficult to use the markers to predict the potential progression of periodontal disease. [83] [91]

Peri-implant crevicular fluid (PICF) volume and composition of healthy implants is similar to natural teeth. [92] [93] [94] Sulcular depth, implant size, and one stage placement surgeries may increase the fluid volume. [95] [96] PICF volume increases

12 during implant loading and may remain elevated without increases in gingival inflammation. [96] [97] A significant increase in PICF volume occurs in response to peri-implant inflammation. [95] [98] [99] PICF fluid volume decreases subsequent to peri-implant therapy. [100] It is thought that PICF volume can be used to help identify peri-implant inflammation.

Nowzari [101] [102] observed similarities in PICF composition when compared to GCF of natural teeth. However, he noted a general increase in pro-inflammatory cytokine concentrations in PICF; suggesting possible implant ion release into the peri- implant tissues may stimulate small changes in cytokine composition. Similar to natural teeth, PICF contains many biological molecules. Possible biological markers include: pro-inflammatory cytokines, enzymes, and catabolic metabolites. (Table 7) These molecules may provide a means to better monitor implant health and disease progression.

Pro-inflammatory molecules found and studied in the dental literature include: interleukin-1β (IL-1β), tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6), interleukin-

8 (IL-8), and prostaglandin E-2 (PGE2). Some of the PICF enzymes present in the dental literature include: matrix metalloproteinases (MMPs), neutral protease (NP), neutrophil elastase (NE), myeloperoxidase, β-glucoronidase, cathepsin, and elastase. Also, several extracellular degradation products have been studied in the dental literature, including: chondroitin-4-sulfate, hyaluronic acid, and carboxytermial telopeptides. [82-112]

The knowledge on remodeling around dental implants following mechanical loading is limited in the dental literature. The purpose of the study was to determine peri-implant characteristics as a response to loading.

13

The working hypothesis states that dental implant loading following conventional osseointegration period affects implant stability, crevicular fluid volume and content.

The specific aims include:

1- Identify peri-implant soft tissue clinical changes following implant loading.

2- Determine crevicular fluid volume changes with respect to implant loading.

3- Determine if RFA can detect changes in implant stability in response to

implant loading.

4- Detect presence of pro-inflammatory wound healing mediators within peri-

implant crevivular fluid prior to and following implant loading.

The results of this study may help to better understand possible clinical and biological markers of remodeling that occur in peri-implant tissues following mechanical loading of a single osseointegrated dental implant. The study may be the first to investigate soft tissue changes during early phase of remodeling immediately following mechanical loading by using a clinical prospective observational study model.

14

CHAPTER 2: MATERIAL AND METHODS

Study Population Study Design

The study design was a prospective longitudinal observational cohort study.

Subjects were recruited from the Graduate Periodontal Clinic at The Ohio State

University between November 2010 and December 2011. The study protocol and informed consent forms were approved by the Institutional Review Board of the Ohio

State University. A written consent form was obtained from all participants before entry to study (OSU IRB Protocol #2007H0261). Recruited patient presented with a need for a one stage single tooth implant placement surgery in a posterior sextant. When a patient met the inclusion criteria he/she received a comprehensive dental exam in the Advanced

Prosthodontic Residency Clinic at The Ohio State University.

The inclusion criteria included: 1) adult patients; 2) single edentulous posterior space with adjacent natural teeth; 3) non-smokers; 4) periodontal and systemic health; 5) possess the ability and were willing to provide informed consent for the study protocol; and 6) surgical placement allowed for a screw retained implant restoration. Exclusion criteria were: 1) patients with untreated periodontal disease; 2) patients with other dental restorative needs including the presence of broken/decayed teeth; 3) implants placed with two-stage protocol; 4) grafted implant sites; 5) patients with major occlusal deficiencies;

6) patients with parafunctional habits; 7) patients with systemic diseases affecting their 15 oral health, eg; uncontrolled diabetes; use of NSAIDs; 8) pregnant women; 9) unable or unwilling to provide informed consent form. Exit criteria were: 1) voluntary withdrawal;

2) non-compliance with study protocol; 3) development of systemic or oral disease that are in the exclusion criteria and 4) implant placement precluding a screw retained restoration. The study design consisted total of nine visits. Figure 1 is a flow chart presenting different phases of the study protocol.

Initial Comprehensive Oral Exam and Details on Surgical /Restorative Phases of Treatment

At the initial visit with the Advanced Prosthodontic Residency Clinic, a comprehensive dental exam was performed; consisting of an extra- and intra-oral head and neck soft tissue examination, TMJ examination, and dental examination. The dental examination included appropriate radiographs, occlusal evaluation, individual tooth evaluation, photos, and diagnostic records. Diagnostic records included maxillary and mandibular alginate d impressions with stock rim lock metal trays, three leaf gauge e centric relation recordsf, and a facebow recordg with polyvinyl siloxaneh. Casts were made using type III dental stone, duplicated with the same type of stone, and mounted in a semi-adjustable articulator using the previously mentioned records. Pathology and disease noted during the exam were explained to the patient and solutions were proposed.

dJeltrate® DENTSPLY International Inc. General Counsel Office World Headquarters Susquehanna Commerce Center 221 West Philadelphia Street, York, PA 17405-0872 e Leaf Gauge Panadent® 580 S. Rancho Ave. Colton, CA 92324 f Aluwax™ ALUWAX DENTAL PRODUCTS COMPANY P.O. Box 87, Allendale, MI 49401 g Slidematic Facbow® Whip Mix Corporation 361 Farmington Avenue P.O. Box 17183, Louisville, KY 40217 h Memoreg®2 Heraeus Kulzer 300 Heraeus Way, South Bend, IN 46614 16

Each patient was given the opportunity to return to their general dentist to address the noted concerns, if they so desired.

Diagnostic Planning and Surgical Guide Fabrication

Duplicate casts were obtained; two were used to fabricate surgical guides (Figure

1); one set was mounted in a semi-adjustable articulator and used for a full contour diagnostic waxing of the missing tooth. The tooth was waxed to ideal contour and articulation; duplicated in stone via an alginate impression and dental stone. The duplicate cast was used as a template to fabricate the implant surgical guides. A clear ethylene-vinyl copolymer 0.080 inch thermoplastic materiali was molded over the waxed tooth duplicate cast using a vacuum former machinej providing a clear matrix of the patient’s teeth. The matrix was trimmed using a heat knife to cement enamel junctions of two teeth mesial and distal of the proposed implant site. The matrix was carefully removed from the cast. The border extensions were lightly heated with a torch and placed on and off the casts repeatedly to ensure ease of removal and insertion during surgery. The borders where then smoothed and polished. This process was repeated to create two clear vacuum formed matrices. The wax tooth was then removed from the diagnostic model and the undercuts on the adjacent teeth were blocked out with light body addition polyvinyl siloxane.k The cast was then covered with a thin layer of

i 0.08 inch ethylene-vinul copolymer clear thermo-forming sheets Henry Schein Inc. 135 Duryea Road, Melville, NY 11747 j Vacuum forming machine 110V Henry Schein Inc. 135 Duryea Road, Melville, NY 11747 k Addition Polyvinyl Siloxane Affinis® Wash Coltene/Whaledent AG Feldwiesenstrasse 20, CH-9450 Altstätten 17 petrolatum. Each matrix was filled with clear methylmethacrylatel, confined to the location corresponding to the edentulous space; then fully seated onto diagnostic cast.

Any excess clear acrylic was immediately removed and the cast with the resin filled matrix was placed into a warm water pressure potm for 20 minutes at 20 psi). The matrices were then removed, trimmed, and polished.

Implant location was determined by using calipers to measure the edentulous space. The ideal location was identified and marked on the cast with a blue and red colored pencil. Lines, indicating the inclination of the adjacent teeth, were drawn on the teeth mesial and distal to the implant site. The cast was placed onto a lab drill pressn and surveyed into the correct angulation and position. The first matrix was placed onto the cast and a 3.0 mm guide hole was prepared through the surgical guide. The first matrix was removed and the second matrix was placed onto the cast. A 3.0 mm guide hole was prepared in the same location and angulation as the previous guide. The first surgical guide remained with a single access hole and was to be used as the pilot hole surgical guide. The lingual wall of the second guide’s access hole was removed to provide ease of access for the surgeon to the proposed implant site. Any flash acrylic was removed and the guide was polished.

Surgical Implant Placement

All implants were placed in the Graduate Periodontal Clinic at The Ohio State

University, by third year residents. Proper aseptic technique was followed and implants l Clear Ortho-jetTM methylmethacrylate Lang Dental, 175 Messner Drive, Wheeling, IL 60090 m AquapressTM 5 inch Lang Dental 175 Messner Drive, Wheeling, IL 60090 n Paraskop® M Multifunctional Milling Unit BEGO Bremer Goldschlägerei, Wilhelm-Herbst-Straße 1 D-28359 Bremen 18 were placed according to the implant manufacture’s protocol, using the surgical guide.

All implants following a one stage protocol. The criterion for one-stage implant placement was an initial ISQ value of 50 and/or a torque value of 35 Ncm. Minimum healing time was 3 months, based on anatomical location and traditionally accepted osseointegration period.

Restoration of Implant

All restorative procedures were performed in the Advanced Prosthodontic

Residency Clinic at the Ohio State University College of Dentistry by the same graduate prosthodontic resident. Crowns were fabricated by the following protocol:

1. Custom open tray polyvinyl siloxane implant impression

Custom Tray Fabrication: visible light curing dental acrylico was adapted over

the full arch with 4 mm internal relief provided by two layers of base plate

waxp. Three vertical stops, to provide proper impression material thickness,

were incorporated on non-functional cusps. All trays were cured following

the manufacture’s recommendations using a light curing unitq. Wax residue

was removed using a laboratory steam cleaner and all borders were rounded

and smoothed. The custom trays were placed back onto the patient’s

diagnostic cast and the location of the implant site was identified and marked

on the external surface of the tray directly above the implant site and a small 3

o Triad® VLC TruTray™ - Dentsply™, DENTSPLY International Inc. 570 West College Ave., York, PA 17405-0872 p TruWax® Base plate wax - Dentsply™ DENTSPLY International Inc. 570 West College Ave., York, PA 17405-0872 q Tiad® VLC Curing Unit - Dentsply™ International Inc. 570 West College Ave., York, PA 17405-0872 19

mm hole through the tray was performed. All trays then bench cured for a

minimum of 24 hours prior to use on a patient

2. Implant Impression:

The healing abutment was removed from the implant and an appropriate direct

implant transfer coping was inserted and hand tightened. A peri-apical

radiograph was obtained to verify complete seating of the transfer coping and

to evaluate the peri-implant bone density and level. The custom tray was

inserted to confirm access to the implant impression post screw through the

tray. The tray was again cleansed and dried. A thin layer of polyvinyl

siloxaine adhesiver was applied to the internal surface of the tray and allowed

to dry for 10 minutes. The patient’s mouth was dried with cotton and light

body vinyl polysioloxaines was syringed around the impression coping in the

patient’s mouth; simultaneously heavy body polyvinyl sioloxainet was

syringed into the custom tray. The tray was seated appropriately in the

patient’s mouth and the excess around the superior aspect of the screw was

wiped away. The impression was carefully held into place until the material

polymerization time was achieved. The transfer abutment screw was loosened

and the impression was removed. An implant lab analog was carefully

attached and hand tightened to the impression post in the impression. The

r Caulk® Tray Adhesive - Dentsply™ DENTSPLY International Inc. 570 West College Ave., York, PA 17405-0872 s Aquasil Ultra Light Body Regular Set Vinyl Polysiloxane Impression Material Aquasil® DENTSPLY Caulk, 38 West Clarke Ave, Milford, DE 19963-1805 t Aquasil Ultra Heavy Body Regular Set Vinyl Polysiloxane Impression Material Aquasil® DENTSPLY Caulk, 38 West Clarke Ave, Milford, DE 19963-1805 20

impression was then disinfected, the healing abutment was re-inserted, and the

appropriate shade was selected using the vita shade guide.

3. Crown Fabrication

Each implant impression was poured with low expansion die stoneu and

mounted using hand articulation with a cast of the opposing dentition that had

previously been mounted in a semi-adjustable. Casts were lightly equilibrated

on the articulator prior to crown fabrication. Appropriate cast-to waxing

abutments were ordered. The mounted case was then sent to a dental lab

(Crown Work Dental Laboratory – Eugene, OR) for fabrication of a screw

retained metal ceramic implant crown with no metal showing. The average

return time from the lab was approximately three and one-half weeks.

4. Crown Insertion:

The screw retained metal ceramic crowns were tried-in. Proximal contacts

were adjusted to lightly drag 8 μm shimstock foilv to ensure complete seating

of the crown. The articulation was then adjusted to be light contact in

maximum intercuspation. Then a periapical radiograph was obtained. The

implant crown was torqued using the manufacturers torque driver, 20 Ncm for

Zimmer®w implant crowns and 15 Ncm for Straumann®x crowns. Initial

insertion torque was lower than the manufactures recommendations knowing u Resin Rock Whip Mix Corporation 361 Farmington Avenue, Louisville, Kentucky, USA 40209 v Shimtock® Metal Foil Whaledent, Wash Coltene/Whaledent AG Feldwiesenstrasse 20, CH-9450 Altstätten w Zimmer Dental 1900 Aston Ave, Carlsbad, CA 92008-7308 x Straumann 60 Minuteman Road, Andover, MA 01810 21

they crowns would need multiple subsequent removals. The access hole was

then filled with a gauze strip and a visible light curing acrylic materialy. At 12

weeks post-loading, a new periapical radiograph was obtained of the implant,

with the crown in place, and crowns were torqued to 30 Ncm (Zimmer) and

35 Ncm (Straumann), respectively.

Peri-Implant Data Collection

Figure 1 presents an outline of the study timeline and peri-implant data collection.

Samples were obtained at baseline (immediately before crown insertion) and at one, three, six, and twelve weeks post insertion. Baseline samples were recorded prior to the insertion of the implant crown. Plaque index (PI), gingival inflammation index (GI), probe depths (PD), crevicular fluid samples, and resonance frequency analysis or ISQ measurements were recorded at each data collection appointment. Peri-implant radiographs were obtained at baseline and twelve weeks as well as attached gingiva measurements.

Plaque and Gingival Indices

At baseline and the post insertion/loading evaluation appointments of one, three, six, and twelve weeks plaque index (PI) and gingival index (GI) [110, 111] [112] for adjacent natural teeth and modified plaque index (mPI) and modified gingival index

(mGI) [113] for dental implant were measured and recorded.

y Fermit® Ivoclar Vivadent AG, Bendererstrasse 2, 9494 Schaan, Principality of Liechtenstein 22

The Silness and Loe plaque index [110] [111] [112] [114] is recorded by measuring the amount of plaque present on four surfaces of a tooth; mesial, distal, buccal, and lingual. Numeric values are assigned to each surface based on the plaque detected.

Based on this scoring system: 0 = no plaque is present, 1 = plaque is present at the free gingival margin, 2 = a moderate amount of plaque present within the gingival sulcus, and

3 = the presence of abundant plaque within the gingival sulcus. The four numbers are then added together and divided by four. The resultant number is the relative plaque score for that tooth. mPI is a modified version of PI for implants.

The Loe and Silness gingival inflammation index [111] [114] is recorded in a similar manner as the plaque index. Numeric values are assigned to each tooth surface analyzed using the following scale: 0 = normal gingiva; 1 = mild gingival inflammation, noted by slight change in color; 2 = moderate inflammation, redness, glazing, and bleeding on probing; 3 = severe inflammation, marked redness, edema, ulceration, spontaneous bleeding. The tooth surface values are added and divided by the total surfaces evaluated.

Modifications, suggested by Mombelli [113], regarding PI and GI detection around implants are related to implants lacking a physiologic gingival margin and CEJ relationship as seen with natural teeth.

Probing Depth and Keratinized Tissue

Probing depths were recorded on the distal buccal, mid-buccal, mesial buccal, mesial lingual, mid-lingual and distal-lingual on the implant and adjacent teeth. Probe

23 depths were recorded on natural teeth using a metal color coded probez and a plastic probeaa was used to record probe depths around the implant.

Keratinized tissue measurements were recorded on the buccal and lingual of the implant crown at baseline and at twelve weeks. The measurement was obtained by recording the distance from the free gingival margin to the mucogingival junction using a color coded metal probebb.

Gingival Crevicular and Peri-Implant Crevicular Fluid Sample Collection

Gingival crevicular fluid (GCF) samples and peri-implant crevicular fluid (PICF) samples were collected by using sterile paper stripscc. Crevicular fluid samples were obtained at baseline and at the subsequent one, three, six, and twelve weeks post-loading data collection appointments. Crevicular fluid was obtained using the following protocol:

1. Fluid Isolation

a. The implant sextant was rinsed and dried with suction and the air water

syringe. Cotton rolls were positioned in buccal vestibule adjacent to the

implant, in the buccal vestibules of the maxilla to block Stenson’s duct,

and sublingually. Gauze was also positioned between the tongue and the

sampling area. The area was again dried with gauze and kept dry with

the help of suction.

2. Crevicular Fluid Sample using Paper Strips

z Qulix™ Color Coded Probes Hu-Friedy Mfg Co. Inc aa Plastic Perio Wise Probe Premeir Dental Products bb Qulix™ Color Coded Probes Hu-Friedy Mfg Co. Inc cc PerioPaper Strips OraFlow Inc. 14 Threepond Road Smithtown, New York 11787 24

a. Paper strips were handled with sterile cotton forceps; they were only

touched on the orange waxed area. A sterile paper strip was inserted into

the gingival sulcus, just below the free gingival margin, of a tooth

adjacent to the implant location. Each paper strip remained in place for

20 seconds and was removed with cotton forceps and placed into a

calibrated electronic volume quantification unit (Periotron 8000®)dd. The

crevicular fluid volume was recorded and the paper strip was immediately

transferred into a properly labeled sterile cyrotube and closed to prevent

excessive evaporation; and then placed on ice. This process was repeated

until four crevicular fluid samples were obtained from each implant and

its adjacent teeth. (two from each tooth adjacent to implant) The same

process was followed to obtain four samples of the peri-implant crevicular

fluid. Samples were obtained from different locations within the gingival

sulcus. Any samples with visible blood contamination with blood or

saliva were discarded. All samples collected were immediately transferred

to a -20°C freezer and later transferred to a -80°C freezer for long term

storage.

Implant Stability Quotient Readings

Implant stability data were recorded at baseline and at one, three, six, and twelve weeks implant post-loading. Implant stability was determined using resonance frequency

dd Periotron Model 8000; OraFlow Inc. 14 Threepond Road Smithtown, New York 11787 25 analysis via the Osstell® ISQ deviceee. The following protocol was used to obtain ISQ data for each implant in the study:

1. After all the peri-implant data were collected, the healing abutment (at

baseline) or the crown (at one, three, six, and twelve weeks) were removed

and the appropriate size SmartPegTM ff was screwed into the implant and

hand tightened.

2. The Osstell® ISQ unit was activated and the wand was held within 2 mm

of the magnetic tip of the smart peg on the buccal side until a numeric ISQ

value was recorded. The wand was then placed on the lingual/palatal side

in a similar manner until a numeric ISQ value was recorded. This was

repeated two times.

3. Four ISQ values, two from the buccal and two from the lingual/palatal

were recorded.

Laboratory Analysis of the Crevicular Fluid

Extraction of crevicular fluid from paper strips

Extraction of crevicular fluid was performed as following: PCR microtubes with small holes at the bottom were prepared and sterilized. These tubes were placed into sterile 1.5 ml Eppendorf tubes. Paper strips were separated from waxed portion by using sterile scissor and were transferred from cyrotubes into PCR microtubes. Paper strips

ee Osstell® ISQ Ostell AB, Gamlestadsvägen 3B, SE 415 02 Göteborg, Sweden ff SmartPegTM Ostell AB, Gamlestadsvägen 3B, SE 415 02 Göteborg, Sweden 26 were soaked in 200 l cold PBS and incubated on ice for 15 minutes with occasional vortexing. Samples then were centrifuged at 13,000 rpm for 10 min. Supernatant was collected from the bottom of the tubes and centrifugation was repeated. Approximately

160 l elution volume was recovered from each tube. Forty l of this volume was loaded into multiplex bead-based assay plate. Remaining volume was used to detect presence of specific bacteria within newly forming peri-implant sulcus.

Multiplex bead-based assay

The multiplex bead-based assay gg was used to detect the presence and the concentration of human pro-inflammatory cytokines and growth factors within crevicular fluid samples including PDGFb, IL-1, IL-1ra, IL-4, IL-6, IL-7, IL-8, IL-10, IL-12, IL-

17, Eotaxin, FGFb, G-CSF, IFN, IP-10, MCP-1, MIP-1b, TNF- and VEGF. This technology uses digital signal processing capable of classifying polystyrine beads

(microspheres) dyed with distinct proportions of red and neared florophores. The spectral addresses for different bead populations where up to hundred different detection reactions can be performed in small volume of samples. The cytokine reagent kit includes assay buffer, antibody diluent, streptavidin-PE, and a 96-well filter plate. Specific kit hh designated as human cytokine group (Group I) was used for cytokines and growth factors.

Microbiology assay: gg Biplex Cytokine Assay, BioRad Laboratories Inc hh BioRad laboratories, Life Science Research, Hercules, CA 27

Targeted bacterial identification was performed by polymerase chain reaction

(PCR) using species specific primers in a nested amplification protocol. [115] Briefly, bacterial DNA was isolated by binding to glass beads and then eluted to Tris-EDTA buffer (pH 8). This community was used as a template to perform two round of PCR. The first round used universal bacterial primers targeted to 16S and 23Sribosomal genes. This initial amplification included part of the 16S gene as well as intergenic spacer region between the 16S and 23S genes. The second amplification used species specific forward primers along with a universal 23S reverse primer in a nested amplification using specific bacterial probes for the following known periodontal microbes (F. neviforme, G. hemolysans, T. forsythia, T. denticola, S. mitis, V. parvola, P. micros and C. gingivalis).

Specific periodontal pathogens were chosen based on a recent publication [115].

Statistical Analysis

The statistical analysis was completed by a statistical consultant at the OSU

Center of Biostatistics. Data were analyzed by using GraphPad Prism version 5ii and

SAS version 9.2jj. Mixed Model Repeated Measures ANOVA was used to analyze the data. Sandwich estimator was used to control the correlation that might exist in the data associated with the same subject. Baseline adjusted model were fitted to the data that compare two groups. T-test were performed within the ANOVA to compare each of the two groups at that time point and compare each post time point with the baseline in that particular group, where comparison was done by applying Bonferroni adjustment method

ii GraphPad Prism® GraphPad Software 2236 Avenida de la Playa La Jolla, CA 92037 USA jj SAS version 9.2® SAS SAS Institute Inc.100 SAS Campus Drive Cary, NC 27513-2414USA 28 to control the overall error rate. Spearman Correlation Coefficient was calculated to determine possible correlations between various clinical parameters and cytokines. The results were considered significant at P value ≤0.05.

29

CHAPTER III: RESULTS

Study population

Sixteen patients missing a posterior tooth with intact adjacent teeth were recruited for the study. Each patient underwent one-stage implant placement surgery. None of the sites were previously grafted. The mean healing time prior to loading was 6.1±0.3 months. Following the osseointegration period, all implants were restored with a screw retained metal ceramic crown. One subject was excluded during restorative phase of the treatment due to implant angulation precluding the use of a screw retained crown. One subject failed to complete the twelfth week appointment and one patient experienced an implant failure at twelfth week (maxillary premolar) appointment. The remaining thirteen subjects completed 12 week follow up study protocol.

Demographic data is presented on Table 2. The study population included seven female and eight male subjects with an overall mean age of 54±3 years. Four implants were maxillary and 11 mandibular. Four implants replaced a premolar teeth and 11 replaced molars. The mean implant diameter was 4.7±0.15 mm and the mean implant length was 12±0.3 mm. Early implant success was 100% while late implant success was

93.3% with one implant failure at 12 weeks post-loading (Table 2).

Clinical measurements

Plaque accumulation 30

Plaque accumulation on newly loaded implant and on adjacent teeth was recorded using PI (Figure 2). Peri-implant plaque accumulation was statistically lower than adjacent periodontal plaque accumulation prior to loading (0.3±0.13 and 0.8±0.2, respectively; P=0.02; median 0 (0-1.5) and 0.5 (0-2) respectively). The PI difference between an implant supported crown and the adjacent teeth remained statistically significant throughout the study (P<0.004). In general, plaque accumulation was well controlled with PI <1 for all surfaces, all observation periods (Figure 2).

Gingival Inflammation

Soft tissue characteristics were studied by recording mGI around implants and GI around adjacent teeth. Prior to loading mGI was statistically lower around implants compared to adjacent GI (Figure 3; 0.4±0.2 and 1±0.1, respectively; P<0.03); median 0

(0-1.8) and 0.9 (0.3-1.6) respectively. It remained lower at the first week following loading, (0.4±0.1 and 0.9±0.1, respectively; P<0.05); median 0.5 (0-1) and 1 (0-1.4) respectively. However, the GI difference between an implant supported crown and adjacent teeth diminished with time. The peri-implant GI approached similar levels as adjacent periodontal GI levels (Figure 3). In general, gingival inflammation was well controlled with GI<1 for all surfaces (Figure 3).

Keratinized Gingiva

Keratinized gingiva width measurements were recorded from the gingival margin to mucogingival junction at the mesial, mid-buccal (mid-lingual), and distal surfaces of

31 the implant site and reported as average measurements for the buccal and lingual surfaces. Before loading mean readings were 3.8±0.3 mm and 4.7±0.6 mm for buccal and lingual surfaces, respectively (Figure 4). Measurements at week 12 were 3.8±0.2 mm and 4.8±0.6 mm for buccal and lingual surfaces, respectively. There was no statistically significant differences between different surfaces and different observation periods (Figure 4; P>0.05).

Probing depth

Peri-implant and adjacent teeth periodontal probing depths were recorded prior to loading and at each time point (Figure 5). Pre-loading peri-implant probing depth were

2.7±0.2 mm for buccal surfaces and 2.6±0.2 mm for lingual surfaces. Similar values were obtained for adjacent periodontal readings prior to loading (2.3±0.13 mm for buccal tooth surfaces and 2.6±0.13 mm for lingual tooth surfaces). There was no statistical significant differences among these initial readings (Figure 5; P>0.05). A slight increase was noted with peri-implant probe depths by 12 weeks compared to before loading values

(Figure 5; 3.1±0.2 mm and 3.3±0.4 mm for buccal and lingual implant surfaces, respectively; P>0.05). Adjacent periodontal probing depths remained fairly stable throughout the study evaluation period (2.5±0.3 mm and 2.8±0.1 mm for buccal and lingual tooth surfaces, respectively).

Gingival Crevicular Fluid (GCF)

32

PICF volume was statistically lower than adjacent GCF volume prior to loading

(Figure 6; 0.7±0.1 ul and 1.1±0.1 ul, for implant and tooth sites, respectively; P=0.03).

As expected with crown delivery, PICF volume significantly increased following loading with each time value being higher than baseline volume (P<0.004). Similarly, PICF volume was statistically higher than GCF volume at each time point following loading

(Figure 6; P<0.04).

Resonance Frequency Analysis

The mean implant stability quotient (ISQ) prior to loading was 76.5±1.7 (Figure

7). A statistically significant increase in ISQ values was observed at 1 week (79.9±1.5;

P<0.001) and at 3, 6 and 12 weeks (Figure 7; 80.67±1.6, 81.5±1.5 and 83.7±1.7, respectively; P<0.0001) compared to baseline (before loading) ISQ values. The differences between post-loading values at each time point were not statistically significant (P>0.05). A mean increase of 7.2 ISQ was observed from baseline to week twelve.

Crevicular Fluid Content Prior to and Post-loading

Several cytokines and growth factors (e.g PDGF, IL-4, IL-6, IL-7, IL-10, IL-12,

IL-17, Eotaxin, FGFb, IFN and TNF-α) were expressed at detectable, but low levels at both peri-implant and adjacent GCF samples, with no differential expression throughout observation period (Table III). IL-1, IL-1ra, IL-8, G-CSF, IP-10, MCP-1, MIP-1b and

VEGF presented higher levels of expression in GCF samples compared to PICF samples

33 prior to loading; although there was no statistically significant difference between these baseline values. IL-1, IL-1ra, IL-8, G-CSF, IP-10, MCP-1, MIP-1b and VEGF were highly expressed within PICF samples following loading and are discussed here.

IL-1 (interleukin-1 beta) is a pro-inflammatory cytokine, thought to work in conjunction with TNF-alpha in bone and soft tissue remodeling. The total amount of IL-

1 within PICF prior to loading was 5.4±0.8 pg, lower than GCF IL-1 levels at baseline

(11±1.2 pg) (Figure 8). PICF IL-1 total amount was doubled by the first week of loading reaching a total amount of 10.4±2 pg and this amount was similar to adjacent

GCF IL-1 levels (10.8±1.5 pg; P>0.05). A decrease in both PICF and GCF IL-1 total amount was noted at 3, 6 and 12 weeks of loading with no statistically significant difference between and within groups (P>0.05).

IL-1ra (interleukin-1ra) is a pro-inflammatory cytokine. The total amount of IL-

1ra within PICF prior to loading was 13179±3266 pg which was more than two fold less than baseline GCF levels (28523±7263 pg) (Figure 9). However, the difference between groups was not statistically significant due to large variation observed in data distribution

(P>0.05). IL-1ra PICF and GCF levels stayed fairly stable throughout observation period with no statistical significance at different time points (Figure 9).

IL-8 (interleukin-8) is a pro-inflammatory cytokine. It acts as a powerful chemoattractent for neutrophils. PICF IL-8 total amount was 43±10 pg prior to loading, close to a twofold less amount than adjacent GCF levels (78±17 pg; Figure 10). The difference between two groups was not statistically significant (P=0.09). A twofold increase in PICF IL-8 was observed at 1 week post-loading (89±28 pg) compared to GCF

34

IL-8 total amount at the same time point (53±9 pg). The difference between groups was not statistically significant due to large variation observed at implant sites (P>0.05;

Figure 10). GCF and PICF IL-8 levels remained at stable levels after 3 weeks post- loading period.

G-CSF (granulocyte-colony stimulating factor) is a pro-inflammatory cytokine thought to be a major regulator of bone marrow progenitor cells, especially polymorphonucleus leukocytes. [117] Preloading PICF G-CSF total amount was 6±1.3 pg which was lower than baseline GCF levels (10±1.4 pg; P=0.08) (Figure 11). A slight increase in PICF G-CSF presence was noted at week one. A more notable increase was observed at week three, post-loading (10.5±3.7pg). The increase observed at week three was not statistically significant, mainly due to large variations observed between subjects

(Figure 11).

IP-10 (interferon-inducible protein-10) is a pro-inflammatory cytokine.

Preloading PICF IP-10 total amount was 22±6 pg. It was lower than baseline GCF levels

(64±24 pg) (Figure 12; P=0.08). PICF IP-10 levels appeared to increase following three weeks post-loading (2.5 fold increase by week 12; Figure 12). However, standard error values were too high to reach statistically significant results (P>0.05).

MCP-1 (monocyte chemotactic protein-1) is a chemotactic pro-inflammatory cytokine involved in recruitment of macrophages. Preloading PICF MCP-1 total amount was 7±3.4 pg with no statistically significant differences between groups at baseline

(Figure 13; P=0.3). PICF MCP-1 levels reached GCF levels by week 1 post-loading and this increase became statistically significant by 12 week post-loading (8.6±2 pg; P=0.01).

35

MIP-1b (macrophage inflammatory protein-1b) is a chemotactic pro- inflammatory cytokine involved in macrophage recruitment during wound healing. Pre- loading PICF MIP-1b total amount was 7±3.5 pg, no significant difference with GCF baseline levels (Figure 14; P>0.05). There was a 1.7 fold increase in PICF MIP-1b levels by 1 week post-loading (12±4 pg). However, this increase was not statistically significant (P>0.05). The increase in PICF MIP-1b total amount remained elevated through week twelve post-loading. The difference between PICF and GCF MIP-1b content became statistically significant by week 12 (4.4±0.5 pg in GCF and 8.5±2 pg in

PICF, respectively; P=0.02).

VEGF (vascular endothelial growth factor) is a pro-inflammatory cytokine involved with the development of capillary formation and permeability during wound healing. The pre-loading difference between GCF and PICF levels of VEGF was not statistically significant (Figure 15; 34±4.5 pg and 22.5±3 pg, respectively; P>0.05). PICF

VEGF total amount increased to similar GCF levels at week 1 post-loading and remained stable for the rest of the twelve week observation period for both PICF and GCF levels

(Figure 15; P>0.05).

Microbiological Findings

Polymerase Chain Reaction (PCR) assays were conducted to detect the presence of eight putative periodontal pathogens previously reported to be related to peri-implant bacterial flora (Figure 16). V. parvula was detected in about 60 to 70% of implant sites at each time point and about 50 to 70% of adjacent tooth sites at each time point. Similarly,

36

S.mitis was commonly found at implant and tooth sites, 50-70% of implant sites and 50-

80% of tooth sites. In addition, P. micros and F. neviforme were moderately detected at both sites 20 to 60%. T. forsythia and C. gingivalis were detected at lower percentages, 0 to 40%. Surprisingly, T.denticola was detected rarely throughout the study. However, its presence increased with a detection rate of 14% of subjects by week 12. This expression was specific to implant only. (Figure 16)

Correlations Between Clinical Parameters and Biomolecules

Correlation analysis among clinical parameters revealed that PI and GI around teeth were highly correlated (P≤0.001; r=0.4). However, similar correlation was not detected between these indices around implants (P=0.58; r=-0.66). (Table 4)

When PICF cytokine levels were analyzed for possible correlation, several were found to be highly correlated with each other. (Table 5)

Correlation analysis comparing peri-implant clinical parameters with PICF cytokine levels revealed that total amount of five (IL-1beta, IL-8, G-CSF, MIP-1b, and

VEGF) out of eight cytokines were statistically correlated with PICF volume. (Table 6)

Also, GI was statistically correlated with PICF MCP-1 expression (P=0.014, r=0.284).

(Table 6) Surprisingly, ISQ was significantly correlated with peri-implant PICF G-CSF,

IP-10, and MCP-1 total amount. (Table 6)

37

Table 1: Study Timeline/Outline

Appointment Description 1 Patient Screening 2 Comprehensive Examination 3 One Stage Implant Surgery Osseointegration Healing Period Implant Impression for crown fabrication 4  PA with impression post in place  Open tray direct transfer impression Base line data collection, crown insertion 1. Gingival Indices a. Plaque Index b. Gingival Inflammation c. Probe Depths d. Attached Gingival Measurements 2. GCF Natural Tooth Sample 5 a. Four 20 second samples from adjacent teeth (two from each tooth) 3. PICF Implant Samples a. Four 20 second samples from the implant (each sample from different sites) 4. Resonance Frequency Analysis (ISQ) a. 4 readings (two buccal two lingual) 5. Samples Stored at -80°C 1 week data collection 1. Gingival Indices a. Plaque Index b. Gingival Inflammation c. Probe Depths 2. GCF Natural Tooth Sample a. Four 20 second samples from adjacent teeth (two from each 6 tooth) 3. PICF Implant Samples a. Four 20 second samples from the implant (each sample from different sites) 4. Resonance Frequency Analysis (ISQ) a. 4 readings (two buccal two lingual) 5. Samples Stored at -80°C 3 week data collection 1. Gingival Indices a. Plaque Index b. Gingival Inflammation c. Probe Depths 7 2. GCF Natural Tooth Sample a. Four 20 second samples from adjacent teeth (two from each tooth) 3. PICF Implant Samples a. Four 20 second samples from the implant (each sample from different sites) 38 Continued

Table 1: Continued 4. Resonance Frequency Analysis (ISQ) a. 4 readings (two buccal two lingual) 5. Samples Stored at -80°C 6 week data collection 1. Gingival Indices a. Plaque Index b. Gingival Inflammation c. Probe Depths 2. GCF Natural Tooth Sample a. Four 20 second samples from adjacent teeth (two from each 8 tooth) 3. PICF Implant Samples a. Four 20 second samples from the implant (each sample from different sites) 4. Resonance Frequency Analysis (ISQ) a. 4 readings (two buccal two lingual) 5. Samples Stored at -80°C 12 week data collection 1. Gingival Indices a. Plaque Index b. Gingival Inflammation c. Probe Depths d. Attached Gingival Measurements 2. GCF Natural Tooth Sample 9 a. Four 20 second samples from adjacent teeth (two from each tooth) 3. PICF Implant Samples a. Four 20 second samples from the implant (each sample from different sites) 4. Resonance Frequency Analysis (ISQ) a. 4 readings (two buccal two lingual) 5. Samples Stored at -80°C

39

Table 2: Description Demographics

Description Demographics Data Total Number Participants In Study 15 Total Complete 15 (100%) Male Participants 8 (53.3%) Female Participants 7 (46.7%) Average Age of Participant 54 ±3 yrs Tooth Type 11 molars / 4 premolar Anatomic Location 11 mandible / 4 maxilla Osseointegration Healing Time (prior to loading) 6.1 ±0.3 months [4-9 months] (mean ±se) [range] Implant diameter (mean ±se) [range] 4.7 ±0.15 mm [3.7-6 mm] Implant length (mean ±se) [range] 12 ±0.3 mm [10-13 mm] Total Successful 14 (93.3%) Total Failed 1 (6.7%) Early (pre-loading) 0 Late (post-loading) 1

40

Table 3: Pro-inflammatory Biomarkers Distribution in PICF and GCF

Before loading 1 Week 3 Weeks 6 Weeks 12 Weeks pg (+/- se) Tooth Implant Tooth Implant Tooth Implant Tooth Implant Tooth Implant

PDGF 2.293 (0.2435) 2.273 (0.3102) 2.527 (0.4923) 2.38 (0.2887) 2.293 (0.4224) 2.36 (0.3422) 2.54 (0.5459) 2.18 (0.3916) 1.979 (0.2667) 2.743 (0.4555)

IL-1B 10.95 (1.215) 5.387 (0.7904) 10.8 (1.541) 10.4 (2.088) 8.673 (1.911) 7.713 (1.704) 7.54 (1.331) 7.427 (1.515) 9.229 (2.003) 6.914 (1.403)

IL-1ra 28523 (7263) 13179 (3266) 21522 (7315) 14788 (3137) 15371 (3157) 13109 (3616) 19078 (3180) 15666 (5812) 21863 (5539) 14627 (2263)

IL-4 0.1053 (0.01693) 0.08933 (0.01446) 0.1047 (0.01864) 0.1173 (0.02108) 0.1 (0.02021) 0.09267 (0.01861) 0.08467 (0.01687) 0.07733 (0.01307) 0.1086 (0.02379) 0.09571 (0.01712)

IL-6 0.804 (0.2129) 0.4513 (0.1192) 0.7427 (0.2004) 0.6613 (0.1965) 0.8013 (0.2002) 0.6927 (0.2011) 0.756 (0.2396) 0.972 (0.4901) 0.7993 (0.2142) 0.5443 (0.1594)

IL-7 0.02533 (0.02033) 0.004 (0.004) 0.09267 (0.05943) 0.158 (0.1171) 0.146 (0.09607) 0.112 (0.07232) 0.028 (0.07232) 0.022 (0.01996) 0.07857 (0.07124) 0.04571 (0.04273)

IL-8 77.8 (16.94) 43.46 (10.09) 52.93 (9.311) 89 (28.16) 52.53 (11.84) 47.6 (11.79) 46.2 (8.052) 112.1 (73.06) 47.14 (8.401) 72.93 (26.89)

IL-10 2.473 (0.3668) 1.993 (0.2607) 2.52 (0.4224) 2.443 (0.3302) 2.393 (0.3836) 2.247 (0.305) 2.165 (0.2756) 2.08 (0.261) 2.254 (0.3366) 2.214 (0.291) IL-12 3.14 (0.6746) 2.117 (0.3796) 3.443 (0.7256) 2.933 (0.559) 3.227 (0.7435) 2.57 (0.5145) 2.524 (0.4606) 2.301 (0.4196) 2.731 (0.5781) 2.185 (0.424) 40 IL-17 p 1.74 (0.1656) 1.953 (0.2509) 1.833 (0.2124) 1.88 (0.2614) 1.8 (0.1746) 1.573 (0.1739) 1.733 (0.1817) 1.727 (0.2121) 1.757 (0.2045) 1.479 (0.1438) Eotaxin 0.8807 (0.211) 1.115 (0.2092) 1.027 (0.2448) 1.027 (0.2262) 1.033 (0.2088) 1.147 (0.2136) 1.227 (0.3258) 1.093 (0.2013) 1.236 (0.368) 1.164 (0.2506) FGFb 2.28 (0.116) 2.26 (0.1767) 2.22 (0.1061) 2.367 (0.3023) 2.207 (0.1217) 2.087 (0.1549) 2.133 (0.1423) 2.04 (0.1077) 2.164 (0.1529) 2.243 (0.1744)

GCSF 9.533 (1.43) 5.962 (1.3040 7.967 (1.245) 6.86 (1.297) 9 (1.278) 10.53 (3.677) 7.187 (0.8657) 6.973 (1.301) 7.1 (1.125) 7.621 (1.051)

IFNg 3.72 (0.2509) 2.887 (0.19) 3.967 (0.3718) 3.687 (0.3135) 3.667 (0.3572) 3.167 (0.3418) 3.507 (0.4159) 3.133 (0.2746) 4.35 (0.8514) 3.493 (0.2833) IP10 64.4 (24.49) 21.93 (5.955) 34.05 (10.53) 22.43 (4.776) 32.75 (9.604) 43.87 (16.49) 26.8 (7.627) 48.21 (21.24) 53 (19.63) 56 (14.84)

MCP-1 10.5 (2.53) 7.167 (1.578) 9.8 (2.377) 9.4 (2.22) 9.313 (2.12) 8.64 (2.054) 8.947 (2.202) 8.587 (1.933) 9.436 (2.545) 8.643 (2.118)

MIP1b 7.62 (3.422) 6.967 (3.459) 4.913 (1.242) 11.77 (3.931) 4.593 (1.094) 8.207 (2.457) 5.587 (1.351) 9.407 (2.817) 4.443 (0.5197) 8.507 (1.953)

TNFa 3.1 (0.3179) 1.887 (0.3389) 2.667 (0.3886) 2.493 (0.419) 2.447 (0.3414) 2.933 (0.3283) 2.6 (0.4085) 2.293 (0.3333) 2.807 (0.3509) 2.554 (0.476)

VEGF 33.67 (4.524) 22.47 (3.025) 31.93 (2.891) 30.33 (3.654) 29.4 (3.11) 24.6 (3.016) 26.27 (3.38) 25.2 (5.372) 28.93 (3.943) 26.36 (4.99) Data labeled in Blue is also presented in Figures 8 through 15.

41

Table 4: Spearman Correlation for Clinical Parameters at Implant and Tooth Sites

IMPLANT PICF PI GI P=0.516 PI r=-0.077 P=0.979 P=0.575 GI r=0.003 r=-0.066 P=0.08 P=0.623 P=0.07 RFA r=0.207 r=0.058 r=0.213

TOOTH GCF PI GCF P=0.557 PI r=0.069 P=0.490 P≤0.001 GI r=0.082 r=0.402

Table 5: Spearman Correlation Analysis for Cytokine Data at Implant Sites

IL-1 IL-1ra IL-8 G-CSF IP-10 MCP-1 MIP-1b P≤0.001 IL-1ra r=0.506 P≤0.001 P≤0.001 IL-8 r=0.645 r=0.525 P≤0.001 P≤0.001 P≤0.001 G-CSF r=0.591 r=0.558 r=0.651 P=0.141 P=0.01 P=0.001 P≤0.001 IP-10 r=0.174 r=0.301 r=0.370 r=0.583 MCP- P=0.005 P=0.07 P=0.743 P≤0.001 P=0.006

1 r=0.325 r=0.214 r=0.039 r=0.490 r=0.318 MIP- P=0.005 P=0.04 P≤0.001 P≤0.001 P≤0.001 P=0.380

1b r=0.323 r=0.242 r=0.719 r=0.485 r=0.505 r=-0.103 P≤0.001 P≤0.001 P≤0.001 P≤0.001 P=0.385 P=0.436 P≤0.001 VEGF r=0.610 r=0.584 r=0.735 r=0.448 r=0.103 r=-0.092 r=0.411

42

Table 6: Spearman Correlation between Clinical Parameters and Cytokines at Implant Sites

PICF PI GI ISQ P=0.004 P=0.225 P=0.276 P=0.257 IL-1 r=0.332 r=0.143 r=-0.128 r=0.135 P=0.689 P=0.888 P=0.690 P=0.097 IL-1ra r=0.047 r=0.017 r=-0.047 r=0.196 P=0.021 P=0.602 P=0.231 P=0.952 IL-8 r=0.269 r=0.062 r=0.141 r=-0.007 P=0.02 P=0.591 P=0.726 P=0.044 G-CSF r=0.271 r=-0.064 r=0.041 r=0.236 P=0.190 P=0.298 P=0.942 P=0.013 IP-10 r=0.155 r=-0.123 r=-0.091 r=0.291 P=0.304 P=0.944 P=0.014 P=0.025 MCP-1 r=0.121 r=0.008 r=-0.284 r=0.262 P≤0.001 P=0.468 P=0.115 P=0.086 MIP-1b r=0.416 r=-0.086 r=0.185 r=0.203 P=0.022 P=0.611 P=0.902 P=0.954 VEGF r=0.265 r=0.06 r=0.015 r=0.007

43

Table 7: Pro-inflammatory Biomolecules

Pro-inflammatory Mediator Sources Target/Action

PDGF Platelets – active Activate platelet cells, aiding in the inflammatory and clot formation process (platelet derived growth factor)

IL-1 Beta Monocyte, Macrophage, T and B Cells, Endothelial cells, T and B cells, Monocytes, Fever, acute phase response, neutrophil production (Interleukin 1 beta) Fibroblasts, Smooth Muscle G-CSF Endothelium and macrophages Proliferation and differentiation of granulocytes (Granulocyte Colony Stimulating Factor) MIP-1b Chemokine that leads to migration of lymphocytes and monocytes during wound healing, Macrophages (Macrophage Inflammatory Factor 1) early implant mucositis

MCP-1 Monocytes, endothelial cells Chemotactic for Monocytes not neutrophils Basophil activator and degranulation induction (Monocyte Chemotactic Protiein – 1) VEGF Endothelial cells Promote capillary formation and dilation (Vascular Endothelial Factor) IP-10 Neutophils stimulated by INF-gamma, TNF-alpha, or IL-1 beta Activator and attractant of T cells May play a role in hypersensitivity reactions (Interferon-inducible Protein – 10)

IL-1 ra Competitively inhibits IL-1a and IL-1b, does not elicite same response as IL-1a/b, role in Monocytes, macrophages, neutrophils fibroblasts, others (Interleukin 1 receptor antagonist) inflammation unclear Growth factor for hematopoietic cells and lymphocytes. Stimulates colony formation of IL-4 Mast cells, T cells, Bone Marrow Cells neutrophils, eosinophils, basophils, erythoid, megakaryocyte, and monoclytic cells. Not 43 (Interleukin 4) lymphocytes IL-6 T and B cells, macrophages, bone marrow stromal cells, Regulates B and T cell functions (Interleukin 6) fibroblasts, endothelial cells Induces acute phase response IL-7 Bone marrow stromal cells, thymic stromal cells, spleen cells Growth factor for B and T cells (Interleukin 7) IL-8 Monocytes, Lymphocytes, Granulocutes, Fibroblasts, Chemokine, Primary action is to attract and activate neutrophils, angiogenic activity, attracts (Interleukin 8) Endothelial cells, and more some lymphocytes IL-10 Proliferation of B cells, thymocytes, and mast cells, Stimulates IgA fabrication of B cells, Activated CD4 and CD8 T cells (Interleukin 10) suppreses Th1 cells IL-12 Macrophages and dendritic cells Differentiation of TH1 T Helper cells, induces INF-γ (Interleukin 12) IL-17 Hematopoiesis by inducing cytokine production by epithelial cells, fibroblast, endothelial, CD4 T cells (Interleukin 17) and stromal cells. Enhances ICAM-1 making cells more adherent Activation, growth and differentiation of T and B cells, macrophages, Upregulates MHC INF-γ CD4 and CD8 expression TNF-alpha Production of adhesion molecules, chemokine, and cytokine inducing IL-1, IL-6, IL-8, and Monocytes, macrophages, other (Tumor Necrosis Factor – Alpha) MMPs. In conjunction with IL-l induces osteoclast activation, 44

Figure 1: Surgical Guide on Diagnostic Cast

1.1 1.2

Buccal view of surgical guide identifying implant Occlusal view of a surgical guide with a lingual access angulation and proposed cervical location of implant crown lot on the diagnostic cast.

1.3

Occlusal view of a surgical guide with a pilot hole on the diagnostic cast.

45

Figure 2: Plaque Accumulation Prior To and Post Crown Delivery

Plaque Accumulation 1.2

1

0.8

s.e) ± 0.6 TOOTH IMPLANT

PI (mean PI 0.4

0.2

0 BL 1 WK 3 WKS 6 WKS 12 WKS

BL: Before Loading; data is presented as mean Plaque Index ± standard error *P<0.004 Difference between implant and teeth at each week **P=0.02 Difference at baseline (before loading) between implant and tooth

46

Figure 3: Soft Tissue Characteristics Prior to and Post Crown Delivery

Soft Tissue Characteristics 1.4

1.2

1 s.e)

± 0.8 TOOTH 0.6 IMPLANT GI (mean GI 0.4

0.2

0 BL 1 WK 3 WKS 6 WKS 12 WKS

Data is expressed as mean GI ± standard error *P<0.05 comparison of implant GI with teeth GI at BL and 1 week **P<0.001 Comparison of baseline implant GI with week 12 implant GI

47

Figure 4: The amount of Keratinized Tissue prior to and post loading

Keratinized Tissue 6

5

4

se) ± 3 BUCCAL

LINGUAL mm (mean mm 2

1

0 BL 12 WEEKS

48

Figure 5: Changes in peri-implant and adjacent periodontal probing depths prior to and following loading

Probing Depth 3.8 3.6

3.4

3.2 s.e)

± TOOTH-BUC 3 IMP-BUC 2.8 TOOTH-LING

mm (mean mm 2.6 IMP-LING 2.4 2.2 2 BL 1 WEEK 3 WEEKS 6 WEEKS 12 WEEKS

49

Figure 6: Peri-implant and Periodontal Crevicular Fluid Volume Changes prior to and following loading

CREVICULAR FLUID VOLUME 2 1.8 1.6

1.4

s.e) 1.2 ± 1 TOOTH

0.8 IMPLANT ul (mean ul 0.6 0.4 0.2 0 BL 1 WK 3 WKS 6 WKS 12 WKS

*P=0.03 baseline PICF and GCF **P<0.004 PICF – difference between BL and WK 1, 3, 6, 12 ***P<0.04 PICF with GCF each time point Data is presented as total amount of protein obtained within 20 second crevicular fluid sampling.

50

Figure 7: Resonance Frequency Analysis prior to and following loading

Resonance Frequency Analysis 88

86

84

se) 82 ± 80

78 ISQ (mean ISQ 76

74

72 BL 1WK 3WK 6WKS 12WKS

**P<0.01 compared to baseline ***P<0.0001 compared to baseline Data is presented as total amount of protein obtained within 20 second crevicular fluid sampling.

51

Figure 8: PICF and GCF IL-1β total amount prior to and following loading

IL-1β 14

12

10

se)

± 8 TOOTH

6 IMPLANT pg (mean pg 4

2

0 BL 1WK 3WKS 6WKS 12WKS

P>0.05 between and within groups Data is presented as total amount of protein obtained within 20 second crevicular fluid sampling.

52

Figure 9: PICF and GCF IL-1ra Total Amount Prior to and Following Loading

IL-1ra 40000

35000

30000

se) 25000 ± 20000 TOOTH IMPLANT

15000 pg (mean pg 10000

5000

0 BL 1WK 3WKS 6WKS 12WKS

P>0.05 between and within group differences Data is presented as total amount of protein obtained within 20 second crevicular fluid sampling.

53

Figure 10: PICF and GCF IL-8 Total Amount Prior to and Following Loading

IL-8 140

120

100

se)

± 80 TOOTH

60 IMPLANT pg (mean pg 40

20

0 BL 1WK 3WKS 6WKS 12WKS

P>0.05 between and within groups Data is presented as total amount of protein obtained within 20 second crevicular fluid sampling.

54

Figure 11: PICF and GCF G-CSF Total Amount Prior to and Following Loading

G-CSF 16

14

12

se) 10 ± 8 TOOTH IMPLANT

6 pg (mean pg 4

2

0 BL 1WK 3WKS 6WKS 12WKS

P>0.05 between and within groups differences Data is presented as total amount of protein obtained within 20 second crevicular fluid sampling.

55

Figure 12: PICF and GCF IP-10 Total Amount Prior to and Following Loading

IP-10 100 90 80

70

se

± 60 50 TOOTH

40 IMPLANT pg (mean pg 30 20 10 0 BL 1WK 3WKS 6WKS 12WKS

P>0.05 between and within group differences Data is presented as total amount of protein obtained within 20 second crevicular fluid sampling.

56

Figure 13: PICF and GCF MCP-1 Total Amount Prior to and Following Loading

MCP-1 16

14

12

se) 10 ± 8 TOOTH IMPLANT

6 pg (mean pg 4

2

0 BL 1WK 3WKS 6WKS 12WKS

*P=0.014 PICF levels at 12 weeks compared to PICF levels at baseline Data is presented as total amount of protein obtained within 20 second crevicular fluid sampling.

57

Figure 14: PICF and GCF MIP-1b Total Amount Prior to and Following Loading

MIP-1b 18 16 14

12 se) ± 10 TOOTH 8 IMPLANT

pg (mean pg 6 4 2 0 BL 1WK 3WKS 6WKS 12WKS

*P<0.02 PICF compared to GCF detection at week 12 Data is presented as total amount of protein obtained within 20 second crevicular fluid sampling.

58

Figure 15: PICF and GCF VEGF Total Amount Prior to and Following Loading

VEGF 45 40 35

30 se) ± 25 TOOTH 20 IMPLANT

pg (mean pg 15 10 5 0 BL 1WK 3WKS 6WKS 12WKS

P>0.05 between and within group differences Data is presented as total amount of protein obtained within 20 second crevicular fluid sampling.

59

Figure 16: Microflora Formation – Implant and Tooth

FLORA FORMATION-IMPLANT

T.dent 400 C.ging 300 G.heam 200 F.nav 100 P.mic % positive subjects positive % 0 V.par S.mit

FLORA FORMATION-TOOTH

450 T.dent

350 C.ging 250 G.heam

150 F.nav P.mic 50 % positive subjects positive % V.par -50 S.mit

60

CHAPTER IV: DISCUSSION

The physiological healing response to implant placement surgery is an acute and well controlled activation of several pro-inflammatory pathways affecting both hard and soft tissue components of the wound. This short term response may become reactivated following occlusal loading of the implant through remodeling procedure. [10] [12] [116]

[117] [118] The ultimate goal of this study was to detect peri-implant tissue changes that occur due to bone and soft tissue remodeling as a response to mechanical loading.

Specifically, an investigation on clinical parameters and the expression of local cytokines and growth factors within peri-implant crevicular fluid and adjacent gingival crevicular fluid was conducted in order to differentiate peri-implant soft tissue response prior to and following occlusal loading. To our knowledge this is the first patient based study evaluating early peri-implant soft tissue responses immediately following mechanical loading following a conventional implant protocol.

Study population included 15 non-smokers with similar gender and age distribution. None of the edentulous sites was previously grafted. It has been reported that peri-implant wound healing is complicated in smokers. [119] [120] Many studies report that smokers may experience higher rates of implant failure. [121] [122] [123]

[124] Thus, smokers were excluded from this study. Although controversial, some studies report differential responses in peri-implant clinical parameters, implant stability,

61 and long term survival rate for implants placed at grafted sites. [125] Thus, grafted sites were excluded from the study. Older age may affect wound healing rate in general. [126]

[127] Most of the patients in need of implant placement are older individuals. Study design recruited patients as old as 75 yrs. Efforts were made to have an equal presentation of adult age groups. Gender differences have not been shown as a possible risk indicator for implant failure. [128] [129] [130] Gender distribution within this study was similar and should not have any effect on results. Patients with a history of parafunctional habits were also excluded since this is a well-known risk factor for dental implant failure. [131]

[132] [133] All implants were placed according to a one-stage surgical protocol to help control the amount of possible scar formation at soft tissue level. In addition, all implant restorations were screw retained to allow evaluation of implant stability with ISQ at post loading evaluation visits. Most of the implants were placed in mandibular posterior sextant. All implants included in the study presented with baseline ISQ values >50 which allowed for one-stage surgical protocol. There was one implant failure observed at the twelfth week evaluation appointment. It was localized in a maxillary posterior sextant which is documented to possess a higher risk of failure. [134] [135] [136] [137]

The average osseointegration period prior to loading was 6.1±0.3 months with a range of four to nine months in the current study. It is possible that the variation in osseointegration healing times may have affected the results of the study. There is controversial data on how peri-implant tissues response to long-term healing period in the absence of occlusal forces. According to the mechanostat theory of bone remodeling, developed by Frost, [26] one might speculate bone atrophy may occur around an implant

62 if the healing period persists long enough without mechanical stimulus to the implant- bone interface. Romanos’ [138] observations of unloaded implant at three months exhibiting less dense peri-implant bone with increased adipose tissue compared to a loaded implant, supports this theory. However, Mattieu et al [139] reported, using a rabbit model, increased bone to implant contact of unloaded implants at fourteen weeks post surgical placement when compared to seven weeks, 69% and 27% respectively.

Similarly, Carr et al [140] observed, using a baboon model, an increase in bone to implant contact at six months when compared to three months, 49.2% and 38.8% respectively. There is sufficient data to suggest healing periods longer than conventional three months in the mandible and six months for maxilla without occlusal loading does not affect implant success and/or survival rate. In addition, though not a direct occlusal function, implants placed with one stage protocol experience some mechanical loading since they are exposed to masticatory forces as well as forces applied by muscles such as cheek and tongue. [117]

Clinical evaluation of the soft tissue response to occlusal loading included: plaque accumulation, gingival inflammation, changes in probing depth, and keratinized tissue width. In general, all plaque scores were exhibited low levels for both implant and adjacent teeth sites. Peri-implant surfaces observed significantly lower plaque accumulation than teeth which is similar to previous reports for plaque accumulation around implant supported restorations. [141] [142] [143] Lower plaque scores around implants have been attributed to the restorative material used. All restorations were metal ceramic restoration with no metal showing in the current study. Dental porcelains have

63 been reported to minimize plaque formation as a result of their highly polished surface and low surface energy. [144] [145] [146] [147]

Peri-implant gingival inflammation, recorded as modified gingival index (GI) score, exhibited an increase following loading in the current study. This is despite to low

PI scores. There was an inverse correlation of implant PI and implant GI scores in the data. Studies investigating peri-implant gingival tissue characteristics generally report similar or slightly higher GI scores around healthy implants compared to healthy teeth.

[142] [148] [149] Gingival inflammation is generally directly correlated to plaque accumulation. It appears that differential GI response detected in the current study prior to and following loading may be directly related to both crown insertion and functional loading.

Peri-implant and adjacent periodontal probing depths were used to evaluate soft tissue response to implant loading. A slight increase was observed in peri-implant probe depths by week twelve after loading compared to pre-loading values. This agrees with previously reported literature. [146] [149] Differences in soft tissue attachment around dental implants compared to gingival attachment around a tooth result in an increased probability that the tip of a periodontal probe will pass through the junctional epithelium and implant-gingival attachment even in health. [50] [83] This will result in deeper peri-implant probe depth measurements. An additional compounding factor is the potential size discrepancy of a full contour implant restoration emerging from a significantly smaller implant fixture. The size discrepancy results in difficulty with probing peri-implant sulci and may result in iatrogenic damage to the peri-implant

64 gingival tissue. [1] [150] Most crowns inserted were greater diameter than the implant and implant healing abutment. The insertion of an implant crown resulted in expansion and pressure on the peri-implant gingival tissues. However, there was no implant surface exposure and/or gingival recession.

The results of the study suggest during the twelve week evaluation period following implant loading little to no changes to keratinized gingival width should be expected. Similar information evaluating the effects of loading on peri-implant keratinized gingiva immediately following loading is lacking in the literature. However, the necessity of keratinized tissue around implants for long-term success has been discussed in depth. Most studies suggest peri-implant keratinized tissue is necessary to maintain long-term peri-implant health. [151] [152] [153] [154, 155] In contrast others

[156] [157] [158] report that keratinized gingival width does not affect the risk of gingival recession or peri-implant gingival health. The lack of keratinized tissue was an exclusion criterion in the current study. Thus, it was not possible to compare peri-implant clinical parameters in the absence of keratinized tissue.

Resonance Frequency Analysis (RFA) was used to determine implant stability in the current study. A statistically significant increase in ISQ levels was noted following implant loading. Literature on ISQ levels at early phases of post loading is very limited.

Most of the available studies compare ISQ values of immediately loaded implants with implants following a conventional osseointegration period. [73] [74] [75] [159] These studies report similar ISQ values following implant surgery. Several in vitro and in vivo studies suggest that implants effectively transfer functional stress to the bone via the

65 intimate bone to implant interface. [18] [35] [34] Thus, the theory of “Wolff’s Law” of bone remodeling can be expected to occur in peri-implant bone which is subject to strains produced by occlusal load. [22] [24] [25] [26] Fischer et al. [160] reports ISQ values of immediately loaded implants after twelve months in function with a slight increase of 3.5

ISQ units from baseline surgery. Grunco et al. [161] reports no change in RFA from baseline after to twelve months of service with conventionally loaded implants and a slight increase from baseline to twelve months of 2.3 ISQ with immediately loaded implants. Our data is the first available data in which implant stability has been measured immediately following conventional implant loading. It is reasonable to speculate that the significant increase in RFA values observed in our study is most likely the result of the bone remodeling process that occurs due to physiologic strain in the peri-implant bone via the dental implant-restoration system.

The reason dental literature may lack data regarding implant stability following crown insertion is likely related to the potential risks involved with regaining access to an implant for RFA measurements once the final restoration has been delivered. Potential risks may arise related to the restoration retaining screw, crestal bone loss, and irretrievability. It is documented in the literature that multiple insertions and removals of titanium retaining screws may be detrimental to a screws implant pre-load. [162] [163]

[164] Changes in screw preload may increase the risk of screw loosening or breakage.

This concern was addressed by not torquing the retaining screws to the recommended final torque values until the final appointment in the current study. The crowns were removed and re-inserted only four times. It has been recommended that screws can be

66 torque to the manufactures recommendations ten times prior to replacement. [162] None of the subjects had screw loosening or screw breakage during the 12 weeks observation period.

There is also concern that repeated removal and insertion of transmucosal crowns/abutments may induce crestal bone changes. Reports from Grandi [165] and

Canullo [166] suggest multiple insertions and removal of an implant abutment is associated with increased crestal bone loss adjacent to an implant. However, these studies report statistically significant differences in crestal bone loss observed with repeated abutment removal; the crestal bone loss reported was 0.1 to 0.2 mm and may not be clinically significant to the long term success of an implant. The crestal bone loss reported with multiple abutment removal and insertions is within the generally accepted implant success criteria presented by Smith and Zarb. [5] There is no evidence to suggest the insertion and removal of the implant crowns as performed in our study will affect the peri-implant soft tissue and characteristics. Peri-implant bone loss/crestal changes could not be evaluated in this study since 12 weeks is too early to differentiate changes in bone level.

A significant increase in peri-implant crevicular (PICF) volume was observed in response to occlusal loading in the current study. Tozum et al. [167] compared PICF to

GCF volumes and reports PICF volume to be greater than GCF volume in patients with gingival health. Our results support this observation. However, Murata et al. [99] report

PICF volume to be less than GCF volumes in healthy subjects and Apse et al. [93] report no difference. Other studies report conflicting data regarding PICF volumes over time

67 with loaded implants. Tozum et al. [168] reports PICF volume fluctuates over a one year period of time with a slight decrease in volume at twelve months of loading. Similarly,

Guncu et al. [161] reports a trend of PICF volume decreasing over time. Both these studies report observations after months of loading. Beck et al. [96] describes a maximal increase in PICF volume at the fourth day immediately following crown insertion, but reports a subsequently decrease in volume. In addition Last et al. [97] suggests PICF volumes increase after restoration of an implant and remain elevated even with a lack of visible inflammation. The current study compares baseline PICF to shorter time intervals post loading. Results show an increase in PICF following crown insertion which then stayed similar to adjacent GCF volume throughout observation time. In that aspect, what is reported as post-loading data is similar to most of the published work.

Several factors influence PICF volume. PICF volume is affected by: implant location [95] [98], implant diameter [95], gingival pocket depth [77] [95] [169], plaque presence [100], implant loading [96] [97], and inflammation [77] [82] [95] [98] [99]

[169]. The most significant and well documented factor is the presence of gingival inflammation. [77] The second most influential factor may be gingival pocket depth or volume. Larger teeth and larger pockets possess a larger area for fluid collection within the gingival sulcus. Thus, GCF volume increases would be expected. Increases in PICF volumes observed in our study are most likely a result of several factors. Initial PICF volumes obtained were the smallest volumes recorded. This volume obtained is related to the tissue height and healing abutment emergence shape. [77] Inherently PICF volumes obtained around an implant healing abutment at baseline would be less than

68

PICF volume obtained after crown insertion. Another possible contributing factor to the increase in PICF volume observed, especially at the week one follow-up appointment may be related to the previously mentioned process of gingival tissue remodeling and related inflammatory process.

PICF and adjacent GCF content were investigated prior to and following loading in the current study. Data on IL-1β, IL-1ra, IL-8, G-CSF, IP-10, MCP-1, MIP-1b, and

VEGF were presented and discussed here due to their consistent detection levels among various samples as well as their the differential expression between PICF and GCF at different time points.

Pro-inflammatory cytokine detection in peri-implant gingival crevicular fluid has been studied and reported in the dental literature. [92] [98] [100] [101] [102] [103] [105]

[104] [107] Most studies discuss changes that may exist related to peri-implant tissue in health and disease [92] [98] [100] [103] [104] [105] [107] or compared to healthy teeth

[101] [102], or surgical implant placement wound healing (Emecen-Huja, MS Thesis

2011). Only a limited number of studies have reported on changes in PICF pro- inflammatory cytokines solely with respect to loading. To our knowledge, this study is the first report on pro-inflammatory cytokine detection in PICF samples obtained immediately following implant loading.

Pro-inflammatory cytokines IL-1β, IL-8, MCP-1, MIP-1b and VEGF all demonstrated a detectable PICF increase the first week after loading when compared to baseline PICF levels. None of the increases were statistically significant. Subsequently, the detectable levels of each cytokine or growth factor decreased at week three and

69 remained at a similar level throughout the evaluation visits. Therefore, it is probable that the increase in IL-1β, IL-8, MCP-1, MIP-1b and VEGF detected at week one following loading was influenced by the simultaneous PICF volume increase, but the trend of subsequent decrease in detectable levels during the following weeks suggests other factors were involved. The increases observed at week one were likely in response to the change in peri-implant environment as a result of the crown insertion and in response to implant loading. These pro-inflammatory cytokines provide evidence for early peri- implant soft tissue response to implant crown insertion and functional loading.

Difference between PICF and GCF levels of IL-1β and VEGF were detected at baseline. IL-1β has been studied extensively in the dental literature. VEGF data is limited. IL-1β is thought to play a significant role in periodontal inflammation and disease as many have demonstrated a detectable increase in IL-1β in inflamed or disease

PICF. [95] [98] [100] [103] [104] [105] IL-1β presence in healthy PICF is similar to detectable levels observed in healthy GCF. [99] [102] Our data support this finding once the implant was loaded. Panagakos et al. [105] report IL-1β’s detectable presence of

59.47 pg/site in healthy PICF. Their samples were obtained using four paper strips at thirty seconds each; slightly longer than our twenty second collection times. Our data reports lower total presence in healthy PICF when compared to Panagokos. The significant difference in IL-1β in PICF and GCF recorded at baseline prior to loading appears to be a unique finding; however, it does correlate to previous data reported by our group (Emecen-Huja, MS Thesis 2011) regarding the twelve weeks period wound healing after implant surgery. This data, regarding Il-1β and VEGF requires further investigation

70 to possibly understand the significance of their elevated levels prior to implant loading when compared to natural teeth.

PICF G-CSF level at week three exhibited a notable, but not statistically significant, increase. Subsequently, at weeks six and twelve the detectable mean levels were returned to similar baseline and one week levels. The dental implant literature is void of references to G-CSF in relation to peri-implant soft tissue. G-CSF is known to be responsible for regulating bone marrow progenitor cells; especially neutrophil progenitor cells. [172] [173] [174] [175] It has also been shown to be involved in angiogenesis.

[176] [177] Deficiency in G-CSF has a strong correlation to neutropenic disorders. It has also been used to help successfully treat congenitally neutropenic people. [117]

[178] Recombinant GCSF has been shown to help improve oral health as conjunctive treatment with periodontal maintenance in patients with cyclic neutropenia. [176] [179]

In our data, G-CSF is detectable in PICF and appears to increase at week three after loading. This increase does not correlate with PICF volume increases. It does have a significant correlation with RFA data. G-CSF is likely expressed as a response to implant loading and requires further investigation.

Mean levels of IP-10 detected in the patient’s PICF also exhibited a change following loading. At week three mean IP-10 levels increased and continued to increase through week twelve. Comparative data from the literature is limited. Therefore, our data cannot be compared or confirmed with respect to PICF. More research is needed.

IP-1 does not correlate with PICF volumes in our data. It does correlate with RFA. It is likely the differential detection of IP-10 in PICF is affected by implant loading.

71

Mean MIP-1b exhibits a similar pattern as IL-1ra, IL-8, IL-1β, VEGF, and MCP-

1. However, this particular pro-inflammatory cytokine at baseline PICF is recorded to be at a similar level as found in GCF at baseline. At week one of loading an increase in mean MIP-1b detection occurs in PICF but not in the GCF. The difference in MIP-1b levels recorded in the PICF and GCF was not statistically significant at week one but became significant at week twelve. Similar to IP-10 and G-CSF, MIP-1b is not well documented in the dental PICF literature. Thus, comparisons are not possible at this time.

MIP-1b is reported as a chemokine for monocytes and lymphocytes and is thought to stimulate and help regulate hematopoietic cell production during inflammatory states.

[180] [181]

There was significant correlation with detectible cytokines in PICF in relation to each other. This is to be expected. Interestingly, there was significant correlation between G-CSF, IP-10, and MCP-1 with the clinical parameter RFA. IL-1β, IL-8, G-

CSF, MIP-1b, and VEGF exhibited a significant correlation with PICF. The correlation does not mean there is a direct cause and effect relationship. However, it does suggest there is a potential relationship with respect to the clinical data we recorded and implant loading, and requires additional investigation.

It is reasonable to speculate that sulcus flora would be affected by the insertion of an implant supported crown. Thus, PCR assays were used to detect the presence of eight specific putative periodontal pathogens.

The change in the peri-implant environment may affect the peri-implant microflora which in turn may additionally affect the peri-implant soft tissue response.

72

Our data reports the percentage of patients with detectable specific bacteria in PICF and

GCF. Evidence suggests microbial colonization of implants occurs early after insertion.

[170] [171] Early peri-implant microbial studies suggest healthy implant microbial composition as similar to healthy natural teeth [143] [172] [173] [174] [175] consisting mainly of cocci and non-motile rods [143] [172] [173] [176]. The prevalence of more anaerobic and motile bacteria appears to occur with increased time in the mouth [141]

[172] [177], increased probe depths [141], and microbial content of adjacent natural teeth

[172] [175]. The presence of known pathogenic bacteria has been reported to be present in both healthy and diseased implant tissues. [172] [175] [177] [178] Kumar et al. [179], in a recent study suggests that despite the similarities between peri-implant and periodontal microflora, peri-implant gingival microflora show specific differences. The putative pathogens studied in the current investigation were based on peri-implant microflora characteristics recently reported by Kumar et al. [115] Based on limited amount of samples, it appears that loading did not have a significant effect on peri- implant sulcus flora. However, a potent periodontal pathogen, Treponema denticola was highly expressed within peri-implant sulcus at 12 weeks compared to adjacent gingival sulcus. Casado [190] reports the presence of T. denticola on both healthy and diseased implant surfaces. Conversely, Lee [184] reports T. denticola is not present on healthy implants. Our data may be unique in that T. denticola was detected in few patients only during the twelve week and on implants.

T. denticola, a spirochete bacterium, is considered to be one of the three “Red

Complex” periodontal pathogens. The “Red Complex” is composed to P. gingivalis, T.

73 forsythia, and T. denticola. It is usually present with P. gingivalis due to its bonding affinity with P. gingivalis. When P. gingivalis and T. denticola are present together a significant increase in periodontal destruction occurs. [180]

One of the limitations of this study is small size of study population. The size of

N=15 was initially chosen by performing a priori sample size determination test. RFA was chosen as main clinical parameter. Previous studies were reported a 4-5 ISQ unit difference as statistically significant difference at P≤0.05. Ideally, a larger study population may help better interpret local cytokine levels.

Conclusions

In summary the following changes were observed with peri-implant parameters after implant loading:

 Peri-implant plaque accumulation was lower than adjacent teeth throughout the

twelve weeks evaluation period.

 Gingival inflammation increased around implants after loading reaching similar

levels as the adjacent natural teeth. Peri-implant gingival inflammation was not

correlated to peri-implant plaque scores.

 Peri-implant keratinized gingiva width remained unchanged.

 Peri-implant probe depths increased slightly after loading. They were slightly

deeper than adjacent teeth by twelve weeks of loading.

 PICF volume experienced a significant increase after loading and remained

elevated through week twelve.

74

 RFA (ISQ) measurements increased significantly from baseline following implant

loading throughout the twelve week observation period.

 Cytokines and growth factors such as: PDGF, IL-4, IL-6, IL-7, IL-10, IL-12, IL-

17, Eotaxin, FGFb, IFN and TNF-α were all detectable in PICF but at low levels

and no differential expression throughout the study.

 IL-1, IL-1ra, IL-8, G-CSF, IP-10, MCP-1, MIP-1b and VEGF exhibited

detectable changes in PICF in response to loading.

 Significant individual variations and fluctuations throughout observation time

were noticed in various PICF cytokine/growth factor levels.

 There was no difference between pre-loading and post-loading peri-implant

sulcus flora members within the limits of specific bacteria. Also, peri-implant

sulcus flora was similar to adjacent teeth sulcus flora of the eight species studied.

Our hypothesis states that dental implant loading following conventional osseointegration period affects implant stability, crevicular fluid volume and content.

The data from our study supports this hypothesis. Dental loading following conventional osseointegration effects implant stability, PICF, and PICF content.

In response to our specific aims:

1. Implant loading effects peri-implant plaque levels, gingival inflammation, and

probe depths.

2. Implant loading significantly affects PICF volume.

3. Implant loading significantly affects implant stability as measured by RFA.

75

4. Implant loading appears to influence selective inflammatory cytokines and

growth factors detectable in PICF.

Future Direction

Future direction may include further investigation of soft tissue response to different types of restorations, different restorative materials, different abutment types, evaluation of different molecules present in PICF, and more in depth investigation of peri-implant microflora.

76

References

[1] M. Esposito, J.-M. Hirsch, U. Lekhom and P. Thomsen, "Biological factors contributing to failures of osseointegrated oral implants (1). success crieteria and epidemiology," European Journal of Oral Science, vol. 106, pp. 527-551, 1998. [2] P. A. Fugazzotto, "Evidenc-Based decision making: replacement of the single missing tooth," Dental Clinics of North America, vol. 53, pp. 97-129, 2009. [3] B. E. Pjetursson, U. Bragger, N. P. Lang and e. al., "Comparison of survival and complication rates of tooth-supported fixed dental prostheses (FPDs) and implant-supported FPDs and single crowns (SCs)," Clinical Oral Implant Research, vol. 18, no. 3, pp. 97-113, 2007. [4] T. Albrektsson, G. Zarb, P. Worthington and e. al., "The long-term efficiency of currently used dental implants: a review and proposed criteria of success," International Journal of Oral Maxillofacial Implants, vol. 1, pp. 11-25, 1986. [5] D. E. Smith and G. A. Zarb, "Criteria for success of osseointegrated endosseous implants," Journal fo Prosthetic Dentistry, vol. 62, pp. 567-572, 1989. [6] "Acedemy of Prosthodontics. The Glossary of Prosthodontic Terms," Journal of Prosthestic Dentistry, vol. 95, no. 1, pp. 1-83, 1005. [7] I. Damjanov, Pathlogy for the health professional, Philidelphia: Elsevier Saunders, 2006. [8] J. Davies, "Mechanisms of endosseous integration," International Journal of Prosthodontics, vol. 11, no. 5, pp. 391-401, 1998. [9] U. M. Wikesjo, R. E. Nilveus and K. A. Selvig, "Significance of early healing events on periodontal repari: a review," Journal of Periotontology, vol. 63, no. 3, pp. 158-165, 1992. [10] D. L. Cochran, J. S. Hermann, R. K. Schenk, F. L. Higginbottom and D. Buser, "Biologic Width around titanium implants. A histormeric analysis of the implant-gingival junction around unloaded and loaded nonsubmerged implants in the canine mandible.," Journal of Periodontology, vol. 68, no. 2, pp. 186-198, 1997. [11] H. P. Weber, D. Buser, K. Donath, J. P. Fiorellini, V. Doppalapudi, D. W. Paquette and R. C. Williams, "Comparison of healed tissue adjacent to submerged and non-submerged unloaded titanium dental implants. A histomeric study in beagle dogs.," Clin Oral Impl Res, vol. 7, pp. 11-19, 1996. [12] J. S. Hermann, D. Buser, R. K. Schenk, F. L. Higginbottom and D. L. Cochran, "Biologic width around titanium implants. A physiologic formed and stable dimension over time.," Clin Oral Impl Res, vol. 11, pp. 1-11, 2000. [13] J. S. Hermann, D. Buser, R. K. Schenk, J. K. Schoolfield and D. L. Cochran, "Biologic width around one- and two-piece titanium implants," Clin Orl Impl Res, vol. 12, pp. 559-571, 2001. [14] H. Ikeda and e. al., "Ultrastructural and immunoelectron microscopic studies of the peri- 77

implant epeithelium-implant (Ti-6Al-4V) interfaces of rat maxilla.," Journal of Periodontology, vol. 71, no. 6, pp. 961-973, 2000. [15] I. Atsuta, T. Yamaza, M. Yoshinari, T. Goto, M. A. Kido, T. Kagiva, S. Mino, M. Shimono and T. Tanaka, "Ultrastructual localization of laminin-5 (gamma 2 chain) in rat peri-implant oral mucosa around atitanium-dental implant by immune-electron microscopy.," Biomaterials, vol. 26, pp. 6280-6287, 2005. [16] R. A. James and R. L. Schultz, "Hemidesmosomes and the adhesion of junctional epithelial cells to metal implants.," Clin Oral Impl Res, vol. 74, no. 4, pp. 294-302. [17] M. A. Listgarten and e. al., "Light and transmission electron microscopy of the intact interfaces between non-submerged titanium-coated epoxy resin implants and bone or gingiva.," Journal of Dental Research, vol. 71, no. 2, pp. 364-367, 1992. [18] C. M. Stanford and R. A. Brand, "Toward an understanding of implant occlusion and strain adaptive bone modeling and remodeling.," Journal of Prosthetic Dentistry, vol. 81, no. 5, pp. 553-561, 1999. [19] H. Roesler, "The history of some fundamental concepts in bone biomechanics.," Journal of Biomechanics, vol. 20, no. 11/12, pp. 1025-1034, 1987. [20] H. V. von Meyer, "Die Architectur der Spongiosa," Reichert und DuBois Raymond Archive, vol. 34, pp. 615-628, 1867. [21] S. A. Cowin, "Selection of the Editor; The graphical statics: Bridges and Bones," ASME News, vol. 23, no. 4, 2004. [22] J. Wolff, "The law of bone remodeling Maquet P and Furlong R translators," Berlin: Springer Verlag, 1986. [23] J. Wolff, "Das Gesetz der Functionellen anpassung," Hirschawald Berlin, 1982. [24] W. Roux, "Der zuchende Kampf der Theile, oder die Thielauslese im organismus Zugleich eine "theorie der functionellen anpassung.," Leipzig: Verlag von Wilhelm Engelmann, 1881. [25] H. M. Frost, "Bone "mass" and the "mechanostat": A proposal.," Journal of Prosthetic Dentistry, vol. 81, no. 5, pp. 553-561, 1999. [26] H. M. Frost, "A determinant of bone architecture: The minimum effective strain.," Clin Orthop, vol. 175, pp. 286-292, 1983. [27] C. H. Turner, I. Owan and Y. Takano, "Mechanostransduction in bone: role of strain rate.," American Journal of Physiology, vol. 269, no. 3, pp. 438-442, 1995. [28] M. R. Forwood and C. H. Turner, "Skeletal adaptations to mechanical usage: results from tibial loading studies in rats.," Bone, vol. 17, no. 4 suppl, pp. 197S-205S, 1995. [29] D. Jones, G. Leivseth and J. Tenbosch, "Mechano-reception in osteoblast-like cells," Biochem Cell Biol, vol. 73, pp. 525-534, 1995. [30] M. G. Mullender and R. Huiskies, "Osteocytes and bone lining cells: which are the best candidates for mechano-sensrs in cancellous bone?," Bone, vol. 20, pp. 527-532, 1997. [31] I. Owan, D. B. Burr, C. H. Turner, J. Qiu, Y. Tu, J. E. Onvia and a. et, "Mechanotransduction in bone: osteoblasts are more responseive to fluid forces than mechanical stain.," American Journal of Physiology, vol. 273, no. 3 pt 1, pp. C810-5, 1997.

78

[32] J. Klein-Nulend, A. van der Plas, C. M. Semeins, N. E. Ajubi, J. A. Frngos and P. J. Nijweide, "Sensitivity of osteocytes to biomechanical stress in vitro," FASEB J, vol. 9, pp. 441-445, 1995. [33] S. B. Doty, "Morphological evidence of pag unctions between bone cells.," Calcif Tissue Int, vol. 33, pp. 509-512, 1981. [34] C.-L. Lin, J. C. Wang, L. C. Ramp and P.-R. Liu, "Biomechanical response of implant systems placed in the maxillary posterior region under various conditions of angulation, bone density, and loading," Int J Oral Maxillofac Imp, vol. 23, no. 1, pp. 57-64, 2008. [35] N. L. Clellend and A. Gilat, "The effect of abutment angulation on stress transfer for an implant.," J Pros, vol. 1, pp. 24-28, 1992. [36] M. W. Bidez and C. E. Misch, "Issues in bone mechanics related to oral implants.," Implant Dentistry, vol. 1, no. 4, pp. 289-294, 1992. [37] E. Kitamura, R. Stegaroiu, S. Nomura and O. Miyakawa, "Biomechanical aspects of marginal bone resorption around osseointegrated implants: considerations based on a three-dimensional finite element analysis.," Clin Oral Impl Res, vol. 15, no. 4, pp. 401-412, 2004. [38] R. Skalak, "Biomechanical considerations in osseointegrated prostheses.," J Pros Dent, vol. 49, no. 6, pp. 843-848, 1983. [39] M. Cruz, T. Wassail, M. Toledo, L. P. da Silva Barra and S. Cruz, "Finite element stress analysis of dental prostheses supported by straigth and angled implants.," Int J Oral Maxillofac Impl, vol. 24, no. 3, pp. 391-403, 2009. [40] G. Papavasiliou, P. Kamposiora, S. C. Bayne and D. A. Felton, "Three-dimensional finite element analysis of stress-distribution around single tooth implants as a function of boney support, prosthesis type, and loading during function.," J Pros Dent, vol. 76, no. 6, pp. 633- 640, 1996. [41] U. Meyer, D. Vollmer, C. Runte, C. Bourauel and U. Joos, "Bone loading pattern around implants in average and atrophic edentulous maxillae: a finite-elemental analysis.," J Cranio-Maxillofac Surg, vol. 29, pp. 100-105, 2001. [42] Piatelli, A et al., "An histologic and histomorphometric study of bone reactions to unloaded and loaded non-submerged single implants in monkeys: a pilot study.," J Oral Impl, vol. 19, no. 4, pp. 314-320, 1993. [43] S. J. Hoshaw, J. B. Brunski and G. V. Cochran, "Mechanical loading of Branemark implants affectes interfacial bone modeling and remodeling.," JOMI, pp. 345-360, 1994. [44] Roberts, W E; et al., "Osseous adaptation to continuous loadign of rigid endosseous implants.," Am J Orthod, vol. 86, no. 2, pp. 95-111, 1984. [45] K. Gotfredsen, T. Berglundh and J. Lindhe, "bone reactions adjacent to titanium implant subjected to static load. A study in the dog (I).," Clin Oral Impl Res, vol. 12, no. 1, pp. 1-8, 2001. [46] K. Gotfredsen, T. Berglundh and J. Lindhe, "Bone reactions adjacent to titanium implanst with different surface characteristics subjected to static load. A study in the dog (II).," Clin Oral Impl Res, vol. 12, no. 3, pp. 196-201, 2001. [47] Asikainen, P; et al., "Titanium implants and lateral forces. An expeimental study with 79

sheep.," Clin Oral Impl Res, vol. 8, no. 6, pp. 465-468, 1997. [48] L. Bakaeen, P. Quinlan, N. P. Lang and D. L. Cochran, "The biologic width around titanium implants: histometric analysis of the implantogingival junction around immediately and early loaded implants.," Int J Perio Rest Dent, vol. 29, no. 3, pp. 297-305, 2009. [49] Assneza, B; et al., "Crestal bone remodeling in loaded and unloaded implants and the microgap: a histologic study.," Impl Dent, vol. 12, no. 3, pp. 235-241, 2003. [50] I. Ericsson and J. Lindhe, "Probing depth at implants and teeth. An experimental study with the dog.," J Clin Periodont, vol. 20, pp. 623-627, 1993. [51] N. P. Lang, A. C. Wetzel, H. Stich and R. G. Caffesse, "Histologic probe penetration in healthy and inflamed peri-implant tissues.," Clin Oral Impl Res, vol. 5, pp. 191-201, 1994. [52] M. G. Newman and T. F. Flemmig, "Periodontal considerations of implants and implant associated microbiota," J Dent Educ, vol. 52, pp. 737-744, 1988. [53] G. R. Bauman, M. Mills, J. W. Rabley and W. H. Hallmon, "Clinical parameters of evaluation during implant maintenace.," Int J Oral Maxillofac Impl, vol. 7, pp. 220-227, 1992. [54] A. Lekholm, R. Adell, J. Lindhe, P.-I. Branemark, B. Eriksson, B. Rockler, A.-M. Lindvall and T. Yoneyama, "Marginal tissue reactions at osseointegrated titanium fixtures. (II) A cross- sectional retrospective study.," Int J Oral Maxillofac Surg, vol. 15, pp. 53-61, 1986. [55] M. E. Gher and A. C. Richardson, "The accuracy of dental radiographic techniqes used for evaluation of implant fixture placement.," Int J Perio Rest Dent, vol. 15, no. 3, pp. 269-283, 1995. [56] A. B. Reiskin, "Implant imaging. Status, controversies, and new developments.," Dent Clin N Am, vol. 42, no. 1, pp. 47-56, 1998. [57] C. C. Wyatt and M. J. Pharoah, "Imaging techniqes and image interpretation for dental implant treatment.," Int J Posth, vol. 11, no. 5, pp. 442-452, 1998. [58] T. Dreiseidler, R. A. Mischkowski, J. Neugebauer, L. Ritter and J. E. Zoller, "Comparison of cone-beam imaging with orthopantomoography and computerized tomography for assessment in presurgical implant dentistry.," Int J Oral Maxillofac Imp, vol. 24, pp. 216- 225, 2009. [59] W. C. Scarfe and A. G. Farman, "What is cone-beam CT and how does it work?," Dent Clin N Am, vol. 52, pp. 707-730, 2008. [60] J. M. Kaneko, "Dynamics of the periotest method of diagnosing the dental implant-bone interface.," Journal of Material Science: Materials in Medicine, vol. 4, pp. 256-259, 1993. [61] J. Olive and C. Aparicio, "Periotest method as a measure of osseintegrated oral implant stability.," Int J Oral Maxillofac Implants, vol. 5, no. 4, pp. 390-400, 1990. [62] C. Aparicio, N. P. Lang and B. Ranger, "Validity and clinical significance of biomechanical testing of implant/bone interface.," Clin Oral Impl, vol. 17, no. 2, pp. 2-7, 2006. [63] N. Meredith, "Assessment of implant stability as a prognostic determinant," Int J Prosthodont, vol. 11, no. 5, p. 501, 1998. [64] N. Meredith, D. Alleyne and P. Cawley, "Quantitative determinatin of the stability of the implant-tissue interface using resonance frequency analysis.," Clin Oral Impl Res, vol. 7, no. 3, pp. 261-267, 1996.

80

[65] Osstell, "www.ostell.com," Ostell AB, Gamlestadsvägen 3B, SE 415 02 Göteborg, Sweden, [Online]. Available: http://www.osstell.com/$-1/25038-03-en-osstell-isq-brochure-lr.pdf. [Accessed 14 April 2012]. [66] N. Meredith, D. Alevne and P. Cawley, "Quantitative determination of the stability of the implant-tissue interface using resonance frequency analysis.," Clin Oral Impl Res, vol. 7, pp. 261-267, 1996. [67] N. Meredith, F. Shagaldi, D. Alvane, L. Sennerdy and P. Cawley, "The application of resonance frequency measurement to study the stability of titanium implants during healing in the rabbit tibia.," Clin Oral Impl Res, vol. 8, pp. 234-243, 1997. [68] P. O. Ostman, M. Hellman, I. Wendelhag and L. Sennerby, "Resonance frequency analysis measurements of implants at placement surgery.," Int J Prosthodont, vol. 19, no. 1, pp. 77- 83, 2006. [69] M. A. Huwiler, B. E. Pietersson, D. D. Bosshart, G. E. Salvi and N. P. Lang, "REsonance frequency analysis in relation to jawbone characteristics and during healing of implant installation.," Clin Oral Impl Res, vol. 18, pp. 275-280, 2007. [70] B. Friberg, L. Sennerby, B. Linden, K. Grondahl and U. Lekhom, "Stability measurments of one-stage Branemark implants during healing in mandibles. A clinical resonance frequency analysis study.," Int J Maxillofac Surg, vol. 28, pp. 266-272, 1999. [71] I. Abrahamsson, E. Linder and N. P. Lang, "Implant stability in relation to osseointegration: an experimental study in the labrador dog.," Clin Oral Impl, vol. 20, pp. 313-318, 2009. [72] M. Sjostrom, S. Lundgren, H. Nilson and L. Sennerby, "Monitoring of implant stability in grafted bone using resonance frequency analysis. A clinical study from implant placement to 6 months of loading.," Int J Clin Oralmaxillo Surg, vol. 35, pp. 45-51, 2005. [73] R. Glauser, L. Sennerby, N. Meredith, A. N. Ree, A. Lundgren, J. Gottlow and C. Hammerle, "Resonance frequency analysis of implant subjected to immediate or early functional loading. Successful vs. failed implants.," Clin Oral Impl Res, vol. 15, pp. 428-434, 2004. [74] M. Bischof, R. Nedir, S. Szmukler-Moncler, J.-P. Bernard and J. Samson, "Implant stability measurement of delayed and immediately loaded implant during healing. A clinical resonance-refrequency analysis study with sandblasted-and-etched ITI implants.," Clin Oral Impl, vol. 15, pp. 529-539, 2004. [75] R. Nedir, M. Beschof, S. Szmukler-Moncler, J.-P. Bernard and J. Samson, "Predicting osseointegration by means of implant primary stability. A resonance-frequency analysis study with delayed and immediately loaded ITI SLA implants.," Clin Oral Impl Res, vol. 15, pp. 520-528, 2004. [76] M. C. Alfano, "The origin of gingival fluid.," J Theor Biol, vol. 47, pp. 127-136, 1974. [77] J. M. Goodson, "Gingival crevicular fluid flow.," Periodont 2000, vol. 31, pp. 43-54, 2003. [78] D. H. Pashley, "A mechanistic analysis of gingival fluid production.," J Peridont Res, vol. 11, pp. 121-134, 1976. [79] H. Loe, "Physiologic aspects of the gingival pocket, an experimental study.," Acta Odont Scand, vol. 19, pp. 387-395, 1961. [80] E. Weinstein and J. D. Mandel, "The fluid of the gingval sulcus.," J Periodont, vol. 2, no. 4, pp. 147-153, 1964. 81

[81] J. Bang, G. Cimansoni, C. Rosenbusch and A. Duckert, "Sodium, potassium, and calcium contents of crevicular exudate: their relations to gingivitis and periodontitis.," J Periodont, vol. 44, no. 12, pp. 770-774, 1973. [82] M. T. Pollanen, J. I. Salonen and V.-J. Uitto, "Structure and function of tooth-epithelial interface in health and disease.," J Periodont, vol. 31, pp. 12-31, 2000. [83] G. Armitage, "Analysis of gingival crevice fluid and risk of progression of periodontitis.," Periodont, vol. 34, pp. 109-119, 2000. [84] McGee, J M; et al., "The relationship between concentrations of proinflammatory cytokines within gingiva and the adjacent sulcular depth.," J Periodont, vol. 69, no. 8, pp. 865-871, 1998. [85] Engebreston, S P; et al., "The influence of interleukin gene polymorphism on expression of interleukin-1 beta and tumor necrosis factor-alpha in periodontal tissue and gingival crevicular fluid.," J Periodont, vol. 70, no. 6, pp. 567-573, 1999. [86] C. M. Champagne, W. Buchanan, M. S. Reddy, J. S. Preisser, D. Beck and S. Offerbacher, "Potential for crevice fluid measures as predictors of risk for periodontal disease.," J Periodont, vol. 31, pp. 167-180, 2000. [87] C. M. Figueredo, M. S. Ribeiro, R. G. Fischer and A. Gustafsson, "Increased interleukin-1 beta concentration in gingival crevicular fluid as a characteristic of periodontitis.," J Periodont, vol. 10, no. 2, pp. 1457-1463, 1999. [88] I. B. Lamster and J. K. Ahlo, "Analysis of gingival crevicular fluid as applied to the diagnosis of oral and systemic disease," Ann NY Acad Sci, vol. 1098, pp. 216-229, 2007. [89] W. V. Giannoblie, K. R. Al-Shammari and D. P. Sarment, "Matrix molecules and growth factors as indicators of periodtnal disease activity.," J Periodont, vol. 31, pp. 125-134, 2000. [90] J. L. Ebersole, "Humoral innume response in gingival crevice fluid: local and systemic implications.," J Periodont, vol. 31, pp. 135-166, 2000. [91] B. G. Loos and S. Tjoa, "Host-derived diagnostic markers for periodontitis: do they exist in gingival crevicular fluid?," J Periodont, vol. 39, pp. 53-72, 2000. [92] F. Javed, K. Al-Hezaimi, Z. Salameh, K. Almas and G. E. Romanos, "Proinflammatory cytokines in the crevicular fluid of patients iwth peri-iplantitis.," Cytokine, vol. 53, pp. 8-12, 2011. [93] P. Apse, R. P. Ellen, C. M. Overall and G. A. Zarb, "Microbiota and crevicular fluid collagenase activity in the osseointegrated dental implant sulcus: a comparison of sites in edentulous and parital edentulous patients.," J Periodontal Res, vol. 24, no. 2, pp. 96-105, 1989. [94] E. Adonogianaki, J. Mooney, J. L. Wennstrom, U. Lekholm and D. F. Kinane, "Acute-phase proteins and immunoglobulin G against Pohphyromonas gingivalis in peri-implant crevicular fluid: a comparison with gingival crevicular fluid.," Clin Oral Impl Res, vol. 6, no. 1, pp. 14-23, 1995. [95] S. Gunday, A. O. Topcu, G. N. Guncu, A. C. Akman, E. Karabulut and N. Yamalik, "Analysis of potential factors affecting peri-implant sulcus fluid (PISF) volume.," Clin Den Res, vol. 10, no. 3, pp. 166-173, 2011. 82

[96] C. B. Beck, G. Embery, M. S. Langley and R. J. Waddington, "Levels of glycosaminoglycans in peri-implant sulcus fluid as a means of monitoring bone response to endosseous dental implants.," Clin Oral Impl Res, vol. 2, no. 4, pp. 179-185, 1991. [97] K. S. Last, S. Smith and N. Pender, "Monitory of IMZ titanium endosseous dental implants by glycosaminoglycan analysis of peri-implant sulcus fluid.," Int J Oral Maxillofac Impl, vol. 10, no. 1, pp. 58-60, 1995. [98] E. G. Kaklamanos and L. Tsalikis, "A review on peri-implant crevicular fluid asays potential in montinroing and predicting peri-implant tissue responses.," J Int Acad Periodont, vol. 4, no. 2, pp. 49-59, 2002. [99] M. Murata, J. Tatsumi, Y. Kato, S. Suda, Y. Nunokawa, Y. Kobayashi, H. Takeda, H. Araki, S. K, O. K, T. Miyata and H. Yoshie, "Osteocalcin, deosypuradinonline, and interluekin-1 beta in peri-implant crevicular fluid of patients with peri-implantitis.," Clin Oral Impl Res, vol. 13, pp. 637-643, 2002. [100] G. Scheirano, G. Pejrone, A. Trombetta, G. Martinasso, G. Preti and R. A. Canuto, "TNF- alpha TGF-beta2, and IL-1beta levels in gingival an dperi-implant crevicular fluid before and after de noe plaque accumulation.," J Clin Periodont, vol. 35, no. 6, pp. 532-538, 2008. [101] H. Nowzari, J. E. Botero, M. DeGiacomo, M. Villacres and S. K. Rich, "Microbiology and cytokine levels around healthy dental implants and teeth.," Clin Imipl Dent Rel Res, vol. 10, no. 3, pp. 166-173, 2008. [102] H. Nowzari, S. Phamduong, J. E. Botero, M. C. Villacres and S. K. Rich, "The profile of inflammatory cytokines in gingival clrevicular fluid around healthy osseointegrated implants.," Clin Impl Dent Rel Res, pp. 1-7, 2010. [103] J. Silness and H. Loe, "Periodontal diseasein pregnancy. II. Correlation between oral hygeine and periodontal condition.," Acta Odontol Scand, vol. 22, pp. 121-135, 1964. [104] H. Loe, "Then gingival index, the plaque index, and the retention index systems.," J Periodont, vol. 38, no. 6, pp. 610-616, 1967. [105] J. Sliness and H. Loe, "Oral Health Database," Malmo University, 28 February 2012. [Online]. Available: http://www.mah.se/CAPP/Methods-and-Indices/Oral-Hygiene- Indices/Silness-Loe-Index/. [Accessed 28 April 2012]. [106] A. Mombelli, M. A. van Oosten, E. Schurch and N. P. Lang, "The microbiota associated with successful or failing osseointerated titanium implants.," Oral Microbiology and Immunology, pp. 145-151, 1987. [107] Newman, Takei, Klakkevold, Carrnaza, J. D. Beck and S. J. Arbes Jr., "Epidemiology of Gingival and Periodontal Disease," in Carranza's Clinical Periodontology, St. Lous, Saunders, 2006, pp. 115-117. [108] P. Kumar, A. Griffen, J. Barton, B. Paster, M. Moeschberger and E. Lays, "New bacterial species associated with chronic periodontitis.," J Dent Res, vol. 82, pp. 338-344, 2003. [109] P. S. Kumar, M. R. Mason, M. R. Brooker and K. O'brien, "Pyrosequencing reveals unique microbial signatures associated with healthy and failing dental implants.," J Clin Periodontol, 2012. [110] K. Hubel, D. Dale and W. Liles, "Therapeutic use of cytokines to modulate phagocyte function... of infectious diseases: current status of granulocyte colony stimulationg factor, 83

granulocye-macrophage colony stimulating factor, macrophage colony-stimulating factor, and interferon-gamma.," J Infect Disease, vol. 185, pp. 1490-501, 2002. [111] Gotfredsen , K; et al., "Histological and histomophomoterical evaluation of tissue reactions adjacent to endosteal implant in monkeys.," Clin Oral Impl Res, vol. 2, pp. 30-37, 1991. [112] Masaaki, S; et al., "The effects of early occlusal loading on one-stage titanium alloy implants in beagle dogas: a pilot study.," J Prosthet Dent, vol. 69, pp. 281-288, 1993. [113] Bakaeen, L; et al., "The biologic width around titanium implants: histomeric analysis of the implantogingival junction around immediately and early loaded implants.," International Journal of Periodontics and Restorative Dentistry, vol. 29, no. 3, pp. 297-305, 2009. [114] S. M. Balaji, "Tobacco smoking and surgical healing of oral tissues: a review.," Indian Joural of Dental Research, vol. 4, no. 344-348, p. 19, 2008. [115] R. J. Weyant, "Characteristics associated with the loss of peri-implant tissue health of endosseous dental implants.," Int J Oral Maxillofac Imp, vol. 9, pp. 95-102, 1994. [116] L. J. Heitz-Mayfield and G. Huynh-Ba, "History of treated periodontitis and smoking as risks for implant therapy.," In J ORal Maxillofac Impl, vol. 19, no. 1, pp. 107-115, 2004. [117] L. M. Gorman, P. M. Lambert, H. F. Morris, S. Ochi and S. Winkler, "The effect of smoking on implant survival at second-stage surgery: DICRGinterim report No 5: Dental implant clinical research group.," Implant Dent, vol. 3, pp. 165-168, 1994. [118] C. A. Bain and P. K. Moy, "The association between the failure of dental implants and cigarette smoking.," Int J Oral Maxillofac Impl, vol. 8, pp. 609-615, 1993. [119] H. de Bruyn and B. Collaert, "The effect of smoking on early implant failure.," Clin Oral Implant Res, vol. 5, pp. 260-264, 1994. [120] J. P. Becktor, S. Isaksson and L. Sennerby, "Survival analysis of endosseous implants in grafted and nongrafted edentulous maxillae.," Int J Oral Maxillofac Impl, vol. 19, no. 1, pp. 107-115, 2004. [121] E. O. Butcher and J. Klingsberg, "Age, gonadectomy, and wound healing in the palatal mucosa.," J Dent Res, vol. 40, p. 694, 1961. [122] P. Holm-Pedersen and H. Loe, "Wound healing in the gingiva of young and old individuals.," Scand J Dent Res, vol. 79, p. 40, 1971. [123] R. A. Smith, R. Berger and T. B. Dodson, "Tisk factors asociated with dental implants in healthy and medically compromised patients.," In J Oral Maxillofac Impl, vol. 7, no. 3, pp. 367-372, 1992. [124] S. R. Bryant, "Osseointegration of oral implants in older and younger adults.," Int J Oral Maxillofac Impl, vol. 13, no. 4, pp. 492-499, 1998. [125] M. J. Meijer, R. H. Batenburg and G. M. Raghoebar, "Ingluence of patient age on the success rate of dental implants supporting and overdentures in an edentulous mandible: A 3-year prospective study.," Int J Oral Maxillofac Impl, vol. 16, no. 4, pp. 522-526, 2001. [126] D. W. Paquette, R. Brodala and T. B. Dodson, "Risk factors for end osseous implant failure.," Dent Clin North Amer, vol. 50, no. 3, pp. 361-374, 2006. [127] C. E. Misch, "The effect of bruxism on treatment planning for dental implants.," The effect of bruxism on treatment planning for dental implants., Vols. 76-81, no. 21, p. 2002, Dent

84

Today . [128] Bragger, U; et al., "Biological and technical complications and failures with fixed partial dentures (FPD) on mplants and teeth after four to five years of function.," Clin Oral Implants Res, vol. 12, pp. 26-34, 2001. [129] N. J. Attard and G. A. Zarb, "Implant prosthodontics management of paritally edentulous patients missing posterior teeth: The Toronto experience.," J Prosthet Dent, vol. 89, no. 4, pp. 352-359, 2003. [130] R. A. Jaffin and C. L. Berman, "The excessive loss of Branemark fixtures in type IV bone: A 5-year analysis.," J Periodontol, vol. 62, pp. 2-4, 1991. [131] Da Silva, J K; et al., "Influence of site on implant survival: 6-year results.," J Dent Res, vol. 71, p. 256, 1992. [132] O. Bahat, "Treatment planning and placement of implants in the posterior maxilla: REport of 732 consecutive Nobelpharma implants.," Int J Oral Maxillofac Impl, vol. 8, no. 2, pp. 151-161, 1991. [133] Romanos, G E; et al., "Bone-implant interface around titanium implants under different loading conditions: a histomorphometrical analysis in the Macaca fascicularis monkey.," J Periodont, vol. 74, no. 10, pp. 1483-90, 2003. [134] Mathieu, V; et al., "Influence of healing time on the echographic response of the bone- implant interface.," J Acoustical Soc Amer, vol. 131, no. 4, p. 3368, 2012. [135] A. B. Carr, D. A. Gerard and P. E. Larsen, "Histomorphomoteric analysis of implant anchorage for 3 types of dentalimplants following 6 months of healing in baboon jaw bones.," Int J Oral Maxillofac Impl, vol. 15, no. 6, pp. 785-91, 2000. [136] Papaiannou, W; et al., "The effect of periodontal parameters on the subgingival microbiota around implants.," Clin Oral Impl Res, vol. 6, no. 4, pp. 197-204, 1995. [137] A. Mombelli and R. Mericske-Stern, "Microbiological features of stable osseointegrated implants used as abutments for overdentures.," Clin Oral Impl Res, vol. 1, no. 1, pp. 1-8, 1990. [138] M. Quirynen and M. Listgarten, "Distribution of bacterialmorphotypes around natural teeth and titanium implants ad modum Branemark.," Clin Oral Impl Res, vol. 1, no. 1, pp. 8-12, 1990. [139] K. Ababnaeh, B. Al-Omari and T. Alawneh, "The effect of dental restoration type andmaterial on periodontal health.," Oral Health Prev Dent, vol. 9, no. 4, pp. 395-403, 2011. [140] C. Chan and H. Weber, "Plaque retention on theeth restored with full-ceramic crowns: a comparative study.," J Prosthet Dent, vol. 56, no. 6, pp. 666-71, 1986. [141] Quirynen, M; et al., "The clnical relevnce of surface energy and froughness on supragingival and subgingival plaque formation in man.," Colloids and Surfaces B Biointerfaces, vol. 2, pp. 25-31, 1994. [142] Teugheles, W; et al., "Effect of materials charachteristics and/or surfacr topography on biofilm development.," Clin Oral Impl Res, vol. 17 (suppl 2), pp. 68-81, 2006. [143] Venza, M; et al., "Changes in inflammatory mediators in peri-implant fluid after implant

85

insertion.," J Periodont, vol. 80, no. 2, pp. 297-306, 2009. [144] Vered, Y; et al., "Teeth and implant surroundings: clincial health indices and microbiologic parameters.," Quintess Int, vol. 42, no. 4, pp. 339-44, 2011. [145] Quirynen, M; et al., "The reliability of pocket probing around screw-type implants.," Clin Oral Impl Res, vol. 2, no. 4, pp. 186-92, 1991. [146] H. Zigdon and E. Machtei, "The dimensions of keratinized mucosa around implants afect clinical and immunological parameters.," Clin Oral Impl Res, vol. 19, no. 4, pp. 387-92, 2008. [147] M. Block and J. Kent, "Factors associated with soft and hard tissue compromise of endosseous implants.," J Oral Maxillofac Surg, vol. 48, pp. 1153-60, 1990. [148] Bouri Jr., A; et al., "Width of keratinized gingiva and the health status of the supporting tissues around dental implants.," Int J Oral Maxillofac Impl, vol. 23, no. 2, pp. 323-6, 2008. [149] Schrott, AR; et al., "Five-year evaluation of the influence of keratinized mucosa on peri- implant soft-tissue health and stability around implants supporting full-arch mandibular fixed prostheses.," Clin Oral Impl Res, vol. 20, no. 10, pp. 1170-7, 2009. [150] D. Chung and e. al., "Significance of keratinized mucosa in maintenacne of dental implants with different surfaces.," J Periodont, vol. 77, no. 8, pp. 1410-20, 2006. [151] J. Wennstrom, F. Bengazi and U. Lekholm, "The influence of the masticatory mucosa on the peri-implant soft tissue condition.," Clin Oral Impl Res, vol. 5, no. 1, pp. 1-8, 1994. [152] F. Bengazi, J. Wennstrom and U. Lekholm, "Recession of the soft tissue margin at oral implants. A 2-year longitudinal prospective study.," Clin Oral Impl Res, vol. 7, no. 4, pp. 303-10, 1996. [153] J. Strub, T. Baberthuel and U. Grunder, "The role of attached gingiva in the health of periimplant tissue in dogs. 1. Clinical findings.," Int J Periodont Rest Dent, vol. 11, pp. 317- 33, 1991. [154] Z. Karabuda, M. Backstrom and L. Sennerby, "Immediate and early loading of oxidized tapered implants in the partially edentulous maxilla: a 1-year prospective clinical, radiographic, and resonance frequency analysis study.," Clin Impl Dent Rel Res, vol. 11, no. 2, pp. 69-80, 2009. [155] K. Fischer, M. Backstrom and L. Sennerby, "Immediate and early loading of oxidized tapered implants in the partially edentulous maxilla: a 1-year prospective clinical, radiographic, and resonance frequency analysis study.," Clinc Impl Dent Rel Res, vol. 11, no. 2, p. 69080, 2009. [156] G. Gunco and e. al., "Relationships between implant sability, image-based measures and nitric oxide levels.," J Oral Rehab, vol. 35, no. 10, pp. 745-53, 2008. [157] K. Guzaitis, K. Knoernschild and M. Viana, "Effect of repeated screw joint closing and opening cycles o nimplant prosthetic scre reverse torque and implant screw thread morphology.," J Prosthet Dent, vol. 106, no. 3, pp. 159-69, 2011. [158] M. Cardosa and e. al., "Torque removal evaluation of prosthetic screws afer tightening and loosening cycles: an in vitro study.," Clin Oral Impl Res, vol. 23, no. 4, pp. 475-80, 2012. [159] C. Ricciardi and e. al., "Effect of repeated torque/mechanical loading cycles on two

86

different abutmetn types in implants with interal tapered connections: an in vitro study.," Clin Oral Impl Res, vol. 20, no. 6, pp. 624-32, 2009. [160] T. Grandi and e. al., "Immediate positioning of definitive abutments versus repeated abutment replacements in immediately loaded implants: effects on bone healing at the 1- year follow-up of a multicentre renadomised controlled trial.," Eur J Oral Impl, vol. 5, no. 1, pp. 9-16, 2012. [161] L. Canullo and e. al., "Immediate positioning of a definitive abutment versus repeated abutment replacements in post-extractive implants: 3-year follow-up of a randomised multicenter clinical trial.," Eur J Oral Impl, vol. 3, no. 4, pp. 285-96, 2010. [162] T. Tozum, A. Akman, N. Yamalik, I. Tulunoglu, I. Turkyilmaz, E. Karabulut, K. Kilinic and M. Cehreli, "Analysis of the inflammatory process around endosseous dental implants and natural teeth: meloperoxidase level and nitric oxide metabolism.," Int J Oral Maxillofac Impl, vol. 22, no. 6, pp. 969-79, 2007. [163] Tozum,T F; et al., "The impact of prosthetic design on the stability, marginal bone loss, peri-implant sulcus fluid volume, and nitric oxide metabolism of conventionally loaded endosseous dental implants: a 12-month clincial study.," J Periodont, vol. 79, no. 1, pp. 55- 63, 2008. [164] G. Griffiths, "Formation, collection, and significance of gingival crevicular fluid.," Periodont 2000, vol. 31, pp. 32-42, 2003. [165] H. Aboyoussef, C. Carter, J. J. Jandinski and F. S. Panagakos, "Detection of prostaglandin E2 and matrix metalloproteinases in implant crevicular fluid.," Int J Oral Max Impl, vol. 13, no. 5, p. 6890696, 1998. [166] F. S. Panagakos, H. Aboyoussef, R. Dondero and J. J. Jandinski, "Detection and measurement of inglammatory cytokines in implant crevicular fluid: a pilot study," Int J Orla Maxillofac Impl, vol. 11, no. 6, pp. 794-799, 1996. [167] H. Ataoglu, N. O. Alptekin, S. Haliloglu, M. Gursel, T. Ataoglu, B. Serpek and E. Durmus, "Interleukin-1 beta, tumor necrosis factor-alpha levels, and neutrophils elastase acivityin peri-implant crevicular fluid.," Clin Oral Impl Res, vol. 13, pp. 470-476, 2002. [168] R. T. Koa, D. A. Curtis, D. W. Richards and J. Preble, "Increased ingerleukin-1 beta in crevicular fluid of diseased implants.," Int J Oral Maxillofac Impl, vol. 10, no. 6, pp. 696- 701, 1995. [169] Spadaccio, C; et al., "A G-CSF functionalized PLLA scafold for wound repair: an in vitro preliniminary study.," Conference Proceedings: Annual INternational Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Conference., vol. 2010, pp. 843-6, 2010. [170] Yamada, T; et al., "Elevation of lymphocyte and hematopeitic stem cell numbers in mice transgenic for human granulocyte CSF.," Lab Invest, vol. 74, p. 384, 1996. [171] Ishida, K; et al., "Bone regeneration properties of granulocyte colony-stimulating factor via neovascularization and osteogenesis," Tissue Engineering Part A, vol. 16, no. 10, pp. 3271- 84, 2010. [172] K. Hubel, D. Dale and W. Liles, "Therapeutic use of cytokines to modulate phagocyte function for the treatment of infectious diseases: current status of granulocyte colony-

87

stimulating factor, granulocyte-macrophage colony-stimulating factor, macrophage colony-stimulating factor, and inte," J Infect Disease, vol. 185, no. 10, pp. 1490-501, 2002. [173] Matarasso, S; et al., "The effect of recombinant granulocyte colony-stimulating factor on oral and periodontal manifestations in a patient with cyclic neutropenia: a case report," Int J Dent, vol. 2009, p. 654, 2009. [174] Mifune, Y; et al., "Local dielivery of ganulocyte colony stimulationg factor-mobilized CD34- positive progenitor cellsusing bioscaffold for modality of unhealing bone fracture.," Stem Cells, vol. 26, p. 1395, 2008. [175] van de Geihn, GJ; et al., "Granulocyte colony-stimulating factor and its receptor in normal hematopoietic cell development and myeloid disease.," Rev Physiol Biochem Pharm, vol. 149, pp. 53-71, 2003. [176] Goultschin, J; et al., "The relationship between peripheral levels of leukocytes and neutrophils and periodontal disease status in a patient with congenital neutropenia," J Periodont, vol. 71, no. 9, pp. 1499-505, 2000. [177] K. Driscol, "Macrophage inflammatory proteins: biology and role in pulmonary inflammation.," Exp Lung Res, vol. 20, no. 6, pp. 473-90, 1994. [178] B. Lord, C. Heyworth and L. Woolford, "Macrophage inflammatory protein: its characteristics, biological properties and role in the regulation of haemopoiesis," J Hematology, vol. 57, no. 3, pp. 197-206, 1993. [179] M. Sanz, M. Newman, S. Nachnani, R. Holt, R. Stewart and T. Flemming, "Characterization of the subgingival microbial flora around endosteal sapphire dental implants in partially ednetulous patients.," Int J Oral Maxillofac Impl, vol. 5, pp. 247-253, 1990. [180] M. Quirynen, R. Vogels, W. Peters, v. Steenburghe, I. Naert and A. Haffajee, "Dynamics of initial subgingival colonization of pristine peri-implant pockets.," Clin Oral Impl Res, vol. 17, pp. 25-37, 2006. [181] K. Lee, M. Maiden, A. Tanner and H. Weber, "Microbiota of successful osseointerated dental implants.," J Periodont, vol. 70, no. 2, pp. 131-8, 1999. [182] W. Keller, U. Bragge and A. Mombelli, "Peri-implant microflora of implants with cemented and screw retained suprastructures.," Clin Oral Impl Res, vol. 9, no. 4, pp. 209-17, 1998. [183] G. Bouman, M. Mills, J. Rapley and W. Hallmon, "Plaque-induced inflammation around implants.," Int J Oral Maxillofac Impl, vol. 7, no. 3, pp. 330-7, 1992. [184] J. Botero, A. Gonzalez, R. Mercado, G. Olave and A. Contreras, "Subgingival microbiota in peri-implant mucosa lesions and adjacent teeth in paritially edentulous patients.," J Periodont, vol. 76, pp. 1490-5, 2005. [185] Adell, R; et al., "Marginal tissue reactions osseointegrated titanium fixtures (I). A 3-year longitudinal prospective study.," Int J Oral Maxillofac Impl, vol. 15, no. 1, pp. 39-52, 1986. [186] Meijndert, L; et al, "Microbiota around teeth and dental implants in periodontally healthy, paritally edendulous patients: is pre-implant microbiolgoic testing relevant?," Eur J Oral Science, vol. 118, no. 4, pp. 357-63, 2010. [187] Casado, PL; et al., "Identificatin of periodontal pathogens in healthy periimplant sites.," Impl Dent, vol. 20, no. 3, pp. 226-35, 2011.

88

[188] P. Kumar, M. Mason, M. Brooker and K. O'Brien, "Pyrosequencing reveals unique microbial signatures associated with healthy and failing dental implants.," J Clin Periodont, vol. 39, no. 5, pp. 425-33, 2012. [189] S. Dashper, C. Seers, K. Tan and E. Reynolds, "Virulence factors of the oral spirochete Treponema denticola.," J Dent Res, vol. 90, no. 3, pp. 691-703, 2011. [190] J. M. Salcetti, J. D. Moriarty, L. F. Cooper, F. W. Smith, J. G. Collins, S. S. Scransky and S. Offenbacher, "The clinical, microbial, and host response characteristics of the failing implant.," Int J Maxillofac Impl, vol. 12, no. 1, pp. 32-42, 1997. [191] J. Gamonal, A. Acevdo, A. Bascones, O. Jorge and A. Silva, "Levels of interleukin-1 beta, -8, and -10 and RANTES in gingival crevicular fluid and cell populations in adult periodontitis patients and the effect of periodontal treatment.," J Periodont, vol. 71, no. 10, pp. 1535- 1545, 2000. [192] S. Lachmann, E. Kimmerle-Muller, D. Axmann, L. Scheideler, H. Weber and R. Haas, "Associations between peri-implant crevicular fluid volume, concentrations of crevicular inflammtory mediators, and composite IL-1A-889 and IL-1B+3954 genotype.," Clin Oral Impl Res, vol. 18, pp. 212-223, 2007.

89