The impact of systemic and local factors on the incidence of oral implant failures

Transcription

1 Katholieke Universiteit Leuven Faculty of Medicine School of Dentistry, Oral Pathology and Maxillofacial Surgery Departement of Periodontology The impact of systemic and local factors on the incidence of oral implant failures Ghada Alsaadi Promoter: Prof. Dr. Daniel van Steenberghe Co-promoter: Prof. Dr. Marc Quirynen Submitted in partial fulfilment of the requirements to obtain the degree of Doctor in Medical Sciences Leuven, Belgium,

2 2008 Alsaadi Ghada All rights reserved. No part of this book may be reproduced or transmitted in any form by any means, electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the author. This thesis was conducted at the Department of Periodontology, School of Dentistry, Oral Pathology and Maxillofacial Surgery, Faculity of Medicine, Catholic University of Leuven. Kapucijnenvoer 33, B-3000 Leuven, Belgium.

3 ACKNOLEDGEMENTS I want to express my gratitude to Prof. A. Oosterlinck, former Rector, Prof. M. Vervenne, current Rector of the Katholieke Universiteit Leuven, Prof. G. Mannaerts, former Vicerector, Prof. M. Waer, current Vice-rector of Biomedical Sciences, Prof. J. Janssens, former Dean, Prof. B. Himpens, current Dean of the Faculty of Medicine, and Prof. F. Vinckier, Chairmen of the School of Dentistry, Oral Pathology and Maxillofacial Surgery, for the opportunity to undertake this PhD project. With great respect, I want to express my appreciation to my promoter Prof. Daniel van Steenberghe, for the opportunities he gave me. It was a great honor for me to be one of his last PhD students in his academic life. I want to thank hem for his guidance and his support during throughout all the past years in Leuven. I want also to thank my co-promoter Prof. Marc Quirynen, for his scientific guidance, and utmost support throughout the last years. I also want to thank the members of the thesis committee: Prof. Reinhilde Jacobs, Prof. Evert Schepers, Prof. Chantal Malevez and Prof. Gerry Raghoebaer for their critical review of this thesis manuscript and their helpful remarks. Numerous people have contributed to the research projects included in this thesis. To name but a few, I would like to thank in particular Dr. Arnôst Komàrek for the statistical analysis of this research, he was always open for any question in any time. Many thank goes to Prof. Reinhilde Jacobs and her past and current students and coworkers for their support and their friendship. Special thanks to Mrs Miranda Maréchal, the assistant manager of our department, the nurses of the operatory rooms and Karina Vranckx for their support and help during all the study period. I would also to thank in particular Mrs. Christel Dekeyser, associate head of department, Dr. Betty Vandekerckhove and the staff of the Department of Periodontology for their contribution and co-operation in the patient treatments. I am grateful to my past and present colleague-assistants and co-workers in the Department of Periodontology for the great working environment.

4 I want to thank Dr. Peter Schüpbach for the permission to use his micro CT as a cover of my thesis manuscript. Finally, I want to thank: my husband for his patience, tolerance and encouragements, my father for his utmost support and encouragements, and my mother, for her soul that is fighting to be alive in spite of being engaged in a sleeping body. From her I learned a lot and still am learning. For all of you --, for freedom, justice and peace --, for Palestine I want to dedicate this thesis. Ghada Leuven 13 November 2007

5 This thesis is based on several (to be published) papers in peer reviewed journals: Quirynen, M., Vogels, R., Alsaadi, G., Naert, I., Jacobs, R. & van Steenberghe, D. (2005) Predisposing conditions for retrograde peri-implantitis, and treatment suggestions. Clinical Oral Implants Research 16, Alsaadi, G., Quirynen, M. & van Steenberghe, D. (2006) The importance of implant surface characteristics in the replacement of failed implants. International Journal of Oral & Maxillofacial Implants 21, Alsaadi, G., Quirynen, M., Michiels, K., Jacobs, R. & van Steenberghe, D. (2007) A biomechanical assessment of the relation between the oral implant stability at insertion and subjective bone quality assessment. Journal of Clinical Periodontology 4, Alsaadi, G., Quirynen, M., Komarek, A. & van Steenberghe, D. (2007) The impact of local and systemic factors on the incidence of oral implant failures, up to abutment connection. Journal of Clinical Periodontology 7, Alsaadi, G., Quirynen, M., Komarek, A. & van Steenberghe, D. The impact of local and systemic factors on the incidence of late oral implant loss. Clinical Oral Implants Research (in press). Alsaadi, G., Quirynen, M., Michiels, K., Teughels, W., Komarek, A. & van Steenberghe, D. The impact of local and systemic factors on the incidence of failures up to abutment connection with modified surface implants. Journal of Clinical Periodontology (in press).

6

7 TABLE OF CONTENTS: CHAPTER I. Introduction Part 1. LOCAL AND SYSTEMIC FACTORS AND IMPLANT FAILURES CHAPTER II. II.1. II.2. II.3. II.4. II.5. II.6. CHAPTER III. III.1. III.2. III.3. III.4. III.5. III.6. CHAPTER IV. IV.1. IV.2. IV.3. IV.4. IV.5. IV.6. CHAPTER V. V.1. V.2. V.3. V.4. V.5. V.6. The impact of local and systemic factors on the incidence of oral implant failures up to abutment connection. Introduction Materials and methods Results Discussion Conclusion References The impact of local and systemic factors on the incidence of failures up to abutment connection with modified surface implants. Introduction Materials and methods Results Discussion Conclusion References The impact of local and systemic factors on the incidence of late oral implant loss. Introduction Materials and methods Results Discussion Conclusion References A biomechanical assessment of the relation between the oral implant stability at insertion and subjective bone quality assessment. Introduction Materials and methods Results Discussion Conclusion References 1

8 Part 2. IMPACT OF IMPLANT SURFACE CHARACTERISTICS ON IMPLANT FAILURES CHAPTER VI. VI.1. VI.2. VI.3. VI.4. VI.5. VI.6. The importance of surface characteristics of implants replacing failed ones Introduction Materials and methods Results Discussion Conclusion References CHAPTER VII. VII.1. VII.2. VII.3. VII.4. VII.5. VII.6. Predisposing conditions for retrograde peri-implantitis, and suggestions for their treatment. Introduction Materials and methods Results Discussion Conclusion References CHAPTER VIII. CHAPTER IX. General discussion General conclusion GENERAL REFERENCES CURRICULUM VITAE 2

9 INTRODUCTION

10

11 CHAPTER I. Introduction Failure and success of oral implants Endosseous oral implants have become a significant factor in oral rehabilitation since the early seventies. When a properly documented implant system with a documented long-term success rate has been used, an implant-supported prosthesis is supposed to give the patient a long lasting rehabilitation (Lindquist et al. 1996). On the other hand, despite the many technical advances, failures of implants remain a significant challenge for both clinician and patient. There are emotional and important financial implications. A failing implant can compromise the planned or achieved oral rehabilitation. Failures of the endosseous implants can be subdivided into early and late failures, depending on whether they occur before or at abutment connection (= early), in other words prior to or after exposure to the oral environment and loading (applying a prosthetic superstructure = late). This subdivision is relevant since the aetiology of failures is different. An early failure results from the lack of establishing an intimate bone -to- implant contact (Esposito et al. 1998). This means that bone healing is impaired or even jeopardized after implant insertion. Both systemic (e.g. Crohn s disease) and local factors (e.g. relative movement of the implant towards the environing bone) can interfere with these cellular events. Under such circumstances the mechanisms that normally lead to a proper wound healing by means of bone apposition onto the implant surface do not take place and a fibrous scar tissue is rather formed in between the implant surface and surrounding bone (Esposito et al. 1999). The latter can allow epithelial downgrowth which results in a so-called saucerization or marsupialisation of the implant. This leads to surinfection, implant mobility or even loss. Thus the anchoring function of the endosseous implant cannot be properly maintained. Late implant failures on the other hand are rather due to either peri-implantitis or overloading or both of them. These failures have both been characterized by peri- 1

12 implantitis either resulting from plaque-induced gingivitis or from occlusal overloading (van Steenberghe et al. 1990, Quirynen et al. 2002). The present thesis wants to identify via both large scale retrospective and prospective analyses of patient data, the influence of systemic, behavioural and (local bone and intra-oral) factors on the occurrence of implant failures (chapter II, III and IV). Besides the impact of some implant characteristics on failures will be investigated. I- Patient related factors: - Systemic and behavioral factors: Systemic factors, such as diseases, or behavioral factors may affect oral tissues, and thus interfere with wound healing. Medications used for the treatment of systemic conditions may also affect the outcome of the osseointegration process and its maintenance. It remains a matter of debate which systemic factors can compromise the achievement of an intimate bone to implant interface and/or its maintenance over time. It is especially during the healing time, up to abutment surgery and thus prior to occlusal and microbial challenges, that systemic factors can be more easily identified. Indeed, cofactors such as the above mentioned which occur after abutment surgery and especially after loading by means of a prosthetic superstructure are not yet present (van Steenberghe et al. 2003). The influence of general health problems on the osseointegration process is insufficiently documented (van Steenberghe et al. 2002). Most studies dealing with the role of systemic factors document the effect on long term maintenance of osseointegration. Less is known concerning factors affecting the initial bone apposition on the implant surface (Kronström et al. 2000, 2001). Many systemic conditions have been listed as potentially critical for implant integration, but no controlled studies are available (Mombelli & Gionca 2006). An overview of the literature dealing with systemic conditions and implant failures identified a series of pathologies which play a role: sclerodermia, Parkinson s disease, Sjögren s syndrome, HIV infection, pemphigus vulgaris, cardiovascular diseases, diabetes mellitus type I and II, osteoporosis, Crohn s disease, hematological diseases. Smoking has a very significant impact (for the latter see review; Bain 1996). 2

13 - Local factors Local factors, such as jaw bone (quality and quantity), type of edentulism, location of the edentulous region where the implant will be installed, may affect implant success rate. Several clinical reports on the use of oral implants mention that poor bone quality, as assessed on preoperative radiographs, leads to a less predictable outcome (Porter & von Fraunhofer 2005). While in well mineralized bone with proper degrees of corticalisation, like the symphyseal area a success rate of 99 % was reported even after 15 years with Brånemark system implants (Lindquist et al.1996), in dorsal areas of the upper jaw it can be substantially lower (Adell et al. 1990, Nevins & Fiorellini 1998). It thus seems relevant to develop measurements of the bone quality, especially referring to its mineral density, as a determinant for the primary stability of endosseous implants. It has been observed indeed that too large micromovements during the healing period can disrupt the bone apposition process on the implant surface and rather lead to fibrous scar tissue formation (Szmukler-Moncler et al. 1998). The assessment of the primary stability at insertion may be another option to measure the bone density. It may allow to determine the prognosis or to decide whether early or even immediate loading can be performed or should be postponed. This alternative allows the bone to implant interface to heal for a few months before being exposed to the oral environment. This was the original Brånemark protocol (Brånemark et al. 1969). Another, but unpractical, technique to determine the bone mineral content is to take biopsies of the jaws preoperatively. The most popular current method of bone quality assessment is that developed by Lekholm & Zarb (1985), who introduced a scale of 1 to 4, based on both the radiographic assessment, and the sensation of resistance experienced by the surgeon when preparing the fixture site. The grading refers to individual experience, and only provides a rough mean value for the entire jaw. Johansson & Strid (1994) described a technique whereby bone quality as a function of density and hardness could be derived from the torque forces needed during the screw-shaped implant insertion. They postulated that the energy used in tapping the site, prior to or during implant placement, is a combination of the thread placement force from the tip of the instrument and the friction created as the remaining part of a tap or 3

14 implant enters the site. It has been demonstrated in an ex vivo human preparations that the cutting resistance during implant installation correlates well with the bone density as assessed by microradiography (Friberg et al. 1995). The absence of fixture mobility, either indicative of a good primary stability at insertion or of an intimate bone-to-implant contact after some healing period, can be objectively determined by an electronic measuring device, the Periotest (Gulden, AG, Bensheim, Germany) (Olivé & Aparicio 1990, Teerlinck et al. 1991, van Steenberghe & Quirynen 1993, van Steenberghe et al. 1995). This apparatus is widely used to assess implant outcome as can be seen from the hundreds of papers referring to it ( periotest.de). The PTV can reveal the increased stiffness over time of the implant-bone continuum (Tricio et al. 1995). Implant stability can also be measured by resonance frequency analysis normally referred to as Implant Stability Quotient, ISQ) (Meredith 1998). The in vivo experimental findings demonstrate resonance frequency is related to implant stiffness in the surrounding tissues, which means a higher bone to implant contact percentage (Rasmusson et al. 1998). Clinically, the increased implant stability demonstrated by repeated ISQ measurements, was attributed to increased miniralization of the surrounding bone (Friberg et al. 1999). The Osstell device (Mentor, Integration Diagnostics AB, Sävedalen, Sweden) allows to register the minute changes in the rigidity of the bone-to-implant contact. The validity of the subjective jaw bone quality assessment, the so-called Lekholm and Zarb index, was evaluated by comparing it to the above mentioned objective parameters: the torque force needed to install implants, and the primary stability of these implants measured either by ISQ or PTV (see chapter V). II- Implant related factors: Implant parameters (length and diameter) and surface characteristics: The use of oral endosseous implants, to retain or to support a dental prosthesis, is a predictable clinical procedure if based on the osseointegration principle (Brånemark et al. 1969). High success rates were reported on consecutive implants both in full and partial edentulism and in the upper and lower jaw. Lindquist et al. (1996) even reported a cumulative success rate of 98.9% for the Brånemark system after 15 years for 4

15 edentulous patients provided with mandibular fixed prostheses on implants. The same patient group eventually included functional implants successful for more than 20 years, with a cumulative survival rate of 98.9% (Ekelund et al. 2003). Similar results were reported for implants retaining an overdenture, again in the symphyseal area (van Steenberghe et al. 2001). The rehabilitation of partially edentulous jaws seemed originally less successful in a multicenter retrospective study (van Steenberghe.1989). Later, it appeared that a learning curve played a certain role. Indead a 10-year prospective multicenter study cumulative success rate of 90.2% and 93.7% were reached for respectively the upper and lower jaws (Lekholm et al. 1999). All these successes were achieved with commercially pure screw-shaped implants with a machined surface (also called a turned surface). Recently, many implant surfaces have been developed using various roughening techniques such as blasting with aluminum oxide particles, grit-blasting with titanium dioxide particles, sandblasting and acidetching, and acid-etching alone (Cochran et al. 2002, Ibanez et al. 2002, Albrektsson et al. 2004). Increased oxidation of the implant surface has also been proposed (Hall & Lausmaa. 2000). The TiUnite implant (Brånemark system, Nobel Biocare, Gothenburg, Sweden) surface is created by anodic oxidation. These modified surfaces have proven to enhance and speed up the bone apposition at the implant interface (Carlsson et al. 1988, Buser et al. 1991, Albrektsson et al. 2000, Henry et al. 2000, Ivanoff et al. 2003). Early occlusal load may be more easily imposed on implants with such modified surfaces, since bone apposition takes place at a faster rate (Rocci et al. 2003, Calandriello et al. 2003, Olsson et al. 2003). Modified or so-called improved, surfaces could also lead to an increased success rate in patients or locations that do not offer optimal bone quality and quantity. For optimal situations, such as the mandibular symphyseal area, where success rates close to 100 % have already been achieved with machined surfaces, the need for improved surfaces can be questioned (Ekelund et al. 2003). 5

16 Another development over the years has been the use of less elaborate surgical approaches, since clinicians with variable surgical skills and training are more and more performing implant insertions into the jaw bone. The variation in clinical skills and experience may need to be compensated for by an improved surface to obtain similarly predictable osseointegration rates. Aseptic surgery has been advocated since the early days (Brånemark et al. 1985). The use of a nose guard to prevent contact of sterile gloves with the highly contaminated nasal skin is highly recommended (van Steenberghe et al. 1997). Today these aspects are often overlooked since surgery is performed even in a non-surgical set-up. Furthermore, even when a less precise drilling trajectory is used or when drilling is performed without sufficient cooling, predictable bone apposition can still be achieved with implants with such improved surfaces. Lazzara et al. (1999) for example, demonstrated histomorphometrically a significantly higher bone implant contact (BIC) rate after 6 months in the posterior maxilla of humans when a double acid-etched surface was used rather than a machined surface (73% versus 34%). A prospective study was set to evaluate the effect of systemic and local factors on the incidence of early failure of implants with TiUnite surface (see chapter III). A retrospective study was set to compare the success rates of implants with an increased oxide layer (TiUnite ) and machined surface implants used to replace failed implants. The replacements implants were thus exposed to the same patient-related risk factors as the failed implants (see chapter VI). A retrospective study evaluated the incidence of clinically relevant inflammatory lesions around both machined and TiUnite implants (see chapter VII). 6

17 Part 1. LOCAL AND SYSTEMIC FACTORS AND IMPLANT FAILURES

18

19 CHAPTER II. The impact of local and systemic factors on the incidence of oral implant failures up to abutment connection II.1. Introduction: When a properly documented implant system with a long-term success rate has been selected, an implant-supported prosthesis is supposed to give the patient a long lasting rehabilitation (Lindquist et al. 1996). A failure of an implant can, on the other hand, compromise the achieved oral rehabilitation. Failures of the endosseous implants can be subdivided into early and late failures, depending on whether they occur before or at abutment connection (= early) or rather after occlusal loading took place by means of a prosthetic superstructure (= late). This subdivision is relevant since the aetiology of failures varies according to the timing. An early failure of an implant results from an inability to establish an intimate bone -to- implant contact (Esposito et al. 1998, Quirynen et al. 2002). This means that bone healing after implant insertion is impaired or even jeopardized. Both systemic and local factors can interfere with these primarily cellular events. The mechanisms that normally lead to wound healing by means of bone apposition do not take place, and rather a fibrous scar tissue is formed in between the implant surface and surrounding bone (Esposito et al. 1999). This can lead to epithelial downgrowth, a so-called saucerization or marsipualisiation of the implant, which results in mobility or even implant loss. Thus the anchoring function of the endosseous implant cannot be properly performed. Late implant failures on the other hand are influenced by both the microbial environment and the prosthetic rehabilitation. These failures have been associated with both peri-implantitis resulting from plaque-induced gingivitis and peri-implantitis and/ or occlusal overloading (van Steenberghe et al. 1990, Quirynen et al. 2002). 9

20 Systemic diseases may affect oral tissues, by increasing their susceptibility to other diseases, or by interfering with wound healing. Medications used for the treatment of systemic conditions may also affect the outcome of implants. It remains a matter of debate which systemic factors compromise the achievement of an intimate bone-implant interface and/or its maintenance over time. It is especially during the healing time, up to abutment surgery, that systemic factors can be most easily identified as risk factors from many other cofactors, which occur after abutment surgery and especially after loading by prosthetic superstructures (van Steenberghe et al. 2003). The influence of general health problems on the osseointegration process is poorly documented (van Steenberghe et al. 2002). Although many studies noted the role of systemic and local factors in the long term maintenance of osseointegration, less is known concerning factors affecting the initial bone apposition up to abutment placement process (Kronström et al. 2000, 2001). The aim of this retrospective study is to assess the influence of systemic and local bone and intra-oral factors on the occurrence of early implant failures up to abutment connection. II.2. Materials and methods Materials The surgical records of 2004 consecutive patients (1212 females, 792 males) treated by means of endosseous implants during the period at the Department of Periodontology of the University Hospital of the Catholic University Leuven were evaluated. It is a general policy of the department to accept all patients who can benefit from implants for their oral rehabilitation even if systemic or local factors can eventually compromise the outcome to a certain extent. These patients received a total of 6946 implants (all Brånemark system, Nobel Biocare, Gothenburg, Sweden). These were inserted under strict aseptic conditions in the operating theatre of the department, according to the well-defined protocol described in the Surgical Manual for the Brånemark system. For a total of 700 patients, patient s files were evaluated to check the reliability of the surgical records. Since a perfect coincidence between both could be 10

21 ascertained, checking in the patient files was abandoned afterwards and only the surgical files were further used. Early failure i.e. prior to and up to abutment connection- was related to the presence of health or behavioral factors, implant length and diameter, bone quality and quantity, implant location, type of edentulism, prescription of antibiotics pre- or immediately after surgery, dehiscence or perforation of the jaw bone during surgery, Periotest (Gulden, A G, Bensheim, Germany) value (PTV) and placement torque measurement (OsseoCare, Nobel Biocare, Gothenburg, Sweden), at the crestal third, the middle third and the apical third at implant insertion. The PTV measures the stability of the implant-bone continuum by tapping with an elctro-magnetically driven rod on the implant. The outcome is expressed in arbitrary units, reaching from 8 to + 50 (Tricio et al. 1995). Implants should lead to values below +5; the more negative, the better the stability. Placement torque measurement was recorded during implant insertion, by means of an electronic torque force measurement device which is a part of a controlled motor device. The latter measures the torque force (Ncm) while tapping or inserting the implant at the crestal third, the middle third, and the apical third of each implant insertion trajectory. The surgical records which were not fully completed led to the patient s file to be examined. For a total of 232 patients because of purely administrative reasons, the files could not be retrieved. Data collection and analysis Only screw-shaped Brånemark system implants (Nobel Biocare, Gothenburg, Sweden) were used either with a machined (n = 6316) or Ti-Unite TM surface (n = 630). A minimal bone height of 7mm was required for implant placement. The general health and the behavioral history of the patient were recorded on the surgery form after thorough questioning of the patients preoperatively. Furthermore two supplementary forms, one dealing with all the information regarding the implants used and bone quality and quantity, and the other one dealing with all the information regarding abutment surgery were available. If the implant failed prior to or at abutment surgery, the failure was recorded. An implant was considered a failure if a peri-implant radiolucency could be detected on the intra-oral radiographs, if an individual implants showed the slightest sign 11

22 of mobility corresponding to a Periotest value (PTV) of > =5, or if the patient showed subjective signs of pain or infection, all these led to implant removal. Thus the expression failed implant in our study means a lost implant. Jaw bone quality and the degree of jaw bone resorption were evaluated by the periodontologist at implant placement. Tactile evaluation during drilling and assessment of the alveolar crest both radiographically and clinically allowed classification according to Lekholm & Zarb index (1985). The following health or behavioral factors were particularly (i.e. by questioning the patient and/or checking his medical records from other departments in the hospital) assessed: smoking habits, hypertension, ischemic cardiac problems, coagulation anomalies, gastric problems such as ulcers, osteoporosis, hypo- or hyperthyroidism, hypercholesterolemia, asthma, diabetes type I or II, Crohn s disease, rheumatoid arthritis, chemotherapy, intake of medication (antidepressants, steroids). If one did smoke, the patient was allocated to one of the following three categories (<10 cigarettes/ day, cigarettes/day or > 20 cigarettes/day). Local bone factors, such as radiotherapy of the area concerned, were also recorded. Finally, a special note was made for patients with claustrophobia. These patients were treated with reduced coverage of the face, often without nose cape and as such with a breach of asepsis (van Steenberghe et al. 1997). As the complication often occurred during surgery, removal of some drapes often led to unavoidable microbial contamination. The type of edentulism was classified according the presence and location of the remaining natural dentition in the oral cavity related to implant location: full edentulism, teeth present only in the antagonistic jaw, teeth present in the same jaw where the implant is but not neighboring it and teeth neighboring the implant. In the department a thorough sterility policy allows limiting the systemic use of antibiotics to well-defined indications such as endocarditis prophylaxis, a remaining infection at the site of surgery, coughing or sneezing by the patient during surgery. Use of antibiotics pre-or immediately after implant surgery- was defined as yes or no. Since the type of edentulism was not mentioned on the surgical forms, the analysis was limited to 676 patients (2448 implants) for whom the entire patient file was scrutinized. PTV and torque force measurements were only performed on a fraction of the patient 12

23 material because of irregular availability of the machinery. Systemic and behavioral factors were available for the entire population. Statistical methods Logistic regression models were used to evaluate the effect of explanatory variables on the early failure of the implant. Generalized estimating equation (GEE) method (Liang and Zeger, 1986; Zeger and Liang, 1986) was used to account for the fact that repeated observations (several implants) were available for a single patient. Firstly, a univariate effect of each implant related, behavioral and local bone factor on the early failure was evaluated by fitting univariate GEE logistic regression model. Odds ratios and their 95% confidence intervals based on the robust standard errors from the GEE logistic regression model were computed. The Wald test based on robust standard errors was used to assess the significance of each factor. For categorical factors with more than 2 levels, robust Wald s P-values adjusted for the multiple comparisons, using the method of Holm (1979) were computed. Secondly, we evaluated a multivariable effect of the health factors when controlled for the behavioral, implant related and local bone factors that were univariately (at 5%) significant. Namely, the following factors were controlled for: smoking habits, bone quality and quantity, implant s site (posterior/anterior), length and diameter. Although univariately significant, type of edentulism, and PTV at implant insertion was not controlled for since the data were available for only a limited subgroup of patients. For the purpose of multivariable analysis, the implants with missing data on bone quality and quantity were removed (more than implants). Consequently, it was not possible to evaluate statistically the effect of chemotherapy on early implant failures, due to the fact that no early failure have been observed in the chemotherapy subgroup of the rest group of implants. Due to the fact that no failure have been observed in the group of patients having a given disease, the effect of diabetes type I and rheumatoid arthritis could not be assessed statistically. Statistical analyses were performed using the R software (R Development Core Team, 2006) and the R package gee (Carey, 2002). 13

24 II.3. Results: From the treated patient s population, a total of 252 implants - of different lengths and diameters - out of the 6946 implants installed, appear to have failed one to six months after placement. This corresponds to an early failure rate of 3.6 %. These failed implants occurred in 178 patients. Early implant failures related significantly to the following implant characteristics: implant diameter, length and location. Early failure rates related significantly to smoking habits and increased with the increasing number of cigarettes smoked per day. Early implant failure related significantly to the type of edentulism. No significant correlation was found between early failure and the torque measurements at placement, either in the crestal, middle or apical third. The Periotest values, when recorded at implant placement, were related to early implant failure. Significantly more early failures occurred with increasing values of PTV which indicate a lowered rigidity. There was no significant effect of the presence of dehiscence or fenestration of the bone tissue at implant site during implant insertion on the early failures. Bone volume (bone quantity) and bone quality as assessed by the use of Lekholm & Zarb index (1985) affected significantly early implant failure. A summary of the Univariate GEE logistic regression for all the above mentioned factors can be found in Table 1:a,b. When a multiple comparison was done; significantly more failures were detected in implants with wide platform (5 mm) when compared to implant with regular platform (3.75 and 4 mm), [p-value = 0.004, 0.02, Odds ratio (95% C. I.)= 2.70 ( ), 2.73 ( ) respectively]. Significantly more early failures were detected with short implants (< 10 mm) when compared to implants with a length ranging from 10 to 15 mm [p-value= 0.04, Odds ratio (95% C. I.) =1.710 ( ). Significantly more early failures were detected in the mandibular and maxillary posterior regions when compared to the mandibular anterior region [p-value = 0.01, 0.03, Odds ratio (95% C. I) = 1.99 ( ), 1.88 ( ) respectively]. 14

25 A significant difference was detected between heavy smoking (>20 cigarettes/day) and no smoking groups [p-value < 0.001, Odds ratio (95% C. I): 2.72 ( )]. Significantly higher failure rate was noticed when the implants neighboring a tooth were compared to implants in full edentulism, or to presence of teeth in the antagonistic jaw only [p-values: 0.01, 0.03, Odds ratio (95% C. I): 2.77 ( ), ( ) respectively]. Significantly more failures in resorption grade E (=extreme) were detected when compared to grade A, B, or C. [p-values: 0.03, < 0.001, 0.009, Odds ratio (95% C. I): 3.43 ( ), 5.21( ), 3.90 ( ) respectively]. Significantly more failures in grade 4 (soft bone with little cortical bone) were detected when compared to grade 2 [p-value<0.001, Odds ratio (95% C. I): 3.05 ( ), and more failures in grade 1 compared to grade 2 [p-value=0.02, Odds ratio (95% C. I): 0.42 ( ). 15

26 Table 1a: Univariate GEE logistic regression: implant related, behavioral and local bone factors, the total number of patients to whom the factors were evaluated and the distribution of the failed and successful implants. * Significant p-value < Univariate GEE Logistic Regression: Summary Odds 95% Factor (Patient/Implant) Successful Failed Ratio Conf. Interval Diameter (2004/6936) p-value: * (96.47%) 209 (3.53%) (94.52%) 4 (5.48%) 1.08 (0.26, 4.43) (96.64%) 27 (3.36%) 0.99 (0.63, 1.55) (91.55%) 12 (8.45%) 2.70 (1.53, 4.79) Length (2004/6946) p-value: * (96.57%) 216 (3.43%) 1 < (93.64%) 29 (6.36%) 1.71 (1.11, 2.64) > (96.28%) 7 (3.72%) 1.21 (0.51, 2.89) Location (2000/6931) p-value: * Mandible, anter (97.66%) 46 (2.34%) 1 Mandible, poster (95.30%) 63 (4.70%) 1.99 (1.29, 3.07) Maxilla, anter (96.59%) 69 (3.41%) 1.48 (0.95, 2.18) Maxilla, poster (95.38%) 74 (4.62%) 1.88 (1.20, 2.94) Smoking (2004/6946) p-value: <0.001 * (96.72%) 198 (3.28%) 1 < (95.15%) 11 (4.85%) 1.42 (0.48, 4.23) (94.69%) 17 (5.31%) 1.87 (1.07, 3.26) > (92.95%) 26 (7.05%) 2.72 (1.63, 4.54) Type edentulism (676/2448) p-value: <0.001 * No teeth 697 (97.08%) 21 (2.92%) 1 In the ant. jaw only 360 (97.83%) 8 (2.17%) 0.57 (0.17, 1.93) In the same jaw 520 (94.55%) 30 (5.45%) 1.97 (0.94, 4.11) Neighb. implant 749 (92.24%) 63 (7.76%) 2.77 (1.45, 5.28) 16

27 Table 1 b: Univariate GEE logistic regression: implant related, behavioral and local bone factors, the total number of patients to whom the factors were evaluated and the distribution of the failed and successful implants. * Significant p-value < Univariate GEE Logistic Regression: Summary Odds 95% Factor (Patient/Implant) Successful Failed Ratio Conf. Interval Crestal third Ncm (138/320) p-value: increase by (95.94%) 13 (4.06%) 1.04 (0.70, 1.55) Middle third Ncm (138/320) p-value: increase by (95.94%) 13 (4.06%) 1.07 (0.96, 1.20) Apical third Ncm (138/320) p-value: increase by (95.94%) 13 (4.06%) 1.07 (0.97, 1.18) PTV (71/189) p-value: 0.05 * increase by (88.89%) 21 (11.11%) 1.13 (1.000, 1.28) Dehiscence (430/1380) p-value: No 1238 (96.79%) 41 (3.21%) 1 Yes 96 (95.05%) 5 (4.95%) (0.571, 4.632) Fenestration (418/1345) p-value: No 1267 (96.79%) 42 (3.21%) 1 Yes 35 (97.22%) 1 (2.78%) (0.14, 6.86) Bone quantity (1759/5800) p-value: * A 998 (96.33%) 38 (3.67%) 1 B 2362 (97.36%) 64 (2.64%) 0.66 (0.41, 1.05) C 1700 (96.48%) 62 (3.52%) 0.88 (0.53, 1.46) D 375 (95.42%) 18 (4.58%) 1.07 (0.51, 2.25) E 159 (86.89%) 24 (13.11%) 3.43 (1.49, 7.89) Bone quality (1759/5782) p-value: <0.001 * (94.86%) 26 (5.14%) (97.87%) 52 (2.13%) 0.42 (0.23, 0.77) (96.38%) 78 (3.62%) 0.70 (0.40, 1.22) (92.67%) 50 (7.33%) 1.28 (0.67, 2.45) Surface (2004/6946) p-value: Machined 6088 (96.39%) 228 (3.61%) 1 Ti-Unite 606 (96.19%) 24 (3.81%) 1 (0.531, 1.88) Systemic diseases and medical therapies were analyzed when controlled for the other diseases and for factors significantly (at the 5% level) related to the early failure; (smoking habits, bone quality and quantity, site, length and diameter). Certain factors, such as cardiac and gastric diseases, controlled diabetes type II, coagulation problems, hypertension, hypo- or hyperthyroidsm, hypercholesterolemia, asthma, radiotherapy of the area concerned, claustrophobia, antidepressant and steroid medication did not lead to an increased incidence of the early failures. 17

28 Crohn s disease and osteoporosis, in contrast, were significantly related to implant failures. Again a significant correlation became evident between early failures and, implant diameter, implant location (anterior/posterior) and smoking habits (Table 2). When a multiple comparison was done; significantly more failures were observed with wide platform implants when compared to regular implant diameters (4 mm) [pvalue=0.02, Odds ratio (95% C.I.): 3.02 ( )], and significantly more failures were detected in the posterior regions when compared to the anterior regions of both jaws [p-value<0.001, Odds ratio (95% C.I): 1.18 ( ). Table 2: Multivariable GEE logistic regression: implant related, behavioral and local bone factors and health factors. * Significant p-value < 0.05 Multivariable GEE Logistic Regression Number of patients = 1757 Number of implants = 5759 Factor Odds Ratio 95% CI p-value Hypertension (yes) 0.97 (0.56, 1.67) 0.91 Cardiac problem (yes) 0.42 (0.15, 1.22) 0.11 Gastric problem (yes) 1.81 (0.55, 5.97) 0.33 Osteoporosis (yes) 2.88 (1.51, 5.48) * Hypothyroid (yes) 1.00 (0.32, 3.16) Hyperthyroid (yes) 1.40 (0.07, 26.51) 0.82 Radiotherapy (yes) 0.36 (0.028, 4.65) 0.43 Crohn s disease (yes) 7.95 (3.47, 18.24) <0.001 * Diabetes II (yes) 0.25 (0.05, 1.20) 0.08 Coagulation (yes) 2.00 (0.93, 4.28) 0.08 Claustrophobia (yes) 2.45 (0.64, 9.39) 0.19 Antidepressant medication (yes) 1.28 (0.64, 2.58) 0.49 Steroid medication (yes) 1.25 (0.32, 4.98) 0.75 Hypercholesterolemia (yes) 1.02 (0.31, 3.35) 0.98 Asthma (yes) 1.92 (0.37, 9.97) 0.44 Smoking (<10) 1.76 (0.60, 5.16) 0.02 * Smoking (10 20) 1.90 (1.007, 3.60) Smoking (>20) 2.18 (1.20, 3.97) Bone quality (2) 0.56 (0.29, 1.05) 0.15 Bone quality (3) 0.82 (0.46, 1.47) Bone quality (4) 1.04 (0.53, 2.07) Bone quantity (B) 0.64 (0.40, 1.02) 0.10 Bone quantity (C) 0.86 (0.51, 1.46) Bone quantity (D) 0.95 (0.4437, 2.06) Bone quantity (E) 2.00 (0.77, 5.21) Site (posterior) 1.81 (1.30, 2.53) <0.001 * Length (<10 mm) 1.04 (0.61, 1.77) 0.99 Length (>15 mm) 1.01 (0.41, 2.48) Diameter (3.3 mm) 1.18 (0.35, 4.04) 0.04 * Diameter (4 mm) 0.75 (0.45, 1.24) Diameter (5 mm) 2.26 (1.20, 4.27) 18

29 II.4. Discussion: Early implant failures occur because, instead of an intimate bone-to-implant contact, a fibrous scar tissue is formed between the bone and the implant surface. A large variety of causes can be imagined which interfere with the normal bone wound healing. The tissue reactions following the insertion of an implant in the bone can be compared with fracture healing. The healing of the tissue starts with a blood clot that forms between the remaining bone and the implant surface. Depending on the environment and the relative immobility of the bone-to-implant interface, pluripotent mesenchymal cells will differentiate either into fibroblasts or osteoblasts, leading to the formation of, respectively, a scar tissue or bone (Sennerby 1991). Conditions of poor vascularity or low oxygen tension may direct the mesenchymal cells to a chondrogenic differentiation. The mechanical stress to which the tissues are subjected may also influence cellular differentiation. Distortional stresses may deform cells, altering their genetic expression and synthetic activity, which explains why micromovements of the implants during the healing phase can affect a correct bone-to-implant bond, rather forming fibrous scar tissue (Ivanoff et al. 1996, Szmukler-Moncler et al. 2000). The role of endogenous factors on cellular turn over and differentiation is less documented. Renouard and Nisand (2006) reported in a review paper, that there is a trend for machined-surface implants for an increased failure rate with short and wide-diameter implants. The highest failure rate for short implants was reported in studies which were performed with routine surgical procedures independently of the bone density, with machined-surface implants and in restricted anatomic sites with poor bone density. The increased failure rate of wide-diameter implants reported in some studies was mainly associated with a learning curve, poor bone density, implant design and site preparation, and the fact that this implant was usually used as rescue implants. It is known that an improved surface like Ti-Unite leads to less early failures. From our present data the limited series of the latter surface did not allow to confirm this. Since however; the experience in our centre confirms this tendency for a reduced failure rate with Ti-Unite surfaced implants (Alsaadi et al. 2006). 19

30 In the present study, significant effect of implant length and diameter were detected, more failures occurring with short and wide-diameter implants. These implants were systematically installed in compromised sites, marked by poor bone quality and quantity. Thus these confounding factors may explain the higher failure rate. Too high and low bone densities, as assessed clinically or radiologically, have also been pointed as two possible reasons for non-integration (Engquist et al. 1988, Friberg et al. 1991, Jaffin & Berman 1991). In our present findings it also appears that bone quality type 1 and 4 according to the Lekholm and Zarb classification are associated with slightly higher failure rates. The effects of the inhaled tobacco smoke can be divided into 2 phases: a volatile and a particulate phase. The volatile phase, accounting for 95% of the cigarette smoke, provides nearly 500 different gases, including nitrogen, carbon monoxide, and carbon dioxide. The roughly 3,500 different chemicals released in the particulate phase include nicotine, nornicotine, anatabine, and anabasine (Hoffmann & Hoffmann 1997). Stripped of water, the particulate matter that remains, or tar, contains the majority of the carcinogens of cigarette smoke. Nicotine has been shown to increase platelet aggregation, to decrease microvascular prostacyclin levels, and to inhibit the function of fibroblasts, erythrocytes, and macrophages (Zevin et al. 1998, Jorgensen et al. 1998). Carbon monoxide binds to hemoglobin many times more easily than oxygen, thus displacing oxygen from the molecule and lowering oxygen tension in tissues (Leow & Maibach 1998). Smoking has been determined to adversely affect bone mineral density, lumbar disk health, the relative risk of sustaining wrist and hip fractures, low back pain, and the dynamics of bone and wound healing (Porter & Hanley 2001). Several studies reported on the negative effect of smoking on osseointegration, and that it is a dose-related effect (Bain 1996), which is in accordance with the present findings. Some studies have shown that systemic antibiotic used prior to implant surgery can reduce the occurrence of infections after surgery and increase the success rates of integration (Dent et al. 1997, Laskin et al. 2000). Another study found no such effect (Gynther et al. 1998), which is in accordance with the present study. The trouble with this kind of studies is that the asepsis cannot be taken for granted. Many people perform 20

31 surgery in this field without a proper surgical background. Often sterility measures do not even involve covering of the nose, the most infected site in this area (van Steenberghe et al. 1997). In the present study a series of 120 patient files (516 implants) were analyzed for the use of antibiotics in the perioperative period. There was no statistical effect on the failure rate (p-value = 0.8). The prescription of antibiotics when sterility is truly respected proves to be unnecessary and even, considering the possible side-effects, it should be discouraged. There should be more emphasis, as in all other medical disciplines on a proper education of those who want to perform surgery so that they understand the importance of sterility once there is a breach of the mucoperiosteal lining. Indeed systematic prescription of antibiotics leads to unnecessary side-effects and costs (Lawler et al. 2005) Crohn s disease can, as it is a generalized autoimmune disease, affect the entire gastrointestinal system, and thus even lead to periodontal lesions (van Steenberghe et al. 1976). Crohn s disease is characterized by the presence of many antibody-antigens complexes, leading to autoimmune inflammatory process in several parts of the body, such as enteritis, vasculitis, recurrent oral ulceration, arthritis or keratoconjuctivitis. The same could occur at the interface with biocompatible implants, normally considered by the host as a part of the body, but which could in Crohn s patients recognized as non-self, thus affecting the outcome of implant osseointegration (van Steenberghe et al. 2002). Moreover, the malnutrition encountered in Crohn s patients could also cause a deficient bone healing around the implant (Esposito et al. 1998). Osteoporosis has been defined as a decrease in bone mass and bone density and an increased risk and/or incidence of bone fracture. However, it has been noted that patients without fractures may have lost a significant amount of bone, while many patients with fractures display levels of bone mass similar to those of control subjects (Cummings 1985, Melton & Wahner 1989, Jacobs et al. 1996) In addition the relationship between skeletal and mandibular or maxillary bone mass is limited (von Wowern & Melsen 1979, von Wowern et al. 1988). The Word Health Organization has established diagnostic criteria for osteoporosis based on bone density measurements determined by dual energy x-ray absorptiometry: A diagnosis of 21

32 osteoporosis is attributed if the bone mineral density level is 2.5 standard deviations below that of a mean young population (Glaser & Kaplan 1997). There are two types of osteoporosis; Type I or high turnover- which mostly occurs in women aged due to the sudden decrease in estrogen as a result of (early) menopause. This causes rapid calcium loss from the bones, making the woman susceptible to hip, wrist, forearm, and spinal compression fractures. Type II or low turnover, age-related, or senile osteoporosis-, it occurs when bone loss and formation are not equal and more bone is broken down than replaced. It affects both men and women. It is associated with leg and spinal fractures in both genders. The disease may have an influence on periodontal attachment loss (Wactawski-Wende et al. 1996). Although no studies prove an association between implant failure and the state of osteoporosis, it has been suggested as a risk factor for implant failure especially for postmenopausal women (Becker et al. 2000). In the present study a significant association was detected between early implant failures and osteoporosis. When using implants in treating partial edentulism, one of the most important questions concerning the clinician is whether the influence of the periodontal and endodontic status of the neighboring jaw bone is of importance. A multicenter study reported that a significant plaque accumulation and gingival inflammation at the time of implant placement seems to increase the risk of failure (van Steenberghe et al. (1990). In a retrospective study van Steenberghe et al. (1999)indicated that some of the early failures may be linked with an endodontic pathology, either remaining after tooth extraction or around neigbouring teeth. The higher incidence of these pathologies for failed implants and/or implants with retrograde peri-implantitis vs. successful implants is obvious (with 3 х or even higher incidence). Quirynen et al. (2005) reported that endodontic pathology of the extracted tooth (scar tissue-impacted tooth) at the site of implant placement or possible endodontic pathology in the vicinity, leads to implant failures. These finding are in accordance with other studies (for review see Quirynen et al. 2003, Quirynen & Teughels 2003). In the present study a significant association was detected between early implant failures and the vicinity with a remaining natural dentition. 22

33 The use of the Periotest at insertion seems relevant since there is a clear-cut difference between high and low primary stability at insertion on the chances to achieve a proper osseointegration. This kind of biomechanical assessment could thus be recommended. II.5. Conclusions: This vast number of consecutive patients allows identifying -because of the homogeneity of the treatment hardware and software- a number of systemic and local factors which interfere with the osseointegration process. Since the study limits the observation to the stage prior to the prosthetic treatment, confounding factors such as loading of the implants or the microbial challenge are eliminated. Some identified factors for failure could be expected, such as smoking while others like Crohn s disease, osteoporosis, and vicinity with the natural dentition are less known. The indication for the use of oral implants should sometimes be reconsidered when alternative prosthetic treatments are available and when possibly interfering systemic or local factors are identified. 23

34 II.6. References: Alsaadi, G., Quirynen, M., & van Steenberghe, D. (2006) The importance of implant surface characteristics in the replacement of failed implants. International Journal of Oral & Maxillofacial Implants 2, Bain, C. A. (1996) Smoking and implant failure--benefits of a smoking cessation protocol. International Journal of Oral and Maxillofacial Implants 11, Becker W., Hujoel P.P., Becker B.E. &Willingham H. (2000) Osteoporosis and implant failure: an exploratory case-control study. Journal of Periodontology 4, Carey, V. (2002) gee: Generalized Estimation Equation solver. R package version , ported to R by Thomas Lumley (version 3.13 & 4.4) and Brian Ripley (version 4.13). Cummings, S.R. (1985) Are patients with hip fractures more osteoporotic? Review of the evidence. The American Journal of Medicine 78, Dent, C.D., Olson, J.W., Farish, S.E., Bellome, J., Casino, A.J., Morris H.F. & Ochi S. (1997) The influence of preoperative antibiotics on success of endosseous implants up to and including stage II surgery: a study of 2,641 implants. Journal of Oral and Maxillofacial Surgery 55, Engquist, B., Bergendal, T., Kallus, T. & Linden, U. (1988) A retrospective multicenter evaluation of osseointegrated implants supporting overdentures. International Journal of Oral and Maxillofacial Implants 3, Esposito, M., Hirsch, J.M., Lekholm, U. & Thomsen, P. (1998) Biological factors contributing to failures of osseointegrated oral implants. (II). Etiopathogenesis. Eurupean Journal of Oral Sciences 3, Esposito, M., Thomsen, P., Ericson, L.E. & Lekholm U. (1999) Histopathologic observations on early oral implant failures. International Journal of Oral and Maxillofacial Implants 6, Friberg, B., Jemt, T. & Lekholm, U. (1991) Early failures in consecutively placed Brånemark dental implants, a study from stage 1 surgery to the connection of completed prostheses. International Journal of Oral and Maxillofacial Implants. 6, Glaser, D.L. & Kaplan, F.S. (1997) Osteoporosis. Definition and clinical presentation. Spine 22, Gynther, G.W., Kondell, P.A., Moberg, L.E. & Heimdahl, A. (1998) Dental implant installation without antibiotic prophylaxis. Oral Surgery, Oral Medicine, Oral Pathology, Oral Pathology, Oral Radiology and endodontics 85, Hoffmann, D. & Hoffmann, I. (1997) The changing cigarette, Journal of Toxicology and Enviromental Health 50, Holm, S. (1979). A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics 6, Ivanoff, C.J., Sennerby, L. & Lekholm, U. (1996) Influence of initial implant mobility on the integration of titanium implants. An experimental study in rabbits. Clinical Oral Implants Research 7,

35 Jacobs, R., Ghyselen, J. & Koninckx, P., van Steenberghe D. (1996) Long-term bone mass evaluation of mandible and lumbar spine in a group of women receiving hormone replacement therapy. European Journal of Oral Sciences 104, Jaffin, R.A. & Berman, C.L. (1991) The excessive loss of Brånemark fixtures in type IV bone, a 5-year analysis. Journal of Periodontology. 62, 2-4. Jorgensen, L.N., Kallehave, F., Christensen, E., Siana, J.E. & Gottrup, F. (1998) Less collagen production in smokers. Surgery 123, Kronström, M., Svensson, B., Erickson, E., Houston, L., Braham, P. & Persson, G.R. (2000) Humoral immunity host factors in subjects with failing or successful titanium dental implants. Journal of Clinical Periodontology 27, Kronström, M, Svenson, B, Hellman & M, Persson, GR. (2001) Early implant failures in patients treated with Brånemark System titanium dental implants: a retrospective study. International Journal of Oral and Maxillofacial Implants 16, Laskin, D.M., Dent, C.D., Morris, H.F., Ochi, S. & Olson, J.W. (2000) The influence of preoperative antibiotics on success of endosseous implants at 36 months. Annals of Periodontology 5, Lawler, B., Sambrook, P.J. & Goss, A.N. (2005) Antibiotic prophylaxis for dentoalveolar surgery, is it indicated? Australian Dental Journal 50, Leow, Y.H. & Maibach, H.I. (1998) Cigarette smoking, cutaneous vasculature, and tissue oxygen. Clinics in Dermatology 16, Lekholm, U. & Zarb, G. A. (1985) Patient selection and preparation. In: Brånemark, P-I., Zarb, G., Albrektsson, T. eds. Tissue integrated prosthesis: Osseointegration in clinical dentistry. p.199, Chicago, Quintessence publishing Co Inc. Liang, K.Y. & Zeger, S.L. (1986) Longitudinal data analysis using generalized linear models. Biometrika 73, Lindquist, L.W., Carlsson, G.E. & Jemt, T. (1996) A prospective 15-year follow-up study of mandibular fixed prostheses supported by osseointegrated implants. Clinical results and marginal bone loss. Clinical Oral Implant Research 7, Melton, L.J. 3rd & Wahner, H.W. (1989) Defining osteoporosis. Calcified Tissue International 45, Porter, S.E. & Hanley, E.N. Jr. (2001) The musculoskeletal effects of smoking. The Journal of the American Academy of Orthopaedic Surgeons 9, Quirynen, M., De Soete, M. & van Steenberghe, D. (2002) Infectious risks for oral implants: a review of the literature. Clinical Oral Implants Research 13, Quirynen, M., Gijbels, F. & Jacobs, R. (2003) An infected jawbone site compromising successful osseointegration. Periodontology , Quirynen, M. & Teughels, W. (2003) Microbiologically compromised patients and impact on oral implants. Periodontolgy , Quirynen, M., Vogels, R. Alsaadi, G., Naert, I., Jacobs, R. & van Steenberghe, D. (2005) Predisposing conditions for retrograde peri-implantitis, and treatment suggestions. Clinical Oral Implants Research 16, Renouard, F. &, Nisand, D. (2006) Impact of implant length and diameter on survival rates Suppl Clinical Oral Implants Research 2:

36 26 R Development Core Team (2005). R: A language and environment for statistical computing, ISBN , R Foundation for Statistical Computing, Vienna, Austria Sennerby, L. (1991) On the Bone Tissue Response to Tintanium Implants. Thesis. Gothenburg, Sweden, Gothenburg University. Szmukler-Moncler, S., Piattelli, A., Favero, G.A. & Dubruille, J.H. (2000) Considerations preliminary to the application of early and immediate loading protocols in dental implantology. Clinical Oral Implants Research 11, Tricio, J., Laohapand, P., van Steenberghe, D., Quirynen, M.& Naert, I. (1995) Mechanical state assessment of the implant-bone continuum, a better understanding of the Periotest method. International Journal of Oral and Maxillofacial Implants. 101, van Steenberghe, D., Vanherle, G.V., Fossion, E. & Roelens, J. (1976) Crohn's disease of the mouth, report of case. Journal of Oral Surgery 34, van Steenberghe, D, Lekholm, U, Bolender, C, Folmer, T, Henry, P, Herrmann, I., Higuchi, K., Laney, W., Linden, U. & Astrand, P. (1990) Applicability of osseointegrated oral implants in the rehabilitation of partial edentulism: a prospective multicenter study on 558 fixtures. International Journal of Oral and Maxillofacial Implants 53, van Steenberghe, D., Yoshida, K., Papaioannou, W., Bollen, C., Reybroeck, G.I. & Quirynen, M. (1997) Complete nose coverage to prevent airborne contamination via nostrils is unnecessary. Clinical Oral Implants Research 8, van Steenberghe, D., Quirynen, M., & Naert, I. (1999) Survival and success rates with oral endosseous implants. In: Lang, N.P., Attstrom, R. & Lindhe, J., eds. Proceedings of the 3de European Workshop on Periodontology, Berlin: Quintessence Publ. Co. Inc. van Steenberghe, D., Jacobs, R., Desnyder, M., Maffei, G. & Quirynen, M. (2002) The relative impact of local and endogenous patient-related factors on implant failure up to the abutment stage. Clinical Oral Implants Research 13, van Steenberghe, D., Quirynen, M., Molly, L. & Jacobs, R. (2003) Impact of systemic diseases and medication on osseointegration. Periodontolology , von Wowern, N. & Melsen, F. (1979) Comparative bone morphometric analysis of mandibles and iliac crests. Scandinavian Journal of Dental Research 87, von Wowern, N., Storm, T.L. & Olgaard, K. (1988) Bone mineral content by photon absorptiometry of the mandible compared with that of the forearm and the lumbar spine. Calcified Tissue International 42, Wactawski-Wende, J., Grossi, S.G., Trevisan, M., Genco, R.J., Tezal, M., Dunford, R.G., Ho, A.W., Hausmann, E. & Hreshchyshyn, M.M. (1996) The role of osteopenia in oral bone loss and periodontal disease. Journal of Periodontology 67, Zeger, S.L. & Liang, K.Y. (1986) Longitudinal data analysis for discrete and continuous outcomes. Biometrics 42, Zevin, S., Gourlay, S.G. & Benowitz, N.L. (1998) Clinical pharmacology of nicotine. Clinics in Dermatology 16,

37 CHAPTER III. The impact of local and systemic factors on the incidence of failures up to abutment connection with modified surface oral implants III.1. Introduction: Failures of endosseous osseointegrating implants can be subdivided into early or late failures, depending on whether they occur up to abutment connection (= early) or rather after the abutments exposure to the oral microbial environment and occlusal loading took place (= late). This subdivision is necessary because the etiology of failures during the two periods may be different. Early implant failure results from an inability to establish an intimate bone -to- implant contact (Esposito et al. 1998). This means that bone healing after implant insertion is impaired or jeopardized. The mechanisms that normally lead to wound healing by means of bone apposition do not take place, and a fibrous scar tissue is rather formed in between the implant surface and surrounding bone (Esposito et al. 1999). The latter leads to epithelial downgrowth, saucerization of the implant, and results in mobility or even implant loss. This will compromise the anchoring function of the endosseous implant. The late implant failures have been associated with both periimplantitis resulting from plaque-induced gingivitis and/ or occlusal overloading (van Steenberghe et al. 1990). It remains a matter of debate which systemic factors compromise the achievement of an intimate bone-implant interface and/or rather its maintenance over time. It is especially during the healing time, up to abutment surgery, that systemic factors can be most easily discriminated from other risk factors, which occur after abutment surgery do not yet apply (van Steenberghe et al. 2003, Mombelli & Gionca 2006). The influence of general health problems on the osseointegration process is still poorly documented (van Steenberghe et al. 2002). Advanced jawbone resorption and poor bone quality have been linked to higher rates of implant failure (Bass & Triplett, 1991; Jaffin & Berman, 1991). The improvement of the surface characteristics of the implant to enhance the bone response, 27

38 was a way to improve the clinical success rate. Numerous experimental studies focusing on roughened implant surfaces (i.e., plasma spraying, grit blasting, acid etching--) have found a faster and improved bone response, in terms of implant-bone contact and removal torque, as compared to turned surface (Carlsson et al. 1988, Buser et al. 1991, Huré et al. 1996). An implant with an increased oxide layer (TiUnite, Nobel Biocare, Göteborg, Sweden) become available some 5 years ago. The surface is created by anodic oxidation (Hall & Lausmaa 2000). Animal studies have shown a stronger bone reaction compared to turned implants, as measured with removal torque tests and histomophometry (Albrektsson et al. 2000, Henry et al. 2000). A histological study in human jawbone demonstrated higher bone response to anodic oxidized titanium implants than for implants with a turned surface (Ivanoff et al. 2003). Although the clinical evaluation of oxidized surface titanium implants (TiUnite ) is limited, the available studies demonstrated that the TiUnite surface implants have a better primary stability and help to achieve secondary stability earlier when compared to the machined surface implants (Rocci et al. 2003). The study is aimed to be exploratory to identify systemic, local bone and other intra-oral factors on the incidence of early implant failure. Although a statistical analysis has been performed, the main purpose of the study limits itself to suggest which factors may influence an early implant failure. To evaluate rigorously their effect, further studies explicitly designed for this purpose are needed. III.2. Material and methods: In agreement with the exploratory nature of the prospective study we enrolled, in a crosssectional way, all patients treated by means of endosseous implants during November June 2005 at the Department of Periodontology of the University Hospital of the Catholic University of Leuven were evaluated. It is a general policy of the department to accept all patients who can benefit from implants for their oral rehabilitation even if systemic or local factors can compromise the outcome. The patient group consisted of 28

39 283 consecutive patients (187 females, 96 males, mean age 56.2, age range 18-86). These patients received a total of 720 MkIII TiUnite implants (Brånemark system, Nobel Biocare, Gothenburg, Sweden). The classical two staged surgical protocol with strict sterility measures was used for all surgeries. At implant insertion, a minimal bone height of 7 mm had to be available. The study fulfills a high degree of homogeneity for both the implant type and the surgical phase. Early failures i.e. prior to and up to abutment connection- were related to the presence of health or behavioral factors, implant length and diameter, bone quality and quantity, implant location, type of edentulism, prescription of antibiotics pre- or immediately after surgery, dehiscence or perforation of the jaw bone during surgery, fenestration of the implant in the sinus or the nasal cavity, immediate insertion of the implant after tooth extraction, apical lesion detection radiographically, torque measurements at the crestal, middle and the apical third during implant insertion. The general health and the behavioral history of the patient were carefully recorded, through questioning the patient pre-operatively according to a printed questionnaire. Moreover, the patient medical status was also evaluated through hospital files, which are available on the hospital intranet system. If not, the house doctor was questioned when doubt arose. The following aspects were particularly assessed: smoking habits, hypertension, cardiac problems, gastric problems, osteoporosis, hypo- or hyperthyroid, hypercholesterolemia, asthma, diabetes type I or II, Crohn s disease, rheumatoid arthritis, chemotherapy, hysterectomy and intake of medication (antidepressants, steroids, hormone replacement). In case of HRT (hormone replacement therapy), the early failure rate was compared between all female patients of 50 years following HRT, and those do not follow HRT. Local bone factors, such as radiotherapy of the concerned area, were also recorded. Finally, a special note was made for patients with claustrophobia. These patients were treated with reduced coverage of the face, often without nose cover and as such with a breach of asepsis (van Steenberghe et al. 1997). As the complication often occurred 29

40 during surgery, removal of some drapes often led to unavoidable microbial contamination of the surgical area. Jaw bone quality and the degree of jaw bone resorption were evaluated by the periodontologist at implant placement. Tactile evaluation during drilling and assessment of the alveolar crest both radiographically and clinically allowed classification according to Lekholm & Zarb index (1985). A copy of this grading system was available while the score was given. Torque measurements were recorded during implant insertion, by means of an electronic torque force measurement device (OsseoCare, Nobel Biocare, Gothenburg, Sweden), which is a part of a controlled motor device. The latter measures the torque force while tapping or inserting the implant at the crestal, middle and the apical third of each implant insertion trajectory. The type of edentulism was classified according to the presence and location of natural teeth in the oral cavity related to implant location: full edentulism, teeth present only in the antagonistic jaw, teeth present in the same jaw where the implant is but not neighboring it and teeth neighboring the implant. An implant was considered a failure if a peri-implant radiolucency could be detected on the intra-oral radiographs, if an individual implants showed the slightest sign of mobility corresponding to a Periotest value (PTV) of > =5, or if the patient showed subjective signs of pain or infection, all these led to implant removal. Thus failed implant in our study is equal to lost implant. In the department a thorough sterility policy allows limiting the systemic use of antibiotics to well defined indications such as endocarditis prophylaxis, a remaining infection at the site of surgery, coughing or sneezing by the patient during surgery. Use of antibiotics pre-or immediately post implant surgery- was defined as yes or no. 30

41 Statistical analysis For exploratory purposes we constructed for each categorical factor a contingency table which cross-classifies the implants with respect to the levels of the factor and the failure status of the implant evaluated at abutments connection. The effect of each factor on the early failure of the implant was further evaluated by the Fisher s exact test of independence (e.g., Le, 2003, Sec. 6.6) in the case of the categorical factors and by the Wilcoxon rank-sum test (e.g., Le, 2003, Sec. 7.4) in the case of the continuous factors. Due to the fact that for most of the patients several implants were inserted and failure status was evaluated (clustering in the data) we cannot directly assume the independence between the failures events of the implants placed in a single patient. Consequently, both Fisher s exact test and Wilcoxon rank-sum test are anti-conservative, that is the P-values obtained by these tests are attenuated towards zero. To correct for clustering a possibility is offered by the significance test of the regression coefficient in the logistic regression model estimated using the generalized estimating equations (GEE) method (Liang & Zeger, 1986; Zeger & Liang, 1986). However, the test relies on the fact that sufficient amount of failures in each cell of the corresponding contingency table should be observed. Nevertheless, in our study, the observed number of failures appeared to be very low (in total only 14 out of 720, i.e. 1.9%). For this reason, we present the results (whenever numerically feasible) of the GEE logistic regression only as a mean of the sensitivity analysis to the Fisher s exact and Wilcoxon rank-sum test. In the following, symbol N.A. indicates that the p-value is not available due to numerical problems. Statistical analyses were performed using the R software (R Development Core Team, 2005) and the R package gee (Carey, 2002). III.3. Results: From the treated patient s population, a total of 14 implants out of the 720 installed implants appeared one to six months after placement as failed. None were lost in the days following abutment surgery. This corresponds to a 1.9% failure rate. These failures occurred in 14 patients (6 males; mean age 48.8, age range: 32-64), (8 females; mean age 31

42 59.4, age range: 54-66). For four patients the failed implants were replaced by new ones of the same type and no failure have been detected for any of them up to abutment connection. Implant length has no significant effect on early implant failures (p-values: Fisher = 0.94, GEE= 0.89) (Table 1). Table 1: Frequency and percentile distribution of length (mm) for the failed and successful implants Implant length (mm) Successful implants 243 (97.98%) 243 (98.38%) 220 (97.78%) Failed implants 5 (2.02%) 4 (1.62%) 5 (2.22%) Implant diameter has no effect on the early implant failures (p-values: Fisher =1.000, GEE= N.A.). Since no failure have been detected for implants with diameter 3.3 mm and 5 mm, the former was grouped with 3.75 mm, and the later with 4 mm, no statistical difference was detected between this two groups (p-values: Fisher =0.86, GEE= 0.86) (Table 2). Table 2: Frequency and percentile distribution of diameter (mm) for the failed and successful implants Implant diameter (mm) 3.3 and and 5 Successful implants 489 ((98.00% 217 (98.19%) Failed implants 10 (2.00%) 4 (1.81%) The implant location has no effect on the early implant failures (p-values: Fisher =0.54, GEE= 0.59) (Table 3). Table 3: Frequency and percentile distribution for the failed and successful implant location. Mandible anterior Mandible posterior Maxilla anterior Maxilla posterior Successful implants 154 (99.35%) 172 (97.18%) 181 (97.84%) 199 (98.03%) Failed implants 1 (0.65%) 5 (2.82%) 4 (2.16%) 4 (1.97%) 32

43 There was no significant effect of the presence of dehiscence or fenestration of the bone tissue during implant insertion on the early implant failures rate (p-values for dehiscence and fenestration: Fisher = 0.14, 0.38, GEE=0.15, 0.45, respectively) (Table 4). Table 4: Frequency and percentile distribution of the dehiscences and fenestrations of the jaw bone during implant insertion phase for the failed and successful implants. Dehiscence Fenestration No Yes No Yes Successful implants 640 (98.31%) 66 (95.65%) 683 (98.13%) 23 (95.83%) Failed implants 11 (1.69%) 3 (4.35%) 13 (1.87%) 1 (4.17%) The effect of implant perforation in the nasal cavity or in the sinus for all implants inserted in the maxilla (total of 383 implants) was evaluated. For 5 implants the perforation was not evaluated. There was no significant effect of implant perforation in the nasal cavity or the sinus on the early implant failures rate (p-values: Fisher =0.43, GEE= 0.25) (Table 5). Table 5: Frequency and percentile distribution of the occurrence of implant perforation in the nasal cavity or the sinus for the failed and successful implants inserted in the upper jaw. Perforation in the nasal cavity/sinus No Yes Successful implants 238 (97.54%) 138 (99.28%) Failed implants 6 (2.46%) 1 (0.72%) For a total of 5 implants, an apical lesion was detected on radiographs. This detection was related to the early implant failures. There was no significant effect of the apical lesion detection on the early implant failures, when a Fisher exact, while a significant effect appears when GEE analysis was used (p-values: Fisher =0.09, GEE= 0.02*) (Table 6). 33

44 Table 6: Frequency and percentile distribution of the apical lesion detection for the failed and successful implants. Apical lesion detection No Yes Successful implants 702 (98.18%) 4 (80.00%) Failed implants 13 (1.82%) 1 (20.00%) A total of nine implants were immediately inserted after tooth extraction, none of them failed at abutment connection. The possible impact of antibiotic prescription pre-or immediately after implant surgery was evaluated for two patients provided with 6 implants, the antibiotic use could not be evaluated-. There was no significant effect of antibiotic use on the early implant failures (p-values: Fisher =1.00, GEE= 0.95) (Table 7). Table 7: Frequency and percentile distribution of antibiotic prescription pre-or immediately after implant surgery for patients with and without early implant failures. Antibiotic prescription No Yes Successful implants 330 (98.21%) 371 (98.15%) Failed implants 6 (1.79%) 7 (1.85%) There was no significant effect of bone volume according to Lekholm & Zarb index (1985) on early implant failures rate. (For one implant bone volume was not evaluated) (p-values: Fisher =0.46, GEE= N.A ). Since no failures have been detected for implants inserted in bone quantity grade C and D, these two grades were grouped with grade E, no statistical difference was detected between these grouped grades and the other grades (pvalues: Fisher = 0.256, GEE= 0.235) (Table 8). Table 8: Frequency and percentile distribution of jaw bone quantity according to Lekholm & Zarb index (1985) for failed and successful implants. Grade A Grade B Grades (C,D,E) Successful implants 127 (98.45%) 320 (97.26%) 259 (99.23%) Failed implants 2 (1.55%) 9 (2.74%) 2 (0.77%) 34

45 There was no significant effect of bone quality according to Lekholm & Zarb index (1985) on early implant failures rate. (For one implant bone volume was not evaluated) (p-values: Fisher =0.73, GEE= N.A ). Since no failures have been detected for implants inserted in bone quality grade 4, this grade was grouped with grade 3. No statistical difference was detected between these grouped grades and the other grades (p-values: Fisher = 0.512, GEE= 0.460) (Table 9). Table 9: Frequency and percentile distribution of jaw bone quality according to Lekholm & Zarb index (1985) for failed and successful implants. Grade 1 Grade 2 Grades (3,4) Successful implants 107 (99.07%) 311 (97.49%) 288 ( 98.63%) Failed implants 1 (0.93%) 8 (2.51%) 4 (1.37%) The type of edentulism affected significantly early failures when the Fisher exact was used. A higher failure rate was noticed when the implants neighboring a teeth. (p-values: Fisher = *, GEE= N.A ). Since no failures have been detected for implants inserted in jaw having teeth but not neighboring the implant-, and in an edentulous jaw antagonizing teeth, these categories were grouped with full edentulism category. Significantly more failures were detected when an implant is neighboring a teeth comparing to the other grouped categories (p-values: Fisher /GEE: <0.001) (Table 10). Table 10: Frequency and percentile distribution of the type of edentulism, for patients with and without early implant failures. Full edentulism/ Teeth neighboring the implant Teeth in the antagonistic jaw only/ Teeth in the same jaw Successful implants 429 (99.54%) 277 (95.85%) Failed implants 2 (0.46%) 12 (4.15%) 35

46 Smoking habits affected significantly the early implant failures, according to the two statistical methods used (p-values: Fisher = <0.001 *, GEE= <0.001 *) (Table 11). Table 11: Frequency and percentile distribution of smoking habits, for patients with and without early implant failures. No smoking Smoking Successful implants 616 (98.88%) 90 (94.44%) Failed implants 7 (1.12%) 5 (5.56%) The early implant failures rate was compared between all female patients of 50 years following HRT, and those did not. No significant effect of HRT on early implant failures was found when Fisher exact was used, but significantly more failures were noted in patients who follow HRT, when GEE analysis was used (p-values: Fisher = 0.06, GEE= <0.001 *) (Table 12). Table 12: Frequency and percentile distribution of the failed and successful implants, for women of 50 years old following HRT or did not. Women 50 years following HRT Women 50 years do not follow HRT Successful implants 19 (90.48%) 362 (98.37%) Failed implants 2 (9.52%) 6 (1.63%) For a total of 274 patients provided with 682 implants, the placement torque measurements at the crestal, middle and apical thirds were evaluated. A total of 12 implants were failed in this group. No statistical effect of placement torque measurements on the early implant failure was detected, by either Wilcoxon rank-sum test, or GEE method [p-values were: (0.87, and 0.90 respectively), (0.81, and 0.79 respectively) and (0.27, and 0.24 respectively) for the crestal third, middle third and apical third respectively (Figure 1). 36

47 Figure 1: Box-plots of the placement torque measurements (Ncm) for the successful and failed implants, and results of the Wilcoxon rank-sum test in the crestal, middle and in apical thirds. Both systemic disease and medical therapies were analyzed using the two statistical methods separately. Since the multivariable, and in many cases also univariate GEE analyses are impossible, these results were given a descriptive character only. Certain factors, such as hypertension, ischemic cardiac disease, osteoporosis, hypo- or hyperthyroidism, controlled diabetes type II, rheumatoid arthritis, coagulation problems, chemotherapy, claustrophobia, asthma, hypercholesterolemia, radiotherapy of the concerned area, and antidepressant and steroid medications did not lead to an increased incidence in the early failures (p-values were > 0.05 with Fisher, and GEE statistical 37

48 methods). A significant effect of gastric problems on early implant failures appeared when Fisher exact and GEE analysis were used (p-values: Fisher = 0.04*, GEE= 0.01*). A significant effect of Crohn s disease on early implant failures appeared when GEE analysis was used (p-values: Fisher = 0.21, GEE= 0.02*). A significant effect of Diabetes type I on early implant failures was detected when Fisher exact was used (p-values: Fisher = 0.02*, GEE= N.A.). Significantly more early failures occurred in women who underwent radical hysterectomy, when Fisher exact and GEE analysis were used (p-values: Fisher = 0.04*, GEE= 0.04*) (Table 13). Table 13: Frequency and percentile distribution of the systemic diseases and therapies for patients with and without early failure, and the p-values of 2 different statistical analyses used. No Yes p-value Factor Successful Failed Successful Failed Fisher GEE Hypertension 589 (98.00%) 12 (2.00%) 117 (98.32%) 2 (1.68%) Cardiac problem 638 (97.85%) 14 (2.15%) 68 (100.00%) 0 (0.00%) 0.38 N.A Gastric problem 667 (98.38%) 11 (1.62%) 39 (92.86%) 3 (7.14%) 0.04 * 0.01 * Osteoporosis 677 (97.97%) 14 (2.03%) 29 (100.00%) 0 (0.00%) 1.00 N.A. Hypothyroid 685 (98.00%) 14 (2.00%) 21 (100.00%) 0 (0.00%) 1.00 N.A Hyperthyroid 702 (98.04%) 14 (1.96%) 4 (100.00%) 0 (0.00%) 1.00 N.A. Chemotherapy 699 (98.04%) 14 (1.96%) 7 (100.00%) 0 (0.00%) 1.00 N.A Radiotherapy 703 (98.05%) 14 (1.95%) 3 (100.00%) 0 (0.00%) 1.00 N.A. Crohn s disease 695 (98.16%) 13 (1.84%) 11 (91.67%) 1 (8.33%) * Diabetes I 706 (98.19%) 13 (1.81%) 0 (0.00%) 1 (100.00%) 0.02 * N.A. Diabetes II 682 (98.13%) 13 (1.87%) 24 (96.00%) 1 (4.00%) Rheumatoid arthritis 693 (98.16%) 13 (1.84%) 13 (92.86%) 1 (7.14%) Coagulation 649 (98.18%) 12 (1.82%) 57 (96.61%) 2 (3.39%) Claustrophobia 694 (98.02%) 14 (1.98%) 12 (100.00%) 0 (0.00%) 1.00 N.A. Antidepressant 647 (98.18%) 12 (1.82%) 59 (96.72%) 2 (3.28%) medication Steroid medication 702 (98.04%) 14 (1.96%) 4 (100.00%) 0 (0.00%) 1.00 N.A. Hypercholesterol 673 (97.14%) 14 (1.94) 33 (100.00%) 0 (0.00 %) 1.00 N.A. Asthma 685 (98.74%) 13 (1.86%) 21 (95.45%) 1 (4.55%) Radical hysterectomy 470 (98.74%) 6 (1.26%) 20 (90.91%) 2 (9.09%) 0.04* 0.04* 38

49 II.5. Discussion: This study revealed a high success rate of 98.1% up to abutment connection, which can be related to the use of implants with an oxidized surface (TiUnite implants). Indeed in two previous studies from the same authors, using exclusively or mostly machined surfaces of the same implant type, the failure rate, up to abutment connection, reached respectively 2.8 % and 3.6 % (van Steenberghe et al. 2002, Alsaadi et al. 2007). Due to the fact that the number of early failed implants was extremely low (14 out of 720, i.e. 1.9%), this study could only identify potentially influential factors for the implant failure and could not draw definitive conclusions. Future studies should direct their attention to the evaluation of the following factors: apical lesions around the recipient site, vicinity with natural dentition, smoking habits, hormone replacement therapy, gastric problems, Crohn s disease, diabetes I, and radical hysterectomy. Although the apical lesions seems to affect significantly the early failure rate with the use of TiUnite implants, this rate is relatively low when compared to machined surfaced implants, which fail to achieve osseointegration and are thus lost. This may be explained by the fact that a faster bone apposition occurring with the TiUnite prevents the spread of the inflammatory cells from the remaining apical lesion along the implant surface. With machined surfaces the bone apposition is much slower and thus this interposition of inflammatory cells can occur reaching up to the coronal end. Such implants become loose because of the saucerization. With a TiUnite surface the inflammatory focus is reactivated by the surgical trauma, but the implant remains osseointegrated (Quirynen et al. 2005). Several studies revealed the negative effect of smoking on osseointegration, and its dose-related effect (for review, Bain 1996). The present findings seem in accordance with the previous reports. Many recent studies which have used surgical preparation adapted to the bone density, modified surface implants have reported survival rates for short implants and for wide diameter implants which were comparable with those obtained with long-implants 39

50 and standard diameter implants (for review see Renouard & Nisand 2006). In the present study there was no effect of implant length on implant failure. Moreover, poor bone quality did not effect the early failure rate, as ascertained by Fisher test for osteoporosis and implant diameter. Crohn s disease can, as it is a generalized autoimmune disease, affect the entire gastro-intestinal system, and thus even lead to periodontal lesions (van Steenberghe et al. 1976). Crohn s disease is characterized by the presence of many antibody-antigens complexes, leading to autoimmune inflammatory process in several parts of the body. Symptoms are enteritis, vasculitis, recurrent oral ulceration, arthritis or keratoconjuctivitis. The same can occur at the interface with biocompatible implants, normally considered by the host as a part of the body. In Crohn s patients, they could be recognized as non-self, thus affecting the outcome of implant osseointegration (van Steenberghe et al. 2002). Moreover, the malnutrition encountered in Crohn s patients can also cause a deficient bone healing around the implant (Esposito et al. 1998). Moy et al. 2005, in a retrospective study, found that women on estrogen replacement had a significantly lower success rate than the healthy population. Postmenopausal women, not on hormone replacement therapy, did not have this increased failure rate. In the present study there is tendency to more failures for women with HRT. Conclusion: From the present study, homogenuously using TiUnite implants, the incidence of early failures was so small that statistical analysis of interfering factors became difficult. It appears for example that poor bone quality did not influence the outcome of osseointegration, as ascertained by Fisher test for osteoporosis and implant diameter. On the other hand gastric problems, Crohn, diabetes type I and radical hysterectomy seem to increase the incidence of early failures. Thus in the presence of such diseases the choice of osseointegrated implants should eventually be put into the perspective of more classical prosthetic approaches which do not suffer such drawbacks. 40

51 III.6. References: Albrektsson, T., Johansson, C., Lundgren, A., Sul, Y. & Gottlow, J. (2000) Experimental studies on oxidized implants. A histomorphometrical and biochemical analysis. Applied Osseointegration Research 1, Alsaadi, G., Quirynen, M., Komarek, A. & van Steenberghe, D. (2007) The impact of local and systemic factors on the incidence of oral implant failures, up to abutment connection. Journal of Clinical Periodontology 7, Bain, C. A. (1996) Smoking and implant failure--benefits of a smoking cessation protocol. International Journal of Oral and Maxillofacial Implants 11, Bass, S. & Triplett, R. (1991) The effects of preoperative resorption and jaw anatomy on implant success. A report of 303 cases. Clinical Oral Implants Research 2, Buser, D., Schenk, R., Steinemann, S., Fiorellini, J., Fox, C. & Stich, H. (1991) Influence of surface characteristics on bone integration of titanium implants. A histomorphometric study in miniature pigs. Journal of Biomedical Materials Research 25, Carey, V. (2002) gee: Generalized Estimation Equation solver. R package version , ported to R by Thomas Lumley (version 3.13 & 4.4) and Brian Ripley (version 4.13). Carlsson, L., Rostlund, T., Albrektsson, B. & Albrektsson, T. (1988) Removal torques for polished and rough titanium implants. International Journal of Oral & Maxillofacial Implants 3, Esposito, M., Hirsch, J., Lekholm, U. & Thomsen, P. (1998) Biological factors contributing to failures of osseointegrated oral implants. I. Success criteria and epidemiology. European Journal of Oral Sciences 106, Esposito, M., Thomsen, P., Ericson, L.E. & Lekholm, U. (1999) Histopathologic observations on early oral implant failure. International Journal of Oral & Maxillofacial Implants 14, Hall, J. & Lausmaa, J. (2000) Properties of a new porous oxide surface on titanium implants. Applied Osseointegration Research 1,5-8. Henry, P., Tan, A., Allan, B., Hall, J. & Johansson, C. (2000) Removal torque comparison of TiUnite and turned implants in the greyhound dog mandible..appllied Osseointegration Research 1, Huré, G., Donath, K., Lesourd, M., Chappard, D. & Basle, M. (1996) Does titanium surface treatment influence the bone-implant interface? SEM and histomorphometry in a 6-month sheep study. International Journal of Oral & Maxillofacial Implants 11, Jaffin, R. & Berman, C. (1991) The excessive loss of Brånemark fixtures in type IV bone: a 5-year analysis. Journal of Periodontology 62,

52 42 Ivanoff, C., Widmark, G., Johansson, C. & Wennerberg, A. (2003) Histologic evaluation of bone response to oxidized and turned titanium micro-implants in human jawbone. International Journal of Oral & Maxillofacial Implants 18, Le, C. T. (2003). Introductory Biostatistics. Hoboken: John Wiley & Sons. ISBN: Lekholm, U. & Zarb, G. A. (1985) Patient selection and preparation. In: Brånemark, P-I., Zarb, G., Albrektsson, T. eds. Tissue integrated prosthesis: Osseointegration in clinical dentistry. p.199, Chicago, Quintessence publishing Co Inc. Liang, K.Y. & Zeger, S.L. (1986) Longitudinal data analysis using generalized linear models. Biometrika 73, Mombelli, A. & Gionca, N. (2006) Systemic disease affecting osseointegration therapy. Clinical Oral Implants Reasearch 17, Moy, P.K., Medina, D., Shetty, V. & Aghaloo, T. (2005) Dental implant failure rates and associated risk factors. International Journal of Oral & Maxillofacial Implants 20, Quirynen, M., Vogels, R. Alsaadi, G., Naert, I., Jacobs, R. & van Steenberghe, D. (2005) Predisposing conditions for retrograde periimplantitis, and treatment suggestions. Clinical Oral Implants Research 16, R Development Core Team (2005). R: A language and environment for statistical computing, ISBN , R Foundation for Statistical Computing, Vienna, Austria. Renouard, F. &, Nisand, D. (2006) Impact of implant length and diameter on survival rates Suppl Clinical Oral Implants Research 2: Rocci, A., Martignoni, M. & Gottlow, J. (2003) Immediate loading of Brånemark System TiUnite and machined-surface implants in the posterior mandible: a randomized open-ended clinical trial. Clinical Implant Dentistry and Related Research 5, van Steenberghe, D., Vanherle, G.V., Fossion, E. & Roelens, J. (1976) Crohn's disease of the mouth, report of case. Journal of Oral Surgery 34, van Steenberghe, D., Lekholm, U., Bolender, C., Folmer, T., Henry P., Herrmann I., Higuchi K., Laney W., Linden U. & Åstrand P. (1990) Applicability of osseointegrated oral implants in the rehabilitation of partial edentulism: a prospective multicenter study on 558 fixtures. International Journal of Oral & Maxillofacial Implants 5, van Steenberghe, D., Yoshida, K., Papaioannou, W., Bollen, C., Reybroeck, G.I. & Quirynen, M. (1997) Complete nose coverage to prevent airborne contamination via nostrils is unnecessary. Clinical Oral Implants Research 8, van Steenberghe, D., Jacobs, R., Desnyder, M., Maffei, G. & Quirynen, M. (2002) The relative impact of local and endogenous patient-related factors on implant failure up to the abutment stage. Clinical Oral Implants Research 13,

53 van Steenberghe, D., Quirynen, M., Molly, L. & Jacobs, R. (2003) Impact of systemic diseases and medication on osseointegration. Periodontology , Zeger, S.L. & Liang, K.Y. (1986) Longitudinal data analysis for discrete and continuous outcomes. Biometrics 42,

54

55 CHAPTER IV. The impact of local and systemic factors on the incidence of late oral implant losses IV.1. Introduction: When a properly documented implant system with a long-term success rate has been selected, long lasting prosthetic survival can be expected. Lindquist et al. (1996) evaluating symphyseal fixed prostheses supported by implants, reported implants cumulative success rate of 98.9% both after 10 and 15 years. Implant failure can compromise the prosthetic rehabilitation. Implant failures can be subdivided into early and late failures, defined as occurring prior to or at abutment connection (= early) or after occlusal loading is established (= late). The aetiology of failures may be different in these two time periods. An early failure of an implant results from an inability to establish an intimate bone -to- implant contact (Esposito et al. 1998, Quirynen et al. 2002). This means that bone healing after implant insertion is impaired or even jeopardized. Both systemic and local factors can interfere with these primarily cellular events. The mechanisms that normally lead to wound healing by means of bone apposition do not take place, and rather a fibrous scar tissue is formed in between the implant surface and surrounding bone (Esposito et al. 1999). This can lead to epithelial downgrowth, a so-called saucerization or marsupialisation of the implant, which often results in mobility or even implant loss. Thus the anchoring function of the endosseous implant cannot be properly performed. Late implant failures on the other hand are influenced by both the microbial environment and the prosthetic rehabilitation. These failures have been associated with both peri-implantitis resulting from plaque-induced gingivitis and peri-implantitis and/ or occlusal overloading (van Steenberghe et al. 1990, Quirynen et al. 2002). Long-term maintenance of a rigid implant to bone interface requires continuous bone remodelling (Robert et al. 1992). This may be influenced by implant, patient and surgery related factors. 45

56 Systemic factors play a prominent role for early failures (van Steenberghe et al. 2002, van Steenberghe et al. 2003, Mombelli & Gionca 2006). For late failures the exposure of the implants and superstructures to the microbial environment and occlusal forces are supplementary risk factors. It renders the detection of impact of systemic and local factors more difficult. The aim of this large scale retrospective study is to assess the influence of these systemic and local factors on the occurrence of late implant losses up to 2 years after abutment connection. In order to achieve a good analysis of these factors the following two questions were set: 1) what is the effect of all the factors on the failure rate of the implants? -> Then the unit for the analysis is the implant and one only has to take into account the fact that implants within each patient are correlated. 2) what is the effect of all these factors on patient's ability to use the implants? -> Then the unit for the analysis is indeed the patient. However, one has to clearly define what failure means. The failure here is at implant level and not patient level. For example since from a prosthetic point of view, a patient who has received 5 or 6 implants (fully edentulous) looses one or even two implants - they will often even not be replaced - since the prosthetic treatment can still be successfully performed. The answer to the first question is the goal of this retrospective study. IV.2. Materials and methods: Materials: The files of 700 patients were collected randomly from the total patient group treated by means of endosseous implants at the Department of Periodontology of the University Hospital of the Catholic University Leuven. It is a general policy of the department to accept for treatment all patients who can benefit from implants for their oral rehabilitation. Thus even if systemic or local factors are detected which can compromise the outcome, this treatment modality is considered versus classical prosthetic approaches. All patients were treated by means of Brånemark 46

57 system implants, (Nobel Biocare, Gothenburg, Sweden). A minimal bone height of 7mm was the requirement for implant placement. Surgery in the department is performed under strict aseptic conditions (van Steenberghe et al. 1997). A thorough sterility policy allows limiting the use of antibiotics to well defined indications such as endocarditis prophylaxis, a remaining infection at the site of surgery, coughing or sneezing by the patient during surgery. It involves the use of two suction devices, a nose cap etc. The end point observation was evaluating the failure rate of the implants up to 2 years after abutment installation. The study involved all implants that did not encounter loss before or at abutment surgery (early losses) and implants for which it was possible to evaluate its status 2 years after abutment surgery. The following implants were not considered for the analysis: Implants which failed before or at abutment surgery. Implants that although not failed but in patients who could not be followed for up to 2 years after abutment surgery. The remaining data of 412 patients (240 females) provided with a total of 1514 implants could thus be analyzed. The later data set implant losses were analyzed and related to health and behavioral factors, implant lengths and diameters, bone quality and quantity, insertion site, type of edentulism, prescription of antibiotics pre- or immediately after surgery, dehiscence or perforation of the jaw bone during surgery, Periotest (Siemens, A G, Bensheim, Germany) values (PTV) at implant insertion (by using a temporary abutment) as well as at abutment connection. The PTV measures the stability of the implant-bone continuum by tapping with an electro-magnetically driven rod on the implant. The outcome is expressed in arbitrary units, reaching from 8 to + 50 (Tricio et al. 1995). Implants should lead to values below +5; the more negative, the better the stability. PTVs were only performed on a fraction of the patient material because of irregular availability of the machinery. 47

58 Data collection and analysis: Only screw-shaped Brånemark system implants (Nobel Biocare, Gothenburg, Sweden) were used either with a machined (n = 1316) or a moderately rough very oxidized Ti-Unite surface (n = 198). The general health and the behavioral history of the patient were carefully recorded on the patient s files. It was obtained after thoroughly questioning the patients preoperatively as a routine procedure. An implant was considered a failure if a peri-implant radiolucency could be detected on the intra-oral radiographs, if an individual implant showed the slightest sign of mobility corresponding to a Periotest value (PTV) of > =5, or if the patient showed subjective signs of pain or infection, all these led to implant removal. Thus failed implant in our study is equal to lost implant. Jaw bone quality and the degree of jaw bone resorption were evaluated by the periodontologist at implant placement. Tactile evaluation during drilling and assessment of the alveolar crest both radiographically and clinically allowed classification according to Lekholm & Zarb index (1985). The following health or behavioral factors were particularly questioned and assessed (often also by consulting the files from other medical departments): smoking habits, hypertension, ischemic cardiac problems, coagulation anomalies, gastric problems such as ulcers, osteoporosis, hypo- or hyperthyroidism, hypercholesterolemia, asthma, diabetes type I or II, Crohn s disease, rheumatoid arthritis, chemotherapy, intake of medication (antidepressants, steroids). Smoking patients were allocated to one of the following three categories (<10 cigarettes/ day, cigarettes/day or > 20 cigarettes/day). Local bone factors, such as radiotherapy of the maxillofacial region involved, were also recorded. Finally, a special note was made for patients with claustrophobia. These patients were treated with reduced coverage of the face, often without a nose cap and as such with a serious breach of asepsia (van Steenberghe et al. 1997). The type of edentulism was classified according the presence and location of natural teeth in the oral cavity related to implant location: full edentulism, teeth present only in the 48

59 antagonistic jaw, teeth present in the same jaw as the implants either in the vicinity or not from the implant. Statistical analysis: Logistic regression models were used to evaluate the effect of explanatory variables on the late loss of the implant. Generalized estimating equation (GEE) method (Liang and Zeger, 1986; Zeger and Liang, 1986) was used to account for the fact that several implants were available for a single patient (repeated observations). The question of the study was: what is the effect of all the factors on the failure rate of the implants? Thus the unit which was considered for the analysis is the implant, and the fact that implants within each patient are correlated was taken into account in the analysis. Firstly, a univariate effect of each implant related, behavioral, and local bone factor on the late losses was evaluated by fitting univariate GEE logistic regression model. Odds ratios and their 95% confidence intervals based on the robust standard errors from the GEE logistic regression model were computed. The Wald test based on robust standard errors was used to assess the significance of each factor. Additionally, we adjusted obtained P-values for multiple testing using the method of Holm (1979). For categorical factors with more than 2 levels, differences in the losses proportions between these levels were evaluated by multiple comparison provided by the robust Wald s P-values adjusted using the method of Holm (1979). Secondly, a multivariable was used to evaluated the effect of the health factors when controlled for the behavioral, implant related and local bone factors that were univariately (at 5%) significant. Namely, the following factors were controlled for: location (mandible, anterior/mandible, posterior/maxilla, anterior/maxilla, posterior), diameter. Although univariately significant, jaw (mandible/maxilla), and site (anterior/posterior) was not controlled for since these together determine location. Furthermore, although univariately significant, bone quality, PTV at implant insertion, and PTV at abutment connection (aptv) were not controlled for since the data were available for only a limited subgroup of patients. Due to the fact that no implant losses have been observed in the group of patients having a given disease, the effect of chemotherapy, diabetes type 49

60 I, diabetes type II, rheumatoid arthritis, claustrophobia, steroid medication, could not be assessed statistically. A statistical analysis was performed using ther software (R Development Core Team, 2006) and the R package gee (Carey, 2002). IV.3. Results: A total of 101 implants of different lengths and diameters out of the 1514 installed implants, appear to have failed between abutment connection and the 2 years follow-up. Table (1) describes the distribution of numbers of implant placed per patient and provides a summary statistics of the number of implants failed. Table 1: The distribution of number of implants placed per patient, and number of patients those experienced one or more implant failures. N implants per patient N patients N patients with 1 or more losses Average number of lost implants per patient Average failure rate per patient

61 When implant characteristics were related to late implant losses, implant diameter and location in the jaw were relevant, while implant length did not (p-value=0.01, < 0.001, 0.34 respectively). Smoking habits could not be related to implant losses (p-value=0.28). The same analysis was applied for the type of edentulism (p-value=0.85). The presence of dehiscence or fenestration of the bone tissue at implant insertion did not affect the late outcome (p-value= 0.58, 0.31 respectively). The same hold true for bone volume (p-vale=0.34), while bone quality affected significantly late implant losses (p-value=0.03). A summary of the univariate GEE logistic regression for all the above mentioned factors can be found in Table 2: a,b. Significantly more late losses occurred with higher PTV values either at implant insertion or at abutment connection, which reflects a lower rigidity (p-value/adjusted p-value <0.001) (Figure 1). When a multiple comparison (among the levels of categorical factors) was done, significantly more losses could detected with implants of wide platform (5 mm) when compared to implants of regular platform, diameter (4 mm) or ( 3.75mm) (p-value = 0.04, respectively) [p-value = 0.004, 0.02, Odds ratio (95% C. I.)= 2.70 ( ), 2.73 ( ) respectively]. There was significantly more losses in bone quality grade 4 (soft bone with little cortical bone) when compared to grade 2 [p-value=0.03, Odds ratio (95% C.I.): 3.92 ( )]. The mandibular anterior region experienced significantly less losses than the posterior region [p-value=0.04, Odds ratio (95% C.I.): 3.42 ( )], and than the maxillary posterior [p-value=<0.001, Odds ratio (95% C.I.): ( )], anterior region [pvalue=0.02, Odds ratio (95% C.I.): 3.99 ( )]. Within the upper jaw significantly more implant losses occurred in the posterior regions when compared to the anterior ones [p-value=0.04, Odds ratio (95% C.I.): 1.71 ( )]. 51

62 Table 2 a: Univariate GEE logistic regression: implant related, behavioral and local bone factors, the total number of patients to whom the factors were evaluated and the distribution of the failed and successful implants. Adjusted P-values corrected for multiple tests given in Table 2 using the method of Holm. * Significant P-value < Univariate GEE Logistic Regression: Summary Factor (Patients/Implants) Successful Failed Odds Ratio 95% Conf. Interval Smoking (412/1514) p-value: 0.28 /Adjusted p-value: (351 patients) 1211 (93.80%) 80 (6.20%) 1 <10 (19 patients) 62 (89.86%) 7 (10.14%) 1.39 (0.38, 5.09) (20 patients) 47 (85.45%) 8 (14.55%) 2.92 (0.97, 8.77) >20 (22 patients) 93 (93.94%) 6 (6.06%) 1.21 (0.39, 3.73) Type edentulism (375/1416) p-value: 0.85 /Adjusted p-value:1.000 No teeth 454 (95.38%) 22 (4.62%) 1 In the ant. jaw only 187 (95.41%) 9 (4.59%) 1.31 (0.39, 4.35) In the same jaw 309 (95.37%) 15 (4.63%) 1.03 (0.42, 2.54) Neighb. implant 397 (94.52%) 23 (5.48%) 1.34 (0.65, 2.75) Dehiscence (230/763) p-value: 0.59 /Adjusted p-value:1.000 No 693 (97.06%) 21 (2.94%) 1 Yes 48 (97.96%) 1 (2.04%) 0.58 (0.08, 4.28) Fenestration (226/752) p-value: 0.32 /Adjusted p-value:1.000 No 714 (97.14%) 21 (2.86%) 1 Yes 16 (94.12%) 1 (5.88%) 2.46 (0.42, 14.52) Bone quality (296/921) p-value: 0.03 * /Adjusted p-value: (94.12%) 4 (5.88%) (97.66%) 8 (2.34%) 0.33 (0.10, 1.00) (92.82%) 27 (7.18%) 0.89 (0.30, 2.59) (90.37%) 13 (9.63%) 1.29 (0.41, 4.05) Bone quantity (297/930) p-value: 0.34 /Adjusted p-value:1.000 A 163 (95.32%) 8 (4.68%) 1 B 412 (96.04%) 17 (3.96%) 0.77 (0.33, 1.77) C 231 (93.52%) 16 (6.48%) 0.95 (0.33, 2.69) D 51 (91.07%) 5 (8.93%) 1.95 (0.56, 6.79) E 21 (77.78%) 6 (22.22%) 2.83 (0.43, 18.46) 52

63 Table 2 b: Univariate GEE logistic regression: implant related, behavioral and local bone factors, the total number of patients to whom the factors were evaluated and the distribution of the failed and successful implants. Adjusted P-values corrected for multiple tests given in Table 2 using the method of Holm.* Significant P-value < Univariate GEE Logistic Regression: Summary Factor (Patients/Implants) Successful Failed Odds Ratio 95% Conf. Interval Jaw (412/1514) p-value: <0.001 * /Adjusted p-value: 0.006* Mandible 671 (96.13%) 27 (3.87%) 1.00 Maxilla 742 (90.93%) 74 (9.07%) 2.59 (1.50, 4.49) Site (412/1514) p-value: <0.001 * /Adjusted p-value: 0.003* Anterior 766 (95.75%) 34 (4.25%) 1.00 Posterior 647 (90.62%) 67 (9.38%) 2.14 (1.43, 3.21) Location (412/1514) p-value: <0.001 * /Adjusted p-value: 0.003* Mandible, anter. 379 (97.93%) 8 (2.07%) 1.00 Mandible, poster. 292 (93.89%) 19 (6.11%) 3.42 (1.29, 9.06) Maxilla, anter. 387 (93.70%) 26 (6.30%) 3.99 (1.58, 10.07) Maxilla, poster. 355 (88.09%) 48 (11.91%) 6.83 (2.65, 17.57) Length (412/1514) p-value: /Adjusted p-value: (93.55%) 88 (6.45%) 1 < (89.17%) 13 (10.83%) 1.42 (0.69, 2.92) Diameter (412/1514) p-value: * /Adjusted p-value: , (93.73%) 81 (6.27%) (93.11%) 14 (6.89%) 1.18 (0.57, 2.41) 5 20 (76.92%) 6 (23.08%) 4.25 (1.64, 11.00) Surface (412/1514) p-value: 0.16 /Adjusted p-value: Machined 1223 (92.93%) 93 (7.07%) 1 Ti-Unite 190 (95.96%) 8 (4.04%) 0.51 (0.19, 1.31) 53

64 Figure The frequency 1: Box-plots distribution of Periotest of successful value at implant and failed insertion implants (PTV), within and at abutment the patient suffering connection from a known (aptv) systemic for the disease successful is described and failed implants. in Table 3. 54

65 This was certified through in a multivariate analysis controlled for the other diseases and other factors which are related to late losses such as implant location and diameter. Table 3: The frequency distribution of successful and failed implants, for patients with and without known diseases or healthy conditions. Factor No Yes N Success Failed N Success Failed patients patients Hypertension (93.23%) 87 (6.77%) (93.86%) 14 (6.14%) Cardiac problem (93.46%) 94 (6.54%) (90.79%) 7 (9.21%) Gastric problem (93.29%) 98 (6.71%) 14 51(94.44%) 3 (5.56%) Osteoporosis (93.64%) 92 (6.36%) (86.76%) 9 (13.24%) Hypothyroidism (93.30%) 94 (6.70%) (93.69%) 7 (6.31%) Hyperthyroidism (93.43%) 98 (6.57%) 6 19 (86.36%) 3 (13.64%) Chemotherapy (93.28%) 101 (6.72%) 3 10 (100.00%) 0 (0.00%) Radiotherapy (93.46%) 98 (6.54%) 2 12 (80.00%) 3 (20.00%) Crohn s disease (93.49%) 98 (6.51%) 2 6 (66.67%) 3 (33.33%) Diabetes I (93.32%) 101 (6.68%) 1 1 (100.00%) 0 (0.00%) Diabetes II (93.18%) 101 (6.82%) 9 33 (100.00%) 0 (0.00%) Rheumatoid arthritis (93.20% 101 (6.80%) 6 28 (100.00%) 0 (0.00%) Coagulation problems (93.17%) 99 (6.83%) (96.92%) 2 (3.08%) Claustrophobia (93.30%) 101 (6.70%) 3 7 (100.00%) 0 (0.00%) Antidepressant (93.39%) 92 (6.61%) (92.62%) 9 (7.38%) medication Steroid medication (93.25%) 101 (6.75%) 5 17 (100.00%) 0 (0.00%) Hypercholesterolemia (93.24%) 97 (6.76%) (94.94%) 4 (5.06%) Asthma (93.35%) 100 (6.65%) 5 9 (90.00%) 1 (10.00%) Certain systemic factors, such as cardiac and gastric diseases, controlled diabetes type II, coagulation problems, hypertension, hypo- or hyperthyroidism, hypercholesterolemia, asthma, osteoporosis, Crohn s disease, claustrophobia, antidepressant medication did not lead to an increased incidence in the late losses (p-value > 0.05). Radiotherapy significantly increased the number of implant losses (p-value=0.003) (Table 4). 55

66 Table 4: Multivariable GEE logistic regression: implant related, behavioral and local bone factors and health factors. * Significant P-value < 0.05 Multivariable GEE Logistic Regression Number of patients = 412 Number of implants = 1514 Factor Odds Ratio 95% CI p-value Hypertension 0.85 (0.29, 2.43) 0.76 Cardiac problem 2.09 (0.65, 6.73) 0.23 Gastric problem 0.99 (0.24, 4.01) 0.98 Osteoporosis 2.73 (0.79, 9.39) 0.11 Hypothyroidism 0.95 (0.29, 3.02) 0.93 Hyperthyroidism 0.57 (0.03, 12.99) 0.73 Radiotherapy 3.32 (1.49, 7.35) * Crohn s disease (0.73, ) 0.09 Coagulation problems 0.29 (0.047, 1.75) 0.18 Antidepressant medication 0.67 (0.22, 2.05) 0.48 Hypercholesterolemia 1.19 (0.37, 3.83) 0.77 Asthma 2.55 (0.65, 10.01) 0.18 Location (mandible, poster.) 3.86 (1.41, 10.54) <0.001 * Location (maxilla, anter.) 3.90 (1.49, 10.22) Location (maxilla, poster.) 7.10 (2.69, 18.78) Diameter (4 mm) 0.91 (0.45, 1.85) 0.09 Diameter (5 mm) 2.98 (1.08, 8.22) IV. 4. Discussion: Renouard and Nisand (2006) reported in a review paper, that there is a trend for machined-surface implants a trend which does not apply to rougher surfaces such as TiUnite for an increased losses rate with short and wide-diameter implants. Although in the present study the statistical analysis revealed no significant difference in late failure rate between the machined surface and TiUnite surface implants, but there is a trend for more implant losses with machined surface. Such increased implant losses of wide-diameter implants was mainly associated with a learning curve, poor bone density, implant design and site preparation, and the fact that it is usually used as rescue implants. The present study, confirmed the trend for more losses occurring with wide-diameter implants. Here too, these implants were mostly installed in sites with poor bone quality and quantity. These confounding factors offer a possible explanation. 56

67 Low bone density -as assessed clinically or radiologically- has also been pointed out as a possible reason for non-integration (Engquist et al. 1988, Friberg et al.1991, Jaffin & Berman 1991). In our present findings bone quality type 4 is indeed associated with slightly higher implant losses. The effect of smoking on the late failure rate is not evidenced in the present study. It is well known to have an impact on early failures rate (Bain & Moy 1993). This may be explained by the effect of smoking on the wound healing process, in early stage of osseointegration (Alsaadi et al. 2007) while other factors are more predominant for the late failures. The PTV do not have a direct relation to the aim of the study, since it is not a systemic or a local factor but a method to measure the stability of the implant and therefore only an indirect measure for the bone quality. Higher PTVs at abutment placement are associated with more late implant failures. The only possible explanation is the less rigid bone to implant interface which may lead under loading to a reversal of the osseointegration process: a differentiation to fibrous scar tissue. Lack of intimate bone apposition can result in implant marsupialisation (Ivanoff et al. 1996, Szmukler-Moncler et al. 2000). The higher implant losses in the upper jaw vs. the lower, and in the posterior vs. the anterior region reflect the thin cortical bone combined with less dense trabecular bone often observed in the upper jaw (Jacobs 2003). The poor degree of bone mineralization will reveal itself on the radiographs (Friberg et al. 1995, 1999). Distal regions also have higher chewing forces when compared to the anterior. On average, clenching forces are 3 times higher in the molar vs. the incisor regions (Helkimo 1977) The long term effects of radiotherapy on bone quality are poorly documented (Keller 1997). The dramatic effect of radiotherapy on late failures should lead to a cautious application of osseointegration in such patients. But even with a higher failure rate the implant retained prosthesis is often the only option because of xerostomia. The decreased bone blood supply can lead to osteoradionecrosis (el Askary et al. 1999, Marx et al. 1987). Crohn s disease can, as it is a generalized autoimmune disease, affect the entire gastrointestinal system, and thus even lead to periodontal lesions (van Steenberghe et al. 1976). 57

68 Crohn s disease is characterized by the presence of many antibody-antigens complexes, leading to autoimmune inflammatory process in several parts of the body, such as enteritis, vasculitis, recurrent oral ulceration, arthritis or keratoconjuctivitis. The same could occur at the interface with biocompatible implants, normally considered by the host as a part of the body, but which could in Crohn s patients recognized as non-self, thus affecting the outcome of implant osseointegration (van Steenberghe et al. 2002). Moreover, the malnutrition encountered in Crohn s patients could also cause a deficient bone healing around the implant (Esposito et al. 1998). In the present study there is a trend for more losses with Crohn s disease. Osteoporosis has been defined as a decrease in bone mass and bone density and an increased risk and/or incidence of bone fracture. The Word Health Organization has established diagnostic criteria for osteoporosis based on bone density measurements determined by dual energy x-ray absorptiometry. A diagnosis of osteoporosis is attributed if the bone mineral density level is 2.5 standard deviations below that of a mean young population (Glaser & Kaplan 1997). The disease may have an influence on periodontal attachment losses (Wactawski-Wende et al. 1996). Although no studies prove an association between implant failure and the state of osteoporosis, it has been suggested as a risk factor for implant failure, especially for postmenopausal women (Becker et al. 2000). In a previous study a significant association was detected between early implant failures and osteoporosis (Alsaadi et al. 2007). In the present study, although osteoporosis has no significant effect on implant losses, there is a trend for more failures. IV. 5. Conclusion: The present study has identified through a multivariate analysis the weighed impact of several factors which contribute to late failures. Implant location in the oral cavity and radiotherapy seem predominant to explain the occurrence of implant losses. On the other hand smoking and systemic health factors seem not prominent players in the etiology of late implant losses. 58

69 IV.6. References: Alsaadi, G., Quirynen, M., Komarek, A. & van Steenberghe, D. (2007) The impact of local and systemic factors on the incidence of oral implant failures, up to abutment connection. Journal of Clinical Periodontology 7, Bain, C. A. (1996) Smoking and implant failure--benefits of a smoking cessation protocol. International Journal of Oral and Maxillofacial Implants 11, Bain, C. A. & Moy, P. K. (1993) The association between the failure of dental implants and cigarette smoking. International Journal of Oral & Maxillofacial Implants 8, Becker W., Hujoel P.P., Becker B.E. &Willingham H. (2000) Osteoporosis and implant failure: an exploratory case-control study. Journal of Periodontology 4, el Askary, A.S., Meffert, R.M. & Griffin, T. (1999) Why do dental implants fail? Part I. Implant Dentistry 8, Engquist B., Bergendal, T., Kallus, T. & Linden, U. (1988) A retrospective multicenter evaluation of osseointegrated implants supporting overdentures. International Journal of Oral and Maxillofacial Implants. 3, Esposito, M., Hirsch, J.M., Lekholm, U. & Thomsen, P. (1998) Biological factors contributing to failures of osseointegrated oral implants. (II). Etiopathogenesis. European Journal of Oral Sciences 3, Esposito, M., Thomsen, P., Ericson, L.E. & Lekholm U. (1999) Histopathologic observations on early oral implant failures. International Journal of Oral and Maxillofacial Implants 6, Friberg, B., Jemt, T. & Lekholm, U. (1991) Early failures in 4.,641 consecutively placed Brånemark dental implants, a study from stage 1 surgery to the connection of completed prostheses. International Journal of Oral and Maxillofacial Implants. 6, Friberg, B., Sennerby, L., Roos, J. & Lekholm, U. (1995) Identification of bone quality in conjunction with insertion of titanium implants. A pilot study in jaw autopsy specimens. Clinical Oral Implants Research. 6, Friberg, B., Sennerby, L., Meredith, N. & Lekholm, U. (1999) A comparison between cutting torque and resonance frequency measurements of maxillary implants. A 20-month clinical study. International Journal of Oral and Maxillofacial Surgery. 28, Glaser, D.L. & Kaplan, F.S. (1997) Osteoporosis. Definition and clinical presentation. Spine 22, Helkimo, E., Carlsson, G.E. & Helkimo, M. (1977). Bite force and state of dentition. Acta Odontol Scand 35, Holm, S. (1979). A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics, 6,

70 60 Ivanoff, C.J., Sennerby, L. & Lekholm, U. (1996) Influence of initial implant mobility on the integration of titanium implants. An experimental study in rabbits. Clinical Oral Implants Research 7, Jacobs, R. (2003) Preoperative radiologic planning of implant surgery in compromised patients. Periodontology , Jaffin R.A. & Berman C.L. (1991) The excessive loss of Brånemark fixtures in type IV bone, a 5-year analysis. Journal of Periodontology 62,2-4. Jisander, S., Grenthe, B. & Alberius, P. (1997) Dental implant survival in the irradiated jaw: a preliminary report International Journal of Oral & Maxillofacial Implants 12, Keller, E.E. (1997) Placement of dental implants in the irradiated mandible: a protocol without adjunctive hyperbaric oxygen. Journal of Oral & Maxillofacial Surgery 55, Lekholm, U. & Zarb, G. A. (1985) Patient selection and preparation. Tissue integrated prosthesis: Osseointegration in clinical dentistry. p.199, Chicago, Quintessence publishing Co Inc. Liang, K.Y. & Zeger, S.L. (1986). Longitudinal data analysis using generalized linear models. Biometrika, 73, Lindquist, L.W., Carlsson, G.E. & Jemt, T. (1996) A prospective 15-year follow-up study of mandibular fixed prostheses supported by osseointegrated implants. Clinical results and marginal bone loss. Clinical Oral Implants Research 7, Marx, R.E. & Johnson, R.P. (1987) Studies in the radiobiology of osteoradionecrosis and their clinical significance. Oral Surgery Oral Medicine and Oral Patholology 64, Mombelli, A. & Gionca, N. (2006) Systemic disease affecting osseointegration therapy. Clinical Oral Implant Research 17, Quirynen, M., De Soete, M. & van Steenberghe, D. (2002) Infectious risks for oral implants: a review of the literature. Clinical Oral Implants Research 13, Roberts, W.E., Simmons, K.E., Garetto, L.P., DeCastro, R.A. (1992) Bone physiology and metabolism in dental implantology: risk factors for osteoporosis and other metabolic bone diseases Implant Dentistry 1, Renouard, F. &, Nisand, D. (2006) Impact of implant length and diameter on survival rates Suppl Clinical Oral Implants Research 2, Szmukler-Moncler, S., Piattelli, A., Favero, G.A. & Dubruille, J.H. (2000) Considerations preliminary to the application of early and immediate loading protocols in dental implantology. Clinical Oral Implants Research 11, Tricio, J., Laohapand, P., van Steenberghe, D., Quirynen, M. & Naert, I. (1995) Mechanical state assessment of the implant-bone continuum, a better understanding of the Periotest method. International Journal of Oral and Maxillofacial Implants. 101, van Steenberghe, D., Vanherle, G.V., Fossion, E. & Roelens, J. (1976) Crohn's disease of the mouth, report of case. Journal of Oral Surgery 34, van Steenberghe, D, Lekholm, U, Bolender, C, Folmer, T, Henry, P, Herrmann, I., Higuchi, K., Laney, W., Linden, U. & Astrand, P. (1990) Applicability of

71 osseointegrated oral implants in the rehabilitation of partial edentulism: a prospective multicenter study on 558 fixtures. International Journal of Oral and Maxillofacial Implants 3, van Steenberghe, D., Yoshida, K., Papaioannou, W., Bollen, C., Reybroeck, G.I. & Quirynen, M. (1997) Complete nose coverage to prevent airborne contamination via nostrils is unnecessary. Clinical Oral Implants Research 8, van Steenberghe, D., Jacobs, R., Desnyder, M., Maffei, G. & Quirynen, M. (2002) The relative impact of local and endogenous patient-related factors on implant failure up to the abutment stage. Clinical Oral Implants Research 13, van Steenberghe, D., Quirynen, M., Molly, L. & Jacobs, R. (2003) Impact of systemic diseases and medication on osseointegration. Periodontolology , Wactawski-Wende, J., Grossi, S.G., Trevisan, M., Genco, R.J., Tezal, M., Dunford, R.G., Ho, A.W., Hausmann, E. & Hreshchyshyn, M.M. (1996) The role of osteopenia in oral bone loss and periodontal disease. Journal of Periodontology 67, Zeger, S.L. & Liang, K.Y. (1986). Longitudinal data analysis for discrete and continuous outcomes. Biometrics 42,

72

73 CHAPTER V. A biomechanical assessment of the relation between the oral implant stability at insertion and subjective bone quality assessment. V.1. Introduction: Several clinical reports on the use of oral implants mention that poor bone quality, as assessed on preoperative radiographs, lead to a less predictable outcome (Porter & von Fraunhofer 2005). While in well mineralized bone with proper degrees of corticalisation, like the symphyseal area a success rate of 99 % was reported even after 15 years with Brånemark system implants (Lindquist et al.1996), in distal areas of the upper jaw it can be substantially lower (Adell et al. 1990, Nevins & Fiorellini 1998). It thus seems relevant to develop measurements of the bone quality, especially referring to its mineral density, as a determinant for the primary stability of endosseous implants. It has been observed indeed that too large micromovements during the healing period can disrupt the bone apposition process on the implant surface and rather lead to fibrous scare tissue formation (Szmukler-Moncler et al. 1998). The assessment of the primary stability at insertion may be another option to determine the prognosis or to decide whether early or even immediate loading can be performed. The alternative is to let the bone to implant interface heal for a few months before being exposed to the oral environment as described in the original P-I Brånemark protocol (Brånemark et al. 1985) One available technique to determine the bone mineral content is to take biopsies of the jaws. This procedure is certainly reliable and safe, but does not seem practical in a routine clinical situation. The most popular current method of bone quality assessment is that developed by Lekholm & Zarb (1985), who introduced a scale of 1 to 4, based on both the radiographic assessment, and the sensation of resistance experienced by the surgeon when preparing 63

74 the fixture site. The grading refers to individual experience, and furthermore, it provides only a rough mean value of the entire jaw. Johansson & Strid (1994) described a technique whereby bone quality as a function of density and hardness could be derived from the torque forces needed during the implant insertion. They postulated that the energy used in tapping the site, prior to or during implant placement, is a combination of the thread placement force from the tip of the instrument and the friction created as the remaining part of a tap or implant enters the site. It has been demonstrated in an ex vivo human preparations that the cutting resistance during implant installation correlates well with the bone density as assessed by microradiography (Friberg et al. 1995) The absence of fixture mobility either indicative of a good primary stability or after a while of an intimate bone-to-implant contact can be objectively determine by an electronic measuring system, the Periotest (Gulden, AG, Bensheim, Germany) (Olivé & Aparicio 1990, Teerlinck et al. 1991, van Steenberghe & Quirynen 1993, van Steenberghe et al. 1995). This apparatus is widely used to assess implant outcome as can be seen from the hundreds of papers referring to it ( periotest.de). The PTV reveals the increased stiffness of the implant-bone continuum over time (Tricio et al. 1995). Implant stability can also be measured by resonance frequency analysis normally referred to as Implant Stability Quotient (ISQ) (Meredith 1998). The in vivo experimental findings demonstrate resonance frequency is related to implant stiffness in the surrounding tissues, which means a higher bone to implant contact percentage (Rasmusson et al. 1998). Clinically, the increase in implant stability has been measured using ISQ, and the increase in mobility were attributed to corticalization of the surrounding bone (Friberg et al. 1999a). The Osstell device (Mentor, Integration Diagnostics AB, Sävedalen, Sweden) has less documentation but also allows registration of minute changes in the rigidity of the bone-to-implant contact. The aim of the study was to evaluate the validity of subjective jaw bone quality assessment by comparing it to an objective parameter: the torque force needed to install implants, besides the primary stability of these implants measured either by ISQ or PTV, 64

75 or both, were also related to the subjective bone quality assessment when the measurement was available V.2. Materials and methods: A total of 298 consecutive patient files (198 females) were analyzed. They represent the total patient population treated by means of implants at the Department of Periodontology of the University Hospital of the Catholic University Leuven between November 2003 and June The mean age was 56.4 years (range: 18-86). All patients have been provided with a total of 761 Mark III TiUnite implants (Brånemark system, Nobel Biocare, Gothenburg, Sweden). At implant insertion, a minimal bone height of 7 mm had to be available. The classical surgical protocol with strict sterility measures as defined by Brånemark was used for all surgeries. Bone quality assessment was performed using the Lekholm and Zarb (1985) index. It consists of a scale of 1 to 4 (Figure 1). A copy of this grading system was available to the surgeon. The score was given immediately after implant placement. Figure 1: Grading system for bone quality assessment (Lekholm & Zarb 1985) Tactile sensation was assessed as such for both the cortical bone and the trabecular part during high speed drilling, as experienced by the surgeon when preparing the fixture site. For the latter a scale, ranging from grade 1 (very thick cortex/dense trabecular bone) to grade 3/4 (thin or very thin cortex/ poorly or very poorly mineralized trabecular bone), 65

76 was introduced (Table 1). Indeed the two last scores grouped because a distinction is laminar. Table 1: Tactile evaluation of the cortical and trabecular bone during surgery Grade 1 Grade 2 Grade 3, 4 Cortical bone thick moderate (very) thin Trabecular bone dense moderate (very) poor Besides, the bone quality was assessed objectively during implant insertion, by means of an electronic torque force measurement device, which is part of a controlled motor device. The latter measures the torque force while tapping or inserting the implant at slow speed (OsseoCare, Nobel Biocare, Gothenburg, Sweden). The OsseoCare motor was developed to insert the implant into the (pre-tapped) bone site with a well controlled insertion torque of 20, 30, 40 or 50 Ncm. (Figure 2: a, b). The software controls and registers the operation of the hand-piece micro-motor, and monitors the torque delivered, as well as the number of turns performed. The software records the cutting torque resistance as mean values (Ncm) at the crestal third, the middle third, and the apical third of each implant insertion trajectory. 66

77 a b Figure 2 a: OsseoCare Unit. The screen shows a graph like Fig. 2b, b: The OsseoCare software curve of the placement torque (Ncm) in the 1 st, 2 d and 3 d third during implant placement The rigidity of the implant-bone continuum was assessed by the resonance frequency analysis method (Osstell Mentor, Integration Diagnostics AB, Gamlestadsvägen 3B, SE , Sweden) (Figure 3; a, b, c). These measurements were done at implant insertion as well as just prior to the abutment insertion (after submerged healing). The RFA technique analyses the resonance frequency (range: Hz) of a peg (Smartpeg, Integration Diagnostics AB, Gamlestadsvägen 3B, SE , Sweden), which can be 67

78 attached to the fixture by the aid of a mount, 4-5 Ncm of torque is enough. Subsequently the probe is hold close to the peg in a vestibular-oral and in a mesio-distal direction during the pulsing time. After the processing time the ISQ value is presented on the display. The resonance frequency values are automatically converted into an arbitrary index called the Implant Stability Quotient, the ISQ, which runs from 1 to 100; the higher the ISQ, the more stable the implant. This index facilitates clinical evaluation (Meredith 1994, Meredith et al. 1996). The device was only available at implant placement for the last 141 patients. Unhappily, because of technical problems encountered at the beginning, measurements could only be made on 71 patients provided with a total of 153 implants. a b c Figure 3 a,b: Fixation of the magnetic pin on the implant, c: Osstell Mentor; stimulation and recording of the resonance frequency of the peg. The rigidity of implant-bone continuum was also recorded by means of a Periotest device (Gulden, A G, Bensheim, Germany) after connecting a temporary abutment (Ceka, Alphadent, nv, Antwerp, Belgium) of 4 mm length. Because of time pressure in 68

79 the OR and/or patient related factors this procedure was only performed in a sub group of 22 patients provided with a total of 44 implants. These Periotest values (PTV) were also recorded at abutment surgery. This device measures the damping characteristics of the implant-bone continuum. It consists of a hand-piece connected to a unit that analyses the braking time of the rod projected onto a surface (Tricio et al. 1995). The rod of the device is placed perpendicular to the abutment at a distance of 2 mm. Then the rod is accelerated electromagnetically. When the rod hits the implant, it is decelerated. The faster the deceleration, the greater the implant stability is in the bone tissue. The values were only accepted when two consecutive measurements did not deviate more than one unit from each other. The arbitrary values can range from 8 (very stable) to +50 (extremely mobile) (Figure 4). Although not useful to assess the biomechanics of teeth it appeared that the Periotest was very useful for the assessment of implant stability (Olivé & Aparicio1990, Teerlinck et al. 1991, van Steenberghe et al. 1995). a b Figure 4 a: The Periotest device with a digital display and the microphone for the artificial voice, b: The rod hits the abutment after acceleration. 69

80 Statistical analysis: Data were statistically analyzed by means of SAS software version 9.1 for Windows. Pearson correlation coefficients were calculated using PROC MIXED fitting a bivariate model. In order to statistically assess mean differences, a linear model was fitted in PROC MIXED with the corresponding response value, placement torque values in the crestal, middle and apical third separately, ISQ and PTV and covariates bone quality according to Lekholm and Zarb index (1985) and bone quality as assessed by the surgeon s tactile sensation for the cortical and trabecular bone. Multiple testing corrections by Tukey procedure for pair wise differences when applicable was used. The ISQ values at implant insertion were dichotomized (cut-off =60). Based on these a comparison for the placement torque measurements within each region separately was performed. The p-value was set to 0.05 to detect the level of significance. V.3. Results: From a total of 761 implants installed in the 298 patients, the placement torque measurement in the crestal, the second and the apical third were recorded for 720 implants installed in 288 patients, and compared to implant position in the jaw. A significant difference was detected for placement torque measurement between anterior and posterior locations (p-value < 0.01). The missing data are due to inadvertent erasing of the Osseocare data or to technical problems with the machinery (Table 2). For one implant, bone quality was not assessed because of the se of bone-filling material, and therefore the placement torque for 719 implants in 288 patients was measured and compared with the bone quality assessment according to Lekholm & Zarb index (1985). A significant relationship was detected between placement torque measurements and the Lekholm & Zarb index (p-value< ) (Table 3). 70

81 Table 2: Frequency distribution of 720 implants in the upper and lower jaws and the corresponding placement torque measurements. *A significant difference was detected for placement torque measurement between upper and the lower jaw (p-values < ). Implant position Crestal third (Ncm) Middle third (Ncm) Apical third (Ncm) N Implants Mean SD Mean SD Mean SD L Ant 157 4,52 2,87 10,15 5,66 15,69 6,96 U Ant 185 4,46 2,51 8,52 4,39 12,25 5,21 L Post 177 4,05 2,24 9,41 5,50 14,64 6,30 U Post 201 3,86 1,76 7,41 3,77 11,30 5,22 LJ 334 4,27 2,56 9,76* 5,58 15,13* 6,63 UJ 386 4,15 2,17 7,95 4,11 11,75 5,23 Total 720 4,21 2,36 8,79 4,93 13,32 6,15 Table 3: Placement torque measurements vs. bone quality assessment grades according to Lekholm and Zarb (1985). A significant relationship was found between the Lekholm & Zarb index and the placement torque measurements (p < ). * A significant difference was detected between the grades (p-value < ). N implants Grade Grade Crestal third (Ncm) Middle third (Ncm) Apical third (Ncm) 4,22 9,58 15,21 4,67 10,03 14,85 Grade ,76 * 7,41 * * 11,39 * Grade ,28 5,49 8,38 71

82 The placement torque measurements of 705 implants were compared with the cortical bone thickness as assessed by the surgeon on basis of his tactile grading. The very few missing data are related to for example placement of implant at the time of tooth extraction, the presence of filling material or accidentally deleted data. For a total of 713 implants the placement torque values were compared with the trabecular bone density as assessed by the surgeon on the basis of his tactile grading. Again the very few missing data are related to factors as mentioned above (Table 4). Table 4: Placement torque measurements related to the grades of bone quality assessment according to the surgeon s tactile sensation. * A significant difference was detected between the grades (p-value < ). N Implants Crestal third (Ncm) Middle third (Ncm) Apical third (Ncm) Cortical bone Trabecular bone Thick (grade 1) 323 4,22 9,70 * 15,06 Moderate (grade 2) Very thin (grade 3/4) 66 3,58 6,85 10,06 Dense (grade 1) 295 4,38 9,71 14,83 Moderate (grade 2) * 8.60 * Poor (grade 3/4) * * 72

83 Comparisons were done between ISQ measurements at implant insertion and at abutment connection for those implants (53) where both measurements were done. The same was done for PTV (17) (Table 5). Table 5: Frequency distribution of implants in the upper and lower jaws and the corresponding PTV and ISQ values at implant insertion and at abutment connection for the same implants. * A significant difference was found between PTV values at implant insertion and at abutment connection (p-value < 0.05), and between ISQ values at implant insertion and at abutment connection as well (p-value < Mean N implants At implant insertion At abutment connection PTV UJ 11-1,00-3,27 LJ 6-5,00-3,50 Total 17-2,41-3,35* ISQ UJ 36 67,78 72,00 LJ 17 72,24 69,53 Total 53 69,21 71,21* To evaluate the relation between two objective assessments of bone quality, i.e. the insertion torque and ISQ measurements, a correlation was calculated on part of the data for which these measurements were available. For a total of 136 implants the insertion torque was measured as well as the ISQ values during surgery (Figure 5). The correlation coefficient between these two variables was calculated. The estimated correlation equals ρ=0.20 (SE=0.08). This coefficient is significantly different (p-value=0.01). From the latter the placement torque measurement correspondent to ISQ < 60 was compared to the placement torque measurement 60 (Table 6). 73

84 Figure 5: Linear regression for placement torque at the apical third (Ncm) and ISQ values at implant insertion. Table 6: The mean of placement torque measurements and corresponding ISQ values with a cut-off at 60 * (p-value = 0.05) Torque Crestal third Middle third Apical third measurements (Ncm) (Ncm) (Ncm) Correspondent to ISQ < 60 Correspondent to * ISQ ISQ and PTV were also compared to the bone quality assessed according to the Lekholm and Zarb index. A significant relationship was detected (p-value =0.01) (Table 7 and 8). 74

85 Table 7: ISQ of 146 implants compared with bone quality assessment according to Lekholm and Zarb index. * A significant difference was detected between the grades (p-value < 0.02) N implants ISQ Grade ,29 Grade ,61 Grade ,15 Grade ,00 * Table 8: PTV values vs. the grades of bone quality assessment according to Lekholm and Zarb for 44 implants. * A significant difference was detected between the grades (p-value < 0.05) N implants PTV Grade Grade ,74 Grade ,61 Grade 4 0 / * Furthermore, ISQ and PTV recorded at implant insertion were also compared to the bone quality assessed according to the surgeon s tactile sensation. A significant relationship was detected between ISQ, PTV and cortical bone grades (p-value =0.02, < respectively), and between ISQ and trabecular bone grades (p-value = 0.01) (Table 9 and 10). Table 9: ISQ of 146 implants compared with bone quality assessment according to surgeon tactile sensation assessment. * A significant difference between grades (p-value<0.05) N implants ISQ Thick (grade 1) 64 70,27 Cortical Moderate (grade 2) bone Thin/very thin (grade 3/4) Trabecular Dense (grade 1) 71 69,42 bone Moderate (grade 2) Poor /very poor (grade 3/4) 23 66,43 * 75

86 Table 10: PTV values vs. the grades of bone quality assessment according to the surgeon tactile sensation assessment for 44 implants. * A significant difference was detected between the grades N implants PTV Thick (grade 1) Cortical Moderate (grade 2) bone Thin/very thin (grade 3/4) Trabecular Dense (grade 1) bone Moderate (grade 2) Poor /very poor (grade 3/4) * V.4. Discussion: Subjective assessments seem to have a limited value when trying to discriminate among bone qualities. The present data indicate that, definitely for the extreme categories (1 and 4 of the Lekholm & Zarb index) the relationship with objective parameters is good. Especially today, where early or immediate loading of endosseous implants is more and more considered, these biomechanical parameters may help the clinician to decide if such deviation of the classical osseointegration protocol can be considered or should be delayed. Indeed, too large micromovements of endossesous implants can lead to fibrous encapsulation rather than bone apposition (Szmukler-Moncler et al. 1998) and this risk increases with a lower degree of bone density. Although biomechanical assessments were only performed on a fraction of the patients for a variety of reasons, these data substantiate the main findings and open new perspectives. The workload and the medical considerations or technical reasons sometimes lead to the need to pursue only the patient treatment, and prevented the data registration. The insertion torque measurements were higher in the lower jaw, especially the symphyseal area, when compared to the upper jaw. The posterior region of the upper jaw has the lowest torque value, which is in agreement with a previous study (Friberg et al. 1999a). In the posterior maxilla, there is indeed frequently a (very) thin cortical bone 76

87 combined with less dense trabecular bone (Jacobs 2003). Thus, clinicians generally observe a poor degree of bone mineralization on the radiographs and a limited bone resistance while drilling in this area (Friberg et al. 1995, 1999b). Johansson et al (2004) also found that cutting torque values correlated with the Lekholm and Zarb index of bone quality. Homolka et al (2002) found a significant correlation between bone mineral density measurements and the insertion torque measurements in cadaver mandibles. A number of studies indicated that the failure rate is greater in category of quality (4 ) bone according to Lekholm and Zarb classification (Engquist et al. 1988, Jaffin & Berman 1991, Friberg et al. 1991). Implant stability means indeed a higher bone to implant contact percentage (Rasmusson et al. 1998), which can explain the better prognosis. In literature the ISQ readings obtained during early phases of osseointegration revealed higher implant stability in the mandible compared to the maxilla (Ersanli et al. 2005). It is striking that this difference seems to decrease in the present study (see table 5) during the osseointegration process. It may indicate that a better marrow content in the upper jaw could speed up the bone apposition. Miyamoto et al (2005) found a significant correlation between ISQ and thickness of cortical bone. Nkenke et al (2003) in a human cadaver study found that resonance frequency analysis values did correlate with the surface of bone-to-implant contact, and with the height of the crestal cortical bone penetrated by the implants in the oral aspects of the implant sites. Ostman et al (2006) found a significant correlation between bone quality and ISQ values, which is in accordance with the present study. Friberg et al (1999b) found a significant correlation between insertion torque and RFA measurements. Previous reports indicate that the mean Periotest values obtained for Brånemark system implants placed in the maxilla were higher than for the mandible, indicating less stability, (Olivé & Aparicio 1990, van Steenberghe et al. 1995). The same applies for ITI implants (Buser et al. 1990) and TPS (Salonen et al. 1993). Bone quality assessment according to Lekholm and Zarb (1985) in the present study could be related to insertion torque measurements, ISQ and PTV. 77

88 Quirynen et al observed that the PTV value of an implant was dominated by the cortical/crestal bone. This is illustrated by peri-apical lesions around implants where PTV values remain low although much of the trabecular bone contact has disappeared. The Periotest showed a correlation with the crestal cortical bone penetrated by the implants in the buccal aspect of the implant site. Previously, van Steenberghe et al. (1995) showed that PTV values were lower for implants with a bicortical vs. a monocortical contact. V.5. Conclusions: The present clinical data illustrate that several objective measurement devices are available to assess bone-to-implant contact and primary or early stability. These measurements seem to be related to the categories of the Lekholm & Zarb index, which subjectively assesses bone quality on the basis of radiological aspects and surgeon s tactile sensation. 78

89 V.5. References: Adell, R., Eriksson, B., Lekholm, U., Brånemark, PI. & Jemt, T. (1990) Longterm follow-up study of osseointegrated implants in the treatment of totally edentulous jaws. International Journal of Oral and Maxillofacial Implants. 5, Brånemark, P. I., Zarb, G.A. & Albrektsson, T. (1985) Tissue integrated prostheses. Osseointegration in clinical dentistry volume 13 p. 211 Chicago, Quintessence Publishing Co Inc. Buser, D., Weber, HP. & Lang, N. P. (1990) Tissue integration of non-submerged implants. 1-year results of a prospective study with 100 ITI hollow-cylinder and hollow-screw implants. Clinical Oral Implants Research. 1, Engquist B., Bergendal, T., Kallus, T. & Linden, U. (1988) A retrospective multicenter evaluation of osseointegrated implants supporting overdentures. International Journal of Oral and Maxillofacial Implants. 3, Ersanli, S., Karabuda, C., Beck, F. & Leblebicioglu, B. (2005) Resonance frequency analysis of one-stage dental implant stability during the osseointegration period. Journal of Periodontology. 76, Friberg, B., Jemt, T. & Lekholm, U. (1991) Early failures in 4.,641 consecutively placed Brånemark dental implants, a study from stage 1 surgery to the connection of completed prostheses. International Journal of Oral and Maxillofacial Implants. 6, Friberg, B., Sennerby, L., Roos, J. & Lekholm, U. (1995) Identification of bone quality in conjunction with insertion of titanium implants. A pilot study in jaw autopsy specimens. Clinical Oral Implants Research. 6, Friberg, B., Sennerby, L., Gröndahl, K., Bergstrom, C., Back, T. & Lekholm, U. (1999a) On cutting torque measurements during implant placement, a 3-year clinical prospective study. Clinical Implant Dentistry and Related Research. 1, Friberg, B., Sennerby, L., Meredith, N. & Lekholm, U. (1999b) A comparison between cutting torque and resonance frequency measurements of maxillary implants. A 20-month clinical study. International Journal of Oral and Maxillofacial Surgery. 28, Homolka P., Beer A., Birkfellner W., Nowotny R., Gahleitner A., Tschabitscher M. & Bergmann H (2002) Bone mineral density measurement with dental quantitative CT prior to dental implant placement in cadaver mandibles: pilot study. Radiology. 224, Jacobs, R. (2003) Preoperative radiologic planning of implant surgery in compromised patients. Periodontology , Jaffin R.A. & Berman C.L. (1991) The excessive loss of Brånemark fixtures in type IV bone, a 5-year analysis. Journal of Periodontology. 62,2-4. Johansson, P. & Strid, K. G. (1994) Assessment of bone quality from placement resistance during implant surgery. International Journal of Oral and Maxillofacial Implants. 9,

90 80 Johansson, B. Back, T. & Hirsch, J.M. (2004) Cutting torque measurements in conjunction with implant placement in grafted and nongrafted maxillas as an objective evaluation of bone density: a possible method for identifying early implant failures? Clinical Implant Dentistry and Related Research. 6,9-15. Lekholm, U. & Zarb, G. A. (1985) Patient selection and preparation. Tissue integrated prosthesis: Osseointegration in clinical dentistry. p.199, Chicago, Quintessence publishing Co Inc. Lindquist, LW., Carlsson, GE. & Jemt, T. (1996) A prospective 15-year follow-up study of mandibular fixed prostheses supported by osseointegrated implants. Clinical results and marginal bone loss. Clinical Oral Implants Research 7, Meredith, N. J. (1994) The application of modal vibration analysis to study bone healing in vivo. Journal of Dental Research 73,793. Meredith, N., Alleyne, D. & Cawley, P. (1996) Quantitative determination of the stability of the implant-tissue interface using Resonance Frequency Analysis. Clinical Oral Implants Research. 7, Meredith, N. (1998) A review of nondestructive test methods and their application to measure the stability and osseointegration of bone anchored endosseous implants. Critical Reviews in Biomedical Engineering. 26, Miyamoto, I., Tsuboi, Y., Wada, E., Suwa, H. & Iizuka, T. (2005) Influence of cortical bone thickness and implant length on implant stability at the time of surgery-clinical, prospective, biomechanical, and imaging study. Bone. 37, Nevins, M. & Fiorellini, J. P. (1998) Placement of the maxillary posterior implants. In, Nevines M., Mellonig IT., editors. Implant therapy., clinical approaches and evidence of success, volume 2, p.153 Chicago, Quintessence. Nkenke E., Hahn, M., Weinzierl, K., Radespiel-Troger, M., Neukam, F. W.& Engelke, K. (2003) Implant stability and histomorphometry, a correlation study in human cadavers using stepped cylinder implants. Clinical Oral Implants Research. 14, Olivé, J. & Aparicio, C. (1990) Periotest method as a measure of osseointegrated oral implant stability. International Journal of Oral and Maxillofacial Implants. 4, Ostman PO, Hellman M, Wendelhag I, Sennerby L. (2006) Resonance frequency analysis measurements of implants at placement surgery. The International Journal of Prosthodontics.19,77-83 Porter, J. A. & von Fraunhofer, J. A. (2005) Success or failure of dental implants? A literature review with treatment considerations. General Dentistry. 6, Quirynen, M., Vogels, R., Alsaadi, G., Naert, I. & van Steenberghe, D. (2005) Predisposing conditions for retrograde peri-implantitis, and suggestions for their treatment. Clinical Oral Implants Research. 16: Rasmusson, L., Meredith, N., Kahnberg, K. E. & Sennerby, L. (1998) Stability assessments and histology of titanium implants placed simultaneously with autogenous onlay bone in the rabbit tibia. International Journal of Oral and Maxillofacial Surgery. 27,

91 Salonen, M. A., Oikarinen, K., Virtanen, K. & Pernu, H. (1993) Failures in the osseointegration of endosseous implants. International Journal of Oral and Maxillofacial Implants. 1, Szmukler-Moncler, S., Salama, H., Reingewirtz, Y.& Dubruille, J. H. (1998) Timing of loading and effect of micromotion on bone-dental implant interface, review of experimental literature. Journal of Biomedical Materials Research. 43, Teerlinck, J., Quirynen, M., Darius, P. & van Steenberghe, D. (1991) Periotest, An objective clinical diagnosis of bone apposition toward implants. International Journal of Oral and Maxillofacial Implants. 1, Tricio, J., Laohapand, P., van Steenberghe, D., Quirynen, M. & Naert, I. (1995) Mechanical state assessment of the implant-bone continuum, a better understanding of the Periotest method. International Journal of Oral and Maxillofacial Implants. 101, van Steenberghe, D. & Quirynen, M. (1993) Reproducibility and detection threshold of peri-implant diagnostics. Advances in Dental Research. 7, van Steenberghe, D., Tricio, J., Naert, I. & Nys, M. (1995) Damping charachteristics of bone-to-implant interfaces. A clinical study with Periotest device. Clinical Oral Implants Research. 1,

92

93 Part 2. IMPACT OF IMPLANT SURFACE CHARACTERISTICS ON IMPLANT FAILURES

94

95 CHAPTER VI. The importance of surface characteristics of implants replacing failed ones VI.1. Introduction: The use of oral endosseous implants to retain or to support a dental prosthesis is a well established clinical procedure if based on the osseointegration principle. The biologic principle was clinically introduced by P-I Brånemark approximately four decades ago (Brånemark et al. 1969). High success rates were reported in studies on consecutive implants, both in full and partial edentulism and in the upper and lower jaw. Lindquist et al (1996) reported a cumulative success rate of 98.9% for the Brånemark system after 15 years for edentulous patients provided with mandibular fixed prostheses. The same patient group eventually included implants successful for more than 20 years in function, with cumulative survival rate of 98.9% (Ekelund et al. 2003). Similar results were reported for implants retaining an overdenture, again in the symphyseal area (van Steenberghe et al. 2001). The rehabilitation of partially edentulous jaws seemed originally less successful in a multicenter retrospective study (van Steenberghe 1989). Later, it appeared that a learning curve played a certain role since in a 10-year prospective multicenter study cumulative success rates of 90.2% and 93.7% were reached for respectively the upper and lower jaws (Lekholm et al. 1999). All these successes were achieved with commercially pure screw-shaped implants with a machined surface (also called a turned surface). Recently, many implant surfaces have been developed using various reducing techniques such as blasting with aluminum oxide particles, grit-blasting with titanium dioxide particles, sandblasting and acidetching, and acid-etching alone (Cochran et al. 2002, Ibanez et al. 2002, Albrektsson et al. 2004). Increased oxidation of the implant surface has also been proposed (Hall & Lausmaa. 2000). The TiUnite implant (Brånemark system, Nobel Biocare, 85

96 Gothenburg, Sweden) surface is created by anodic oxidation. These modified surfaces have proven to enhance and speed bone apposition (Carlsson et al. 1988, Buser et al. 1991, Albrektsson et al. 2000, Henry et al. 2000, Ivanoff et al. 2003). Early occlusal load may be imposed on implants with such modified surfaces, since bone apposition takes place at a faster rate (Rocci et al. 2003, Calandriello et al. 2003, Olsson et al. 2003). Modified or so-called improved surfaces could also lead to an increased success rate in patients or locations that do not offer optimal bone quality and quantity. For optimal situations, such as symphyseal areas, where success rate close to 100 % have already been achieved the need for improved surfaces can be questioned (Ekelund et al. 2003). Another development over the years has been the use of less elaborate surgical approaches, since clinicians with variable skills and training are performing implant placement surgeries. Because today s implants are not being placed by a small pool of highly experienced clinicians, as the early experimental implants were, the variation in clinician skill and experience may need to be compensated for by an improved surface to obtain similarly predictable osseintegration. Aseptic surgery has been advocated since the early days (Brånemark et al. 1985). The use of a nose guard to prevent contact of sterile gloves with the highly contaminated nasal skin is highly recommended (van Steenberghe et al. 1997). Today these aspects are often overlooked since surgery is today performed in even a non-surgical set-up. Furthermore, even when a less precise drilling trajectory is used or when drilling is performed without coolant, predictable bone apposition can still be achieved with implants with improved surface that leads to a more intense osteoblastic reaction. Lazzara et al. (1999) demonstrated histomorphometrically a significantly higher bone implant contact (BIC) rate after 6 months in the posterior maxilla of humans when a double acid-etched surface was used rather than a machined surface (73% versus 34%). In the present study, implants with an increased oxide layer (TiUnite ) were compared with machined-surface implants by comparing the success rates of TiUnite and machined implants used to replace failed implants. The replacements implants were thus exposed to the same patient-related risk factors as the failed implants were. 86

97 VI.2. Materials and methods: The files of 578 patients ( 326 female and 252 male), all treated during 3 consecutive years by means of oral implants at the Departement of Periodontology of the University Hospital of the Catholic University Leuven were evaluated. The observation time ranged from 9 to 49 months. All patients had been provided with Brånemark system implants (Nobel Biocare AB, Gothenburg, Sweden). Within this group two implant surfaces were used: a machined also called turned surface and an improved surface by increased oxidation (TiUnite ). The geometry of the machined implants was either standard (Mark I), Mark II or Mark III. In the Brånemark system, standard implants are used after pre-tapping the drilled alveolus, while Mark II and III are self-tapping screw-shaped implants. All are made of commercially pure titanium. The distribution of the implant types with their surface characteristics can be found in Table 1. For each patient the treatment history in file has been carefully analyzed. In cases where questions remained, the treating staff member was asked for further information. At implant placement, a minimal bone height of 7 mm had to be available. The same surgical protocol with strict sterility measures were used in all surgeries, also during the replacement surgeries. A thorough departmental sterility policy allows the use of antibiotics to be limited to well-defined indications, e.g. endocarditis prophylaxis, remaining infection at the site of surgery, or coughing or sneezing by the patient during surgery. Data collection and analysis focused solely on the replacement implants which replaced non-osseointegrated, i.e. failed implants. The failure rate of the original implants was calculated, as well as concomitant health or behavioral factors of the patients involved. Smoking habits, osteoporosis, hypo- or hyperthyroid states, and intake of antidepressant or steroid medication were of particular interest (van Steenberghe et al. 2003). Smoking patients were allocated to two categories: 1 to10 cigarettes per day and > 10 cigarettes per day). 87

98 Statistical Analysis: Data were statistically analyzed by means of the standard Statistica for Windows software version 5.1. (Statsoft, Tulsa, Oklahoma, USA); a Fisher exact P test was used. The p- value was set 0.05 to detect a level of significance. VI.3. Results: A total of 41 patients (18 female, 23 male), aged 24 to 84 years, experienced nonintegration and had to receive new implants to replace the failed ones. Some patients needed to have more than one implant replaced. Implant replacement was carried out 4 and 6 months after implant removal. Twenty-nine implants with a machined surface were replaced by machined-surface implants. Of the 29 replacement implants, 6 failed (5 within one year and 1 after two years of insertion). Nineteen implants with a machined surface were replaced by implants with a TiUnite surface. Of the 19 replacement implants, 1 implant failed. Ten implants with a TiUnite surface have been replaced by implants with the same surface. None of these have failed. The distribution of implant lengths, diameters and locations in the replacement group, and frequency of implant failure are shown in tables 1a to 1c The failure rate of the machined-surface replacement implants was significantly higher compared with that of the TiUnite replacement implants (p-value = 0.05) (Table 2). 88

99 Table 1a: Distribution of replacement implant lengths vs. previous implant lengths in the replacement implant group, and incidence of implant failures. No. of replacement implants No. of replacement implants that failed. Length of replacement implant vs. previous implant Equal Greater than Less than Table 1b: Distribution of replacement implant diameters vs. previous implant diameters and incidence of replacement implant failures. No. of replacement implants No. of replacement implants that failed. Diameter of replacement implant vs. previous implant Equal Greater than Less than Table 1c: Distribution of replacement implant location and incidence of replacement implant failure. Location No. of replacement implants No. of replacement implants that failed. Maxilla Mandible Anterior Posterior Anterior Posterior

100 Table 2: Distribution of the replacement implants, and frequency of failures, the difference in failure rate between machined-surface replacement implants and Ti-Unite implants was statistically significant (* p= 0.05). Replaced by machined surface implants (n =29) (second failure = 6*) No. of implant s replaced Failed implants Type New implants No. of replaceme nt implants that failed Time implant observed (mo) Mean Range No. of replacement implants that failed Time of failure of replacement implants (mo post replacement) 3 Mk I Mk I Mk II Mk I Mk I Mk II / 32 / 2 Mk II Mk II Mk III Mk III Mk II Mk III / / 4 Mk I Mk III / / Replaced by 5 Mk I TiUnite / / with TiUnite 4 Mk II TiUnite / / implants (n =29) (second failure = 1*) Mk III TiUnite TiUnite TiUnite 1 / / 12 Out of 326 female patients, 32 patients (9.8%) were smokers. Twenty smoked more than 10 cigarettes /day. Among the 252 male patients, 51 (20.2%) patients were smokers. Forty-three smoked more than 10 cigarettes /day. In the replacement group, 2 of 18 female patients were smokers (1 more than10 cigarettes /day) and 10 of 23 males were smokers (all more than 10 cigarettes /day) (Table 3). 90

101 Table 3: Distribution of smokers and non-smokers in the study population and in replacement group. (the daily consumption of cigarettes is indicated). Study population (n = 578) Replacement group (n = 41) Non smokers Smokers Female Male Total Female 10 /day 12 1 > /day Total 32 2 Male 10 /day 8 0 > /day Total Twenty-nine out of 578 patients had a known osteoporosis, compared to 2 of 41 patients in the replacement group. The numbers are too small to indicate a tendency for a higher incidence of osteoporosis in the replacement group. The respective frequency of other relevant diseases can be found in Table 4. Of the 58 implants in the replacement group, 33 were placed in patient who did not receive antibiotics at the replacement surgery. Four of these implants were failed. The remaining 25 implants were replaced in patients who received antibiotics immediately prior to and 1 or more days after surgery; 3 of these implants failed. 91

102 Table 4: Distribution of some systemic diseases in the total patient population and in the replacement group. Study population (n = 578) Replacement group (n = 41) Female Male Female Male Total Osteoporosis Hyperthyroid Hypothyroid Medication Antidepressant Steroid VI.4. Discussion: The experimental hypothesis appeared to be confirmed. When failures of a machined-surfaced implant occurred, replacement at the same site by a TiUnite implant surface with similar geometry led to a higher success rate. This was not the case when a failed implant was replaced by another machined-surfaced implant. Randomization was not applied in the present study; the replacement of the failed implant by either a TiUnite or a machined-surface implant was forced choice. Indeed, the TiUnite surface was only used in the department at a certain date which fell in the middle of the 3 years period. Thus any bias could be excluded. An implant newly placed at a site where an implant previously failed is again subjected to the same systemic and local compromising factors. Thus a comparison between the 2 groups leads to the identification of more or less the only significant variable, namely the implant surface. The other uncontrolled variable is that, at replacement, one is confronted with a non-pristine bone site, where latent inflammation or scar tissue can remain since the previous surgical intervention. This probably explains the higher failure rate for implant with a machined surface used for replacing failed ones. It has been reported that the Brånemark system implants with a TiUnite surface, experience a faster bone apposition which allow it to achieve a proper fixation even if a remaining endosseous lesion tends to compromise the osseointegration process 92

103 (Quirynen et al. 2005). Thus while this would lead to early loss (ie, before or at abutment surgery) of the implant with a machined surface, with a TiUnite surface integration is already achieved in the coronal parts before the apical inflammatory process can compromise the ongoing bone apposition. This difference in bone apposition may be very relevant for replacing implants. The impact of smoking habits on the outcome of osseointegration is again evidenced in the present study, as the number of smokers in the replacement group is high comparing with the total patient population. This is in agreement with previous studies (for review Bain 2003). The other systemic factors or medications (van Steenberghe et al. 2003), although known to play an important role, could not be properly analyzed since the incidence frequencies were too small. Often, so-called sterile surgery is not in fact sterile; although sterile drapes and gowns may be used, the chain of sterility is often breached by lack of proper surgical training. The small incidence of failures which is characteristic for the presently used implant system renders statistical analyses of success rates difficult. Nevertheless, the present analysis of the fate of implants inserted at the same site where an implant recently failed allowed the improved surface to be assessed as having higher predictability. VI.5. Conclusion: An improved implant surface such as TiUnite may offer a better prognosis when a failed implant has to be replaced. 93

104 VI.6. References: Albrektsson, T., Johansson, C., Lundgren, A., Su, Y. & Gottlow, J. (2000) Experimental studies on oxidized implants. A histomorphometrical and biochemical analysis. Applied Osseointegration Research 1, Albrektsson, T. & Wennerberg, A. (2004) Oral implant surfaces: Part 2 review focusing on clinical knowledge of different surfaces. International Journal of Prosthodotics 17, Bain, C.A. (2003) Implant installation in the smoking patients. Periodontolology , Brånemark, P-I., Adell, R., Breine, U., Hansson, B.O., Lindström, J. & Ohlsson, Ǻ (1969). Inta-osseos anchorage of dental prostheses. I. Experimental studies. Scandinavian Journal of Plastic and Reconstructive Surgery 3, Brånemark, P-I., Zarb, G.A. & Albrektsson, T. (eds) (1985). Tissue integrated prostheses: Osseointegration in clinical dentistry. p , Chicago, Quintessence publishing Co Inc. Buser, D., Schenk, R., Steinemann, S., Fiorellini, J., Fox, C. & Stich, H. Influence of surface characteristics on bone integration of titanium implants (1991) A histomorphometric study in miniature pigs. Journal of Biomedical Materials Research 25, Calandriello, R., Tomatis, M., Vallone, R., Rangert, B. & Gottlow, J. (2003) Immediate occlusal loading of single lower molars using Brånemark System Wide-Platform TiUnite implants: an interim report of a prospective openended clinical multicenter study. Clinical Implant Dentistry and Related Research 1, Carlsson, L., Rostlund, T., Albrektsson, B. & Albrektsson, T. (1988) Removal torques for polished and rough titanium implants. International Journal of Oral & Maxillofacial Implants 3, Cochran, D.L., Buser, D., ten Bruggenkate, C.M., Weingart, D., Taylor, T.M., Bernard, J.P., Peters, F. & Simpson, J.P. (2002) The use of reduced healing times on ITI implants with a sandblasted and acid-etched (SLA) surface: early results from clinical trials on ITI SLA implants. Clinical Oral Implants Research 13, Ekelund, J.A., Lindquist, L.W., Carlsson, G.E. & Jemt, T. (2003) Implant treatment in the edentulous mandible: a prospective study on Brånemark system implants over more than 20 years. International Journal of Prosthodontics 16, Hall, J., & Lausmaa, J. (2000) Properties of a new porous oxide surface on titanium implants. Applied Osseointegration Research 1,5-8. Henry, P., Tan, A., Allan, B., Hall, J. & Johansson, C. (2000) Removal torque comparison of TiUnite and turned implants in the greyhound dog mandible. Applied Osseointegration Research 1,

105 Ibanez, J.C., & Jalbout, Z.N. (2002) Immediate loading of osseotite implants: two-year results. Implant Dentistry 11, Ivanoff, C., Widmark, G., Johansson, C. & Wennerberg, A. (2003) Histologic evaluation of bone response to oxidized and turned titanium micro-implants in human jawbone. International Journal of Oral & Maxillofacial Implants 18, Lazzara, R.J., Testori, T., Trisi, P., Porter, S.S. & Weinstein, R.L. (1999) A human histologic analysis of osseotite and machined surfaces using implants with 2 opposing surfaces. International Journal of Periodontics & Restorative Dentistry 19, Lekholm, U., Gunn, J., Henry, P., Higuchi, K., Linden, U., Bergstrom, C. & van Steenberghe, D. (1999) Survival of the Brånemark implant in partially edentulous jaws: a 10-year prospective multicenter study. International Journal of Oral &Maxillofacial Implants 14, Lindquist, L.W., Carlsson, G.E. & Jemt, T. (1996) A prospective 15-year follow-up study of mandibular fixed prostheses supported by osseointegrated implants. Clinical results and marginal bone loss. Clinical Oral Implants Research 7, Olsson, M., Urde, G., Andersen, J.B. & Sennerby, L. (2003) Early loading of maxillary fixed cross-arch dental prostheses supported by six or eight oxidized titanium implants: results after 1 year of loading, case series. Clinical Implant Dentistry & Related Research 1, Quirynen, M., Vogels, R., Alsaadi, G., Naert, I. & van Steenberghe, D. (2005) Predisposing conditions for retrograde peri-implantitis, and suggestions for their treatment. Clinical Oral Implant Research 16, Rocci, A., Martignoni, M. & Gottlow, J. (2003) Immediate loading of Brånemark System TiUnite and machined-surface implants in the posterior mandible: a randomized open-ended clinical trial. Clinical Implants Dentistry & Related Research 5, van Steenberghe, D. (1989) A retrospective multicenter evaluation of the survival rate of osseointegrated fixtures supporting fixed partial prostheses in the treatment of partial edentulism. Journal of Prosthetic Dentistry 61, van Steenberghe, D., Yoshida, K., Papaioannou, W., Bollen, C.M.L., Reybrouck, G. & Quirynen, M. (1997) Complete nose coverage to prevent airborne contamination via nostrils is unnecessary. Clinical Oral Implant Research 8, van Steenberghe, D., Quirynen, M., Naert, I., Maffei, G. & Jacobs, R. (2001) Marginal bone loss around implants retaining hinging mandibular overdentures, at 4-, 8- and 12-years follow-up. Journal of Clinical Periodontolology 28, van Steenberghe, D., Quirynen, M., Molly, L. & Jacobs, R. (2003) Impact of systemic diseases and medication on osseointegration. Periodontology ,

106

107 CHAPTER VII. Predisposing conditions for retrograde peri-implantitis, and suggestions for their treatment. VII.1. Introduction The term retrograde peri-implantitis has just recently been introduced trough several case reports (McAllister et al. 1992, Shaffer et al. 1998, Jalbout & Tarnow 2001, Chaffee et al. 2001, Ayangco & Sheridan 2001, Quirynen et al. 2003). It is defined as a clinically symptomatic peri-apical lesion (diagnosed as a radiolucency) which develops shortly after implant insertion while the coronal portion of the implant achieves a normal bone to implant interface (for review, see Quirynen et al. 2003). A retrograde periimplantitis is often accompanied by symptoms of pain, tenderness, swelling, and/or the presence of a fistulous tract (Figure 1). It should be distinguished from a clinically a- symptomatic, peri-apical radiolucency, which is usually caused by placing implants that are shorter than the drilled cavity or by a heat-induced aseptic bone necrosis (McAllister et al. 1992, Reiser & Nevins 1995, Ayangco & Sheridan 2001). The latter, inactive lesions do not need further treatment, unless their size increases. Retrograde periimplantitis can result from bacterial contamination during insertion, premature loading leading to bone micro-fractures, or the presence of a pre-existing inflammation (bacteria, inflammatory cells, and/or remaining cells from a cyst, granuloma). Such lesions appear to initiate at the implant apex but exhibit the capacity of spreading coronally, proximally, and facially. 97

108 Figure 1: Despite a thorough cleaning of a peri-apical lesion after tooth extraction (a), with a favorable healing visualized on the 6-months post extraction radiograph (b), the patient complained of a local irritation in the vicinity of the apex of the implant some weeks after implant insertion (15 mm Brånemark system MkIII implant). A new radiograph (c), with a gutta-percha cone in the fistula see arrow-, pointed towards the implant apex as the origin of the infection. After elevation of a full-thickness flap (d, see location of the fistula) granulation tissue was found around the apex of the implant (e). The granulation tissue was removed meticulously and the implant was cleaned vigorously (f). The radiograph immediately afterwards (g) with a provisional abutment in place, showed an integrated upper part of the implant (Periotest value of -3). After 6 months the final abutment and crown were placed (h). Both the radiograph (i) and the clinical image after 1 year of loading (j) seemed to indicate a stable condition. The latter was confirmed by the radiograph (k) taken after 18 months. 98

109 Retrograde peri-implantitis should also be distinguished from non-integration, e.g. when the apex of the implant touches the tooth and/or when the implant is inserted in an active endodontic lesion from an adjacent tooth. Under the latter conditions the implant often exfoliates spontaneously or becomes mobile, revealing non-integration. Sussman and Moss (1993) presented 2 basic pathways for a common lesion between tooth and implant. When an implant insertion results in devitalisation of the neighbouring tooth (e.g. due to insufficient distance, overheating of the bone during drilling, cutting off the blood supply to the pulp), it can cause an endodontic pathology that consequently may infect the integrating implant (pathway I). A peri-apical lesion from a nearby devitalized tooth on the other hand (pathway II), can encroach upon the implant and contaminate it (e.g. reactivation of a dormant peri-apical lesion or removal of the peri-apical endodontic seal). Brisman et al (2001) reported that even asymptomatic endodontically treated teeth with a normal peri-apical radiographic appearance could be the cause of an implant failure. They suggested that micro-organisms may persist even though the endodontic treatment was considered radiographically successful, because of inadequate obturation or incomplete seal. Information on the incidence and treatment of retrograde peri-implantitis is scarce and can only be retrieved from some case reports. McAllister et al (1992) reported on two patients in whom peri-apical lesions appeared around implants inserted immediately after extraction of periodontally involved teeth. They hypothesized that the microbial source was likely to be from the extracted teeth. Shaffer et al (1998) suggested that an extension of pathology (probably the bacteria) from a large peri-apical lesion around a neighbouring tooth (untreated or failing root canal-treated tooth), formed the source of their apical implantitis cases. A persistent peri-apical endodontic lesion indeed can be inflammatory in nature (Green et al. 1997). All three patients documented by Ayangco & Sheridan (2001) had a history of failed endodontic and apicectomy procedures, which finally led to extraction of the involved teeth and subsequent placement of implants after a short healing time (1-4 months). It would appear that even after thorough debridement and irrigation of the extraction sockets and a short healing time, bacteria (or cysts/granuloma) remain in the bone, and lead to the initiation of retrograde peri- 99

110 implantitis. Jalbout & Tarnow (2001) reported on four patients with retrograde periimplantitis without a clear explanation on the etiology. The first aim of this study was to estimate, via a retrospective analysis of all solitary implants inserted at the department of Periodontology, the incidence of retrograde periimplantitis. The second aim was to evaluate longitudinally the initial outcome of 2 treatment options. VII.2. Materials and methods: All solitary implants (n = 539) inserted at the Department of Periodontology of the University Hospital of the Catholic University in Leuven (from 1987 on, but primarily after 1993) were included in this retrospective evaluation. They had all been installed in an operatory room environment with stringent asepsia measures, always following the initial guidelines of Brånemark (Adell et al. 1985). Since the mid-1990s, however the use of a nose cape was introduced (van Steenberghe et al. 1997). Al implants had been left submerged for several months before abutment connection. The prosthethic phase was performed by the Department of Prosthetic Dentistry or by the referring dentist. For each implant a series of parameters were considered (if available) including (see Table I): implant characteristics (length, diameter, and surface modification), surgical aspects (guided bone regeneration, fenestration, dehiscency), recipient site parameters (bone quality and quantity), reason for tooth loss (caries, periodontitis, severe peri-apical lesion, trauma), congenital absence of a tooth, and condition of adjacent teeth (endodontic therapy, peri-apical lesion, periodontitis). An implant was considered a failure if one or more of the following observations were recorded: implant mobility [evaluated both clinically and by means the Periotest device (Gulden, Germany)], radiographic evidence of radiolucency around the implant, or subjective pain. For all implants with retrograde peri-implantitis, additional information was retrieved (if available) on complications during tooth extraction, radiographic evaluation of the bone healing, and bone appearance at implant insertion. When available, the torque needed to insert the implants measured by means of the Osseocare (Nobel Biocare, Gothenburg, Sweden) was also taken into account. 100

111 Table 1: Characteristics of successful and failed implants, implants with retrograde peri-implantitis, and difference between machined and TiUnite surfaces. Data in bold = clearly different from success. Characteristics Implant type Implant outcome All implants Machined TiUnite Failed Success Retrograde periimplantitis Upper jaw Number Mean age at insertion (years) 42,0 42,2 40,8 43,1 41,9 45,7 Mean implant length (mm) 13,7 13,7 13,7 13,0 13,8 14,3 Mean bone quality* 2,57 2,52 2,83 2,79 2,56 2,86 Mean bone quantity* 1,98 1,95 2,12 2,26 1,96 2,29 Implant loss (%) 5,6 6,4 1, Guided bone regeneration (%) 14,8 15,3 11,9 33,3 13,6 14,3 Fenestration (%) 6,1 6,7 3,0 0,0 6,5 14,3 Dehiscency (%) 8,9 8,9 9,0 8,3 9,0 0,0 Reason tooth loss (%) Caries 19,5 20,1 16,4 16,7 19,7 0,0 Periodontitis 12,4 12,5 11,9 4,2 12,9 28,6 Fracture 28,4 28,4 28,4 33,3 28,1 28,6 Apical lesion 8,2 8,4 7,5 25,0 7,2 28,6 Agenesis 13,4 11,7 22,4 4,2 13,9 14,3 Trauma 16,7 17,9 10,4 16,7 16,7 14,3 Left neighbour (%) Endo treatment 11,3 10,6 14,9 8,3 11,4 28,6 Apical lesion 2,1 2,2 1,5 8,3 1,7 14,3 Periodontitis 6,3 6,7 4,5 8,3 6,2 14,3 Right neighbour (%) Endo treatment 12,7 13,4 9,0 16,7 12,4 14,3 Apical lesion 3,1 3,1 3,0 20,8 2,0 0,0 Periodontitis 5,2 5,3 4,5 4,2 5,2 14,3 Retro peri-implantitis (%) 1,9 0,3 9,0 0 1,7 100 Lower jaw Number Mean age at insertion (years) 40,0 40,1 39,2 41,7 39,9 41,0 Mean implant length (mm) 11,5 11,5 11,5 11,6 11,5 10,5 Mean bone quality* 2,39 2,35 2,80 2,50 2,38 2,0 Mean bone quantity* 1,76 1,68 2,50 1,75 1,76 2,0 Implant loss (%) 8,0 8,0 7, ,3 Guided bone regeneration (%) 2,7 3,0 0,0 0,0 2,9 0,0 Fenestration (%) 3,5 4,0 0,0 0,0 3,8 0,0 Dehiscency (%) 5,3 5,0 7,7 11,1 4,8 0,0 Reason tooth loss (%) 101

112 Caries 25,7 27,0 15,4 0,0 27,9 100 Periodontitis 8,8 8,0 15,4 0,0 9,6 0,0 Fracture 12,4 13,0 7,7 11,1 12,5 0,0 Apical lesion 10,6 9,0 23,1 33,3 8,7 100 Agenesis 29,2 30,0 23,1 22,2 29,8 0,0 Trauma 4,4 5,0 0,0 11,1 3,8 0,0 Left neighbour (%) Endo treatment 7,1 8,0 0,0 0,0 7,7 0,0 Apical lesion 1,8 2,0 0,0 0,0 1,9 0,0 Periodontitis 8,0 9,0 0,0 0,0 8,7 0,0 Right neighbour (%) Endo treatment 13,3 11,0 30,8 22,2 12,5 67,0 Apical lesion 2,7 2,0 7,7 11,1 1,9 0,0 Periodontitis 7,1 8,0 0,0 0,0 7,7 0,0 Retro peri-implantitis (%) 2,7 1,0 15,4 22,2 1,0 100 *Based on Lekholm & Zarb classification (1985) Data in bold are at least 3 X higher than observation for successful implants Prospective evaluation: All implants with a retrograde peri-implantitis were followed longitudinally via intraoral, parallel technique radiographs, to estimate both changes in the marginal bone level and progression of the apical extension of lesion. The biomechanical stability was scored with the Periotest device. The treatment protocol of retrograde lesions in the upper jaw included: elevation of a fullthickness flap, complete removal of all accessible granulation tissue with hand instruments (special attention to reach both apical and oral part of the implant surface), and curettage of the bony cavity walls. In half of the defects deproteinised bovine bone mineral (Bio-Oss, Geistlich, Schlieren, Switzerland) was used as bone substitute (at the decision of the surgeon), while the other defects were left empty. In the lower jaw an explorative flap mostly revealed an absence of a perforation of the cortex so that a trepanation of the bone had to be performed. Most of these interventions were performed under antibiotic coverage (ß-lactamase resistant penicillin). Further details are summarized in Table

113 Table 2: Characteristics of implants with retrograde peri-implantitis per patient (initials). Parameter Patient initials DW De Te Ja Sn VD Vho Vhe He Va Position Tooth history Endo therapy n n y n y y y y - y Peri-apical lesion n > > y n < < > - > Terminal periodontitis n n n y n n n n n n Impacted y n n n n n n n n n Extract history Loss buccal plate n n GBR n n y n n - y Buccal fenestration n tomo n tomo n n y n - n Implant surgery Delayed n y y n n n n n n n Healing time # y 10 m 12 m 18 m 7 m 29 m 6 m 7 m 8 m 6 m Buccal concavity y n y n n n y GBR n n n Remaining infection n n y n n y y n n n Implant type TiU MKIII TiU TiU TiU TiU TiU TiU st TiU Implant length (mm) , OsseoCare (Ncm) Crestal third Middle third Apical third Antibiotics no 1 g no no no 4 d 4 d 2 g no no First symptoms ab i+3m i+1m i+3m i+3m i+1m i+1m ab ab i+2w Timing symptoms Fistula n y n y y y y n n y Obvious pus n n n y y n y n n y Pain n n y n n y n n n n Swelling n n y n n y y n n y Defect size Height large Width large PTV At abutment Treatment 103

114 Curretage ab ab i+1m ab x2 i+5m i+1m i+1m expl - ab+18 m expl - Bone substitute y n y n y n y n n N Membrane n n n y n n n n n N Complication n n n n n n! n i loss N!: fistula from implant to endodontically treated tooth abbreviations: GBR = guided bone regeneration, tomo = tomography, m = months, TiU = TiUnite, st = standard, g = grams, d = days, I = implant insertion, ab = abutment connection, w = weeks. Statistical analysis: In order to verify the significance of the impact of the implant surface on the incidence of retrograde peri-implantitis and implant failure, a Chi-square test (2 by 2 table) was performed. The threshold level for significance was set at VII.3. Results: Overall data: A total of 426 solitary implants in the upper jaw and 113 in the lower jaw were included in this retrospective evaluation. All implants (n = 539) were of the Brånemark system type (Nobel Biocare, Gothenburg, Sweden) with the following distribution for upper and lower jaw respectively: standard implants (26/27), selftapping implants (14/0), MKII (256/64), MKIII machined (63/9) and TiUnite MKIII implants (67/13). A total of 24 (5.6 %) and 9 (8.0 %) implants were lost in upper and lower jaw respectively, mostly before, at or soon after abutment connection (17/24, 7/9). In the upper jaw, implant losses were four times higher for machined implants (23/359, 6.4%) when compared to TiUnite implants (1/67, 1.5%), although known parameters for implant failures were similar (e.g. bone quality and quantity, implant length, surgical complications (see Table 1). When compared to successful implants, failed implants were characterized by a 2.5x higher frequency of guided bone regeneration (upper jaw only), a 3x higher incidence of 104

115 a peri-apical lesion around the extracted tooth, and a >5x higher incidence of endodontic pathology on the adjacent teeth (the later only for the upper jaw). Retrograde peri-implantitis: Ten implants showed retrograde peri-implantitis, 7 in the upper jaw (1.6 %) and 3 in the lower jaw (2.7%), respectively. The intra-oral radiographs taken at the first symptoms (Table 2) are depicted in Figure 2. They all showed a clear radiolucency around the apical part of the implant, while the bone apposition at the coronal part seemed intact. Figure 2: Intra-oral long-cone radiographs made at the moment the first symptoms of retrograde peri-implantitis could be registered or at abutment connection (Ja & Va). The letters give the initials of the patient (as in Table 2 and figures 3 and 4). Sometimes a gutta percha cone illustrated the direction of the fistula. 105

116 In the upper jaw, six lesions were detected before, and 1 at abutment connection (the patient with an impacted canine, DW). A fistula was present in 5 out the 6 patients, but pus was only seen in 4 of them. Swelling was observed in 3 of the 7 patients. When an explorative flap was raised, a large perforation of the buccal bone plate was always present (7/7). The implant was always surrounded by granulation tissue but pus was never observed at surgery- (7/7). The extension of bone destruction could reach 10 by 5 mm (see Table 2 and Figure 3). Figure 3: Some clinical pictures of retrograde peri-implantitis. In the upper jaw the cortical plate was always found to be penetrated. After removal of all granulation tissue, the apex of the implant was found to be no longer in contact with bone. Nevertheless all implants remained osseointegrated as confirmed by the negative PTV values. The letters give the initials of the patient (as in Table 2 and figures 2 and 4) The apex of the implant was always involved with the bone being millimeters higher up. In several patients the palatal bone plate was also perforated. None of the implants was mobile, even though they remained only with the most coronal 4 to 5 threads in contact with the bone. In the lower jaw 2 of the 3 patients with retrograde per-implantitis were only detected via the radiograph taken at abutment connection. The third patient presented a local swelling 2 weeks after implant insertion. An explorative flap did not reveal any cortical except in one 1 patient (Va, small lingual perforation). 106

117 The incidence of retrograde peri-implantitis was significantly higher (p < ) for TiUnite implants (8/80, 10%) than for machined implants (2/459, 0.4%). Moreover, when the characteristics of their recipient sites were compared with those of successful implants (Table 1), it became obvious that the incidence of peri-apical lesion on extracted tooth and endodontic pathology/therapy on both extracted and adjacent teeth were clearly higher. Table 2 shows additional information for these 10 implants. Except in 1 situation (DW, with a previously removed impacted canine and possible signs of remaining scar tissue), all extracted teeth had been endodontically treated and most of them showed a small (n = 3) or obvious (n = 2) peri-apical lesion (no information for 1 patient). For 4 of the 6 upper implants (impacted canine site excluded) the implant insertion had been delayed (n = 2) due to incomplete healing, and/or remaining inflammatory tissue had been removed per-operatively (n = 3). The healing time between extraction and implant insertion was less than 1 year for most implants (7/10). The Osseocare data did not reveal extreme bone resistance during any implant insertion. Seven different periodontologists were involved with the surgical procedures all following the same protocol. Individual variation between surgeons could not be detected. Treatment options of retrograde peri-implantitis: The lesions and implants apices in the upper jaw were all thoroughly curetted (Figure 3, Table 2), and in 4 of the 7 lesions a bone substitute was used. Most lesions healed uneventfully (Figure 4), although 2 patients showed a small fistula (without pus) afterwards. In 1 patient (VHo) the endodontically involved adjacent tooth was responsible, for the other (Ja) a second intervention was successful. 107

118 Figure 4: Intra-oral long-cone radiographs made at the patient s latest visit. The letters give the initials of the patient (as in Table 2 and figures 2 and 3). Both the marginal bone levels as well as the apical part of the implant show either a stable or an improved aspect. In patient VHo an apexectomy had been performed together with the cleaning of the defect. This was the only case with complications at a later stage (fistula that made contact between peri-implant pocket and apex of the neighbouring tooth). Since the defect in the lower jaw could not be reached, improvements could not be detected. In the lower jaw, explorative flaps did not show a perforation of the cortical bone so that a trepanation of the defect and antibiotics were the first option. In the single patient with longer follow-up, this treatment did not arrest the progression of the lesion and the implant was lost after 18 months (He). 108

119 VII.4. Discussion Even with improved implant surfaces and stringent guidelines for implant surgery and follow-up, still not all implant insertions are successful (van Steenberghe et al. 1999). This retrospective study indicates that some of the early failures may be linked with an endodontic pathology, either remaining after tooth extraction or around neighbouring teeth. The higher incidence of these pathologies for failed implants and/or for implants with retrograde peri-implantitis versus successful implants is clear cut. Our observations are supported by previous case reports on peri-apical lesions around implants and/or on early implant failures (McAllister et al. 1992, Sussman & Moss 1993, Reiser & Nevins 1995, Piattelli et al. 1998, Shaffer et al. 1998, Ayangco & Sheridan 2001, Chaffee et al. 2001, Brisman et al. 2001, Shabahang et al. 2001, Jalbout & Tarnow 2001, Quirynen et al. 2003). Histological evaluations of peri-apical tissues after endodontic therapy in cadavers, animals or man clearly highlight that although radiographs can indicate an optimal healing, the apex of endodontically treated tooth often exhibits histological signs of inflammation or persisting microorganisms (Rowe & Binnie 1974, Green et al. 1997, Seltzer 1999). Green et al. (1997) reported that 26% of endodontically treated teeth with a normal radiographic appearance still had histological signs of inflammation. Seltzer (1999) removed in children a small block section containing the root tip and surrounding tissues from 14 teeth, 6 to 30 months after endodontic treatment. He found histological evidence of peri-apical chronic inflammatory lesions in at least 50% of the specimens. Rowe and Binnie (1974) performed endodontic treatment on 180 teeth in beagles. Of the 129 teeth that displayed no radiographic abnormality, 47 percent demonstrated a histological apical inflammatory response. From these studies it is obvious that bacterial endotoxins, inflammatory cells, or bacteria themselves can be encountered around endodontically treated teeth and as such can be responsible for the early contamination of an implant surface. Not all endodontic treatments are clinically successful. Ørstavik and Ford (1998) reported a 89% success rate for endodontic treatments when performed by specialists, with higher failure rates for lateral upper incisors, the first upper premolars and the first lower molars. After extraction of an endodontically involved tooth a radicular cyst might remain/develop in the bone. Reiser and Nevins (1995) associated the 109

120 higher frequency of peri-apical implant lesions in the maxilla with the known higher frequency of radicular cysts in this jaw. Bhaskar (1966) indeed reported a 10:1 ratio of radicular cysts in the maxilla compared to the mandible and speculated that the maxilla contains far more epithelial debris (epithelial rests of Malassez), as well as epithelium left in the wake of fusion of the facial processes. Therefore, it is conceivable that a periapical granuloma has a greater potential of becoming a radicular cyst in the maxilla than it has in the mandible. Such cysts, although often sterile, can compete with the osseointegration process at the apical part of the implant. The TiUnite implants with enhanced surface characteristics (micropores obtained via electrochemical oxidation) were introduced by Nobel Biocare to accelerate the osseointegration process. Several studies indeed clearly confirmed this accelerated healing and bone to implant contact (Albrektsson et al. 2000, Gottlow et al. 2000, Rompen et al. 2000, Friberg & Billström 2002, Ivanoff et al. 2003, Rocci et al. 2003). The higher incidence of retrograde peri-implantitis in our clinic for TiUnite implants can thus be explained. While the machined implants, when getting in contact with a cysts or endodontic pathology, will soon be completely surrounded by granulation tissue (and subsequently fails), the TiUnite surface implants will not have the same fate because of the accelerated bone apposition. As such the coronal part of the TiUnite implant still integrates before the fibrous encapsulation can reach this area. The machined implant would already have failed by that time. The higher incidence of implant failure at sites with a history of a peri-apical granuloma or in the neighbourhood of teeth with endodontic pathology justifies a more detailed analysis of the radiographs before implant insertion. Unfortunately, remaining pathologies both in upper or lower jaw are often not detectable on radiographs. An endosseous defect in the mandible or maxilla can indeed only be detected radiographically when the cortex or the junction area is destructed (van der Stelt 1985, Bianchi et al. 1991). Even on tomograms or CT scan images these lesions are difficult to diagnose (Velvart et al. 2001). 110

121 VII.5. Conclusion: The treatment of peri-apical peri-implantitis is still empiric. The longitudinal data from this study, together with the outcome in some case reports, seem to indicate that the removal of all granulation tissue is sufficient to arrest the progression of the bone destruction. The removal of the apical part of the implant does not seem mandatory. Our data also indicate that implants from which only the coronal part is osseointegrated can successfully resist occlusal load for years, at least in the single tooth replacement condition. 111

122 VII.6. References: Adell, R., Lekholm U. & Brånemark, P-I. (1985) Surgical procedures. In: Brånemark, P-I., Zarb, G.A. & Albrektsson, T., eds. Tissue-Integrated Prostheses, Osseointegration in Clinical Dentistry, Chicago: Quintessence Publ Co. Albrektsson, T., Johansson, C., Lundgren, A.K., Sul, Y. & Gottlow, J. (2000) Experimental studies on oxidized implants. A histomorphometrical and biomechanical analysis. Applied Osseointegration Research 1, Ayangco, L. & Sheridan, P.J. (2001) Development and treatment of retrograde periimplantitis involving a site with a history of failed endodontic and apicoectomy procedures: a series of reports. International Journal of Oral & Maxillofacial Implants 16, Bhaskar, S. (1966) Periapical lesions - types, incidence and clinical features. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology & Endodontics 21, Bianchi, S.D., Roccuzzo, M., Cappello, N., Libero, A., Rendine, S. (1991) Radiological visibility of small artificial periapical bone lesions. Dentomaxillofacial Radiology 20, Brisman, D.L., Brisman, A.S. & Moses, M.S. (2001) Implant failures associated with asymptomatic endodontically treated teeth. Journal of the American Dental Association 132, Chaffee, N.R., Lowden, K., Tiffee, J.C. & Cooper, L.F. (2001). Periapical abscess formation and resolution adjacent to dental implants: a clinical report. Journal of Prosthetic Dentistry 85, Friberg, B. & Billström, C. (2002) Preliminary results of a prospective multicenter clinical study on TiUnite implants. Applied Osseointegration Research 3, Gottlow, J., Henry, P., Tan, A., Allan, B., Johansson, C. & Hall, J. (2000) Biomechanical and histologic evaluation of the TiUnite and Osseotite implant surfaces in dogs. Applied Osseointegration Research 1, Green, T.L., Walton, R.E., Taylor, J.K. & Merrell, P. (1997) Radiographic and histologic periapical findings of root canal treated teeth in cadaver. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology & Endodontics 83, Ivanoff, C-J., Widmark, G., Johansson, C. & Wennerberg, A. (2003) Histological evaluation of the bone response to oxidised and turned titanium micro implants in human jawbone. International Journal of Oral & Maxillofacial Implants 3, Jalbout, Z.N. & Tarnow, D.P. (2001). The implant periapical lesion: four case reports and review of the literature. Practical Procedures & Aesthetic Dentistry 13, McAllister, B.S., Masters, D. & Meffert, R.M. (1992) Treatment of implants demonstrating periapical radiolucencies. Practical Procedures & Aesthetic Dentistry 4, Ørstavik, D. & Ford T.R. (1998) Essential Endodontology-Prevention and treatment of apical periodontitis, 1 st ed, Blackwell Science, Great Britain, 410pp. 112

123 Piattelli, A., Scarano, A., Balleri, P. & Favero, G.A. (1998) Clinical and histologic evaluation of an active "implant periapical lesion": a case report. International Journal of Oral & Maxillofacial Implants 13, Quirynen, M., Gijbels, F., Jacobs, R. (2003) An infected jawbone site compromising successful osseointegration. Periodontology , Reiser, G.M. & Nevins, M. (1995) The implant periapical lesion: etiology, prevention, and treatment. Compendium of Continuing Education in Dentistry 16, Rocci, A., Martignoni, M. & Gottlow, J. (2003) Immediate loading of Brånemark System with TiUnite and machined surfaces in the posterior mandible: A randomised, open-ended trial. Clinical Implant Dentistry and Related Research 5, Rompen, E., DaSilva, D., Lundgren, A.K., Gottlow, J. & Sennerby L. (2000) Stability measurements of a double-threaded titanium implant design with turned or oxidized surfaces. An experimental resonance frequency analysis study in the dog mandible. Applied Osseointegration Research 1, Rowe, A.H. & Binnie, W.H. (1974) Correlation between radiological and histological inflammatory changes following root canal treatment. Journal of the British Endodontics Society 7, Seltzer, S. (1999). Long-term radiographic and histological observations of endodontically treated teeth. Journal of Endodontics 25, Shabahang, S., Bohsali, K., Caplanis, N. & Torabinejad, M. (2001) Effects of periradicular lesions on osseointegration of dental implants. Journal of Endodontics 27, Shaffer, M.D., Juruaz, D.A. & Haggerty, P.C. (1998) The effect of periradicular endodontic pathosis on the apical region of adjacent implants. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology & Endodontics 86, Sussman, H.I. & Moss, S.S. (1993) Localized osteomyelitis secondary to endodontic-implant pathosis. A case report. Journal of Periodontology 64, van der Stelt, P.F. (1985) Experimentally produced bone lesions. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology & Endodontics 59, van Steenberghe, D., Yoshida, K., Papaioannou, W., Bollen, C.M.L., Reybrouck, G. & Quirynen M. (1997). Complete nose coverage to prevent airborne contamination via nostrils is unnecessary. Clinical Oral Implants Research 8, van Steenberghe, D., Quirynen, M. & Naert, I. (1999). Survival and success rates with oral endosseous implants. In: Proceedings of the 3 de European Workshop on Periodontology. Lang, N.P., Attstrom, R. & Lindhe, J. (Eds.) Quintessence Publ Co Inc, Berlin, pp Velvart, P., Hecker, H. & Tillinger, G. (2001) Detection of the apical lesion and the mandibular canal in conventional radiography and computed tomography. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology & Endodontics 92,

124

125 GENERAL DISCUSSION

126 118

127 CHAPTER VIII. General discussion This thesis achieved the aims that were set at the beginning of the study by identifying: I- The systemic and local factors that may interfere with osseointegration of oral implants, either before (early) or after (late) abutment connection. The fact that there is a general policy at the department to accept all patients who can benefit from implants for their oral rehabilitation -even in the presence of systemic or local factors which can compromise the outcome- the incidence of such eventually compromising factors was not negligible. The reliability of the incidence was very high since the department is part of a very large university hospital where many of the patients are in treatment at other medical departments. Thus full access to laboratory tests and patients files is guaranteed. Furthermore the house doctor was consulted in case of remaining questions. Since one and the same implants hardware (Brånemark system, Nobel Biocare, Gothenburg, Sweden) was used for two decades in the department, this allowed to reduce the number of covariables thus disclosing more easily the impact of systemic and/or local factors. Both retrospective and prospective studies were performed. II- The impact of implant surface characteristics on implant losses. This was done by using two different surfaced implants when replacing lost implants by new ones, and by evaluation of the incidence of retrograde periimplantitis with both of them. 117

128 Part I. Effect of systemic and local factors on implant loss Early failures (up to abutment connection): Chapter II: A retrospective study with mostly machined surfaced implants. This study analyzed a very large number of consecutive patients (n = 2004) provided with a total of 6946 implants. The above mentioned homogeneity of hardware and the liberal uptake of patients with systemic and local factors which may interfere with osseointegration allowed identifying their impact. Since the observation was limited to the healing period up to the abutment connection, confounding factors like forces acting on the implants and microbiological challenges were eliminated. All patients treated by means of endosseous implants during the period at the Department of Periodontology of the University Hospital of the Catholic University Leuven were included in the study. All implants which were followed had been inserted applying the classical two-stage surgical protocol with a period of several months between implant insertion and implant connection. Thus implants which were for example immediately loaded, like the Brånemark Novum and the Litorim systems, were excluded from this study. The surgical records which were not fully completed led to the examination of the patient s file. For a total of 232 patients, because of purely administrative reasons, the files could not be retrieved. This led to a total of 2004 patients which could be included in this study. It is known that anodized titanium surface like TiUnite leads to less early failures. The limited series of the latter surface observed in the present study, did not allow confirming this. A later more focused study (see chapter VI) confirmed the reduced failure rate with Ti-Unite surfaced implants. The same applies to the large scale prospective study (see chapter III). In the present retrospective study, significant effects of implant length and diameter were detected, more failures occurring with short and wide-diameter implants. One should realize however that these implants were systematically installed in 118

129 compromised sites, marked by poor bone quality and quantity. Thus these confounding factors may explain the higher failure rate. This is in accordance with other studies (review see Renouard & Nisand 2006). Too high bone densities, as assessed clinically and/or radiologically, have also been pointed as a possible reason for non-integration (Engquist et al. 1988, Friberg et al. 1991, Jaffin & Berman 1991). Our present findings confirm that bone quality type 1 and 4 the very dense and the very soft- (Lekholm & Zarb 1985) are associated with slightly higher failure rates. Several previous studies drew the attention on the negative effect of smoking on osseointegration, and its dose-related effect (for review see Bain 1996). Our retrospective study confirms this. Crohn s disease is characterized by the presence of many antibody-antigens complexes and autoimmune inflammatory reactions in several parts of the body leading to enteritis, vasculitis, recurrent oral ulcerations, arthritis or keratoconjuctivitis. The same probably occurs at the bone to implant interface when even biocompatible implants are inserted with a surgical trauma. The hyperreactive process in Crohn s patients can lead to a hyperreactive response, thus affecting the outcome of implant osseointegration (van Steenberghe et al. 2002). Moreover, the malnutrition encountered in Crohn s patients can also contribute to a deficient bone healing around the implant (Esposito et al. 1998). Although no studies did prove an association between implant failures and the state of osteoporosis, it has been suggested as a risk factor especially for postmenopausal women (Becker et al. 2000). The present retrospective study reveals a significant correlation between early implant failures and osteoporosis. The same study indicates a significant association between early implant failures and the vicinity of natural teeth. It had been previously reported in a multicenter study that a significant plaque accumulation and gingival inflammation, at the time of implant placement, increases the risk of implant failure (van Steenberghe et al. 1990). A retrospective study indicated that some of the early failures may be linked with an endodontic pathology, either remaining after tooth extraction or around neigbouring teeth (van Steenberghe et al. 1999). 119

130 In another retrospective study (chapter VII) some of the early failures were linked to periapical pathologies of either neighbouring teeth or from the extracted teeth. These findings were in accordance with other studies (for review see Quirynen et al. 2003). The vast majority (about 90 %) of the more than 6200 implants considered in the present study had a machined surface. In order to evaluate the effect of modified surface characteristics on implant outcome by investigating systemic and local factors, a prospective study was undertaken (chapter III). In the later only TiUnite implants were used. Chapter III: a prospective study with exclusively TiUnite implants When a group of 283 consecutive patients provided with 720 MkIII TiUnite implants was followed up to abutment connection, only 1.9% of implants failed. Due to the very low failure rate, this study could only identify potentially influential factors for the implant failure but could not draw definitive conclusions. The statistical analysis has therefore considered descriptive. This association with failure rate had been also observed in the previous chapter for both types of implant surfaces: machined and TiUnite. Other factors like osteoporosis, bone quality and quantity, implant length, diameter and location in the oral cavity, seem not to affect the early failure rate in the present study. This may be due to the faster and improved bone response, in terms of implant-bone contact and removal torque, as compared to what occurred with machined surfaces (Carlsson et al. 1988, Buser et al. 1991, Huré et al. 1996). On the other hand other factors, which were not evaluated in chapter II, could be detected in the present prospective study. These are apical lesions around the recipient site, hormone replacement therapy, radical hysterectomy, gastric problems and diabetes type I. Although the apical lesions, i.e. infrabony lesions at the tip of the implant, seem to affect significantly early failure rate in TiUnite implants, this rate is relatively low when compared to machined surface implants. The latter were often lost while the TiUnite surfaces demonstrated an apical lesion but could be maintained (see chapter VII). 120

131 A previous retrospective study (Moy et al. 2005) revealed that women on estrogen replacement had a significantly lower success rate than healthy population. Postmenopausal women not on hormone replacement therapy did not have this increased failure rate. In the present study there is tendency for more failures in women with hormone replacement therapy To find out about the eventual effect of all these factors on late implant failures, which means after the implant has been exposed to loading and the microbiological challenge of the oral cavity, a retrospective study with a prolonged observation period was undertaken (see chapter IV) Late failures (up to two years after abutment connection): Chapter IV: A retrospective study on a random selection of more than 400 patients From the total set of patient group evaluated in chapter II, 700 files were randomly selected. The following implants were not considered for the analysis: Implants which failed before or at abutment surgery. Implants which, although not failed, belonged to patients who could not be followed for the present follow-up period of 2 years after abutment connection. The remaining data consisted of 412 patients (240 females) who were provided with a total of 1514 implants. The end point (two years) has been randomly chosen. It is also related to the date of introduction of TiUnite surfaced implants in the department (end of 2002). By choosing a sufficient follow-up time we made sure that a relatively sufficient number of implants with TiUnite surfaces would be available within the group. Although in the present study the statistical analysis revealed no significant difference in late failure rate between the machined surface and TiUnite surface implants, a trend for more implant losses with the former was detected. The present study, confirmed the trend for more losses occurring with wide-diameter implants. Such increased implant loss of wide-diameter implants was mainly associated 121

132 with a learning curve, poor bone density, implant design and site preparation, and the fact that it is usually used as rescue implants (Renouard & Nisand 2006). Here too, these implants were mostly installed in sites with poor bone quality and quantity. These confounding factors offer a possible explanation. Low bone density -as assessed clinically or radiologically- has also been pointed out as a possible reason for non-integration (Engquist et al. 1988, Friberg et al. 1991, Jaffin & Berman 1991). In our present findings bone quality type 4 was indeed associated with slightly higher implant loss. The effect of smoking on the late failure rate could not be evidenced in the present study. Smoking is on the other hand well known to have an impact on early failure rate (Bain & Moy 1993). This may be explained by the effect of smoking on the wound healing process, in early stage of osseointegration (chapter II) while other factors became more predominant for the late failures. Higher PTVs at abutment, reflecting a less rigid implant-bone continuum, was associated with more implant losses at a later stage. The only possible explanation is the less rigid bone to implant interface which may lead under loading to a reversal of the osseointegration process: a dedifferentiation to fibrous scar tissue. Lack of intimate bone apposition results in implant marsupialisation (Ivanoff et al. 1996, Szmukler-Moncler et al. 2000) and finally loss of the implant. The higher implant loss in the upper jaw vs. the lower, and in the posterior vs. the anterior region reflect the thinner cortex combined with less dense trabecular bone (Jacobs 2003). The poor degree of bone mineralization reveals itself on the radiographs (Friberg et al. 1995, 1999). Dorsal regions are also subjected to higher (x 3) chewing forces when compared to the anterior (Helkimo 1977). The long term effects of radiotherapy on bone quality are well known: a progressive decrease of vascular supply. The decreased blood supply to the bone can lead to osteoradionecrosis. (el Askary et al. 1999, Marx et al. 1987, Keller, 1997). The dramatic effect of radiotherapy on late failures should lead to a cautious application of osseointegration in such patients unless of hyperbaric oxygen is applied. On the other hand, even with a higher failure rate, the implant retained prosthesis is often the only option because of xerostomia. The latter impairs the use of dentures. 122

133 Since bone quality affects the outcome in some instances (chapter II and chapter IV) it was logical to define whether preoperative assessment as is routinely done on radiographs of the jaws is a valid and discriminating method. We undertook a study trying to confirm the validity of this subjective assessment by relating it to several biomechanical measuring devices (see chapter V). Chapter V: A biomechanical assessment of the relation between the subjective bone quality assessments on oral implant stability at insertion The currently most popular clinical method of bone quality assessment was developed by Lekholm & Zarb (1985). They introduced a scale of 1 to 4, based on both the radiographic jaw bone assessment and the tactile sensation experienced by the surgeon when preparing the fixture site. The grading refers to individual experience, and furthermore, it provides only a subjective and rough mean value for the entire jaw. A series of objective measurement devices, related to bone quality are presently available. They are: the insertion torque measurements (Osseocare ), the Periotest value and the resonance frequency analysis (Osstell ). Although for a variety of reasons (clinical workload and technical problems with the devices) biomechanical assessments were only performed on a fraction of the patients, the present data substantiate the relationship between such objective measurements and the Lekholm and Zarb index. It opens new perspectives for their clinical application. The insertion torque measurements were higher in the lower jaw, when compared to the upper, especially in the symphyseal area. The posterior region of the upper jaw has the lowest torque value. This is in agreement with a previous study (Friberg et al. 1999a). In the posterior maxilla, indeed, a (very) thin cortical bone is often combined with less dense trabecular bone (Jacobs 2003). Thus, clinicians generally observe a poor degree of bone mineralization on the radiographs and a limited bone resistance while drilling in this area (Friberg et al. 1995, 1999b). This is ascertained by the fact that PTV were significantly lower in the lower jaw comparing to the upper jaw indicating more rigidity. Several studies confirmed this (Olivé & Aparicio 1990, van Steenberghe et al. 1995) 123

134 In the present study a significant correlation was detected between the Lekholm & Zarb index and the torque force measurements, ISQ, and PTV. This is in accordance with previous studies (Johansson et al. 2004, Homolka et al.2002, Friberg et al.1999). Very thick cortical bone was significantly related with low PTV and high ISQ values. Previously van Steenberghe et al. (1995) showed that PTV values were lower (i.e. more rigidity) for implants with a bicortical vs. a monocortical contact. Since in chapter II no significant difference -but a tendency- was observed between implants with different surface characteristics, machined vs. TiUnite, we felt this should be further analysed. The very small failure rate observed in chapter III for TiUnite implants, encouraged us to further investigate the impact of surface characteristics on implant failures. In a retrospective study we analysed which influence TiUnite surface may have on the prognosis of implants replacing failed ones (see next chapter) and on the incidence of so-called retrograde periimplantitis (see chapter VII). Part II. The impact of implant surface characteristics on implant loss Chapter VI: The importance of surface characteristics of implants to replace failed ones. In the present study, when failure of a machined-surfaced implant occurred, replacement at the same site by a TiUnite implant surface which has identical but with more oxidized surface geometry led to a higher success rate. This was not the case when a failed implant was replaced by another machined-surfaced implant. Randomization could not apply because of the retrospective nature of the study. The replacement of the failed implant by either a TiUnite or a machined-surface implant was on the other hand forced choice. Indeed, the TiUnite surface was only used in the department at a certain date (end 2002) which fell in the middle of the 3 years period. Thus any bias could be excluded. An implant newly placed at a site where an implant previously failed is again subjected to the same systemic and local compromising factors, the local factor being eventually less favorable because it is not a pristine site anymore. Thus a comparison between the 2 124

135 groups leads to the identification of more or less the only significant variable, namely the implant surface. The other uncontrolled variable at replacement is eventual latent inflammation or scar tissue which remains since the previous surgery. This probably explains the higher failure rate for implant with a machined surface used for replacing failed ones. The impact of smoking habits on the outcome of osseointegration is again evidenced in the present study, as the number of smokers in the replacement group is high comparing with the total patient population. This is in agreement with previous studies (for review see Bain 2003). The small incidence of failures which is characteristic for the presently used implant system renders statistical analyses of failure, i.e. loss, rates difficult. Nevertheless, the present analysis of the fate of implants inserted at the same site where an implant recently failed demonstrated a higher predictability for the improved surface. Chapter VII: The importance of surface characteristics on retrograde peri-implantitis. Even with improved implant surfaces and stringent guidelines for implant surgery and follow-up, still not all implant insertions are successful (van Steenberghe et al. 1999). The present study indicates that some of the early failures may be linked with an endodontic pathology, either remaining after tooth extraction or around neighbouring teeth. The higher incidence of these pathologies for failed implants and/or for implants with retrograde peri-implantitis versus successful implants is clear cut. Histological evaluations of peri-apical tissues after endodontic therapy in cadavers, animals or man clearly highlight that although radiographs can indicate an optimal healing, the apex of endodontically treated tooth often exhibits histological signs of inflammation or persisting microorganisms (Rowe & Binnie 1974, Green et al. 1997, Seltzer 1999). Such remaining foci can be responsible for the early contamination of an implant surface inserted in the vicinity. After extraction of an endodontically involved 125

136 tooth a radicular cyst might remain/develop in the bone. Reiser and Nevins (1995) associated the higher frequency of peri-apical implant lesions in the maxilla with the known higher frequency of radicular cysts in this jaw. Bhaskar (1966) indeed reported a 10:1 ratio of radicular cysts in the maxilla compared to the mandible and speculated that the maxilla contains far more epithelial debris (epithelial rests of Malassez), as well as epithelium left in the wake of fusion of the facial processes. Therefore, it is conceivable that a peri-apical granuloma has a greater potential of becoming a radicular cyst in the maxilla than it has in the mandible. Such cysts, although often sterile, can compete with the osseointegration process at the apical part of the implant. The TiUnite implants with enhanced surface oxidaization were introduced to accelerate the bone apposition. This accelerated healing and establishment of a bone to implant contact was confirmed several times (Albrektsson et al. 2001, Gottlow et al. 2000, Rompen et al. 2000, Friberg & Billström 2002, Ivanoff et al. 2003, Rocci et al. 2003). The higher incidence of retrograde peri-implantitis in the present study for TiUnite implants can be explained as follows: when machined implants get into contact with a cyst or endosseous inflammation, they will soon be completely surrounded by granulation tissue, and subsequently lost. With the TiUnite surface, because of the accelerated bone apposition, the coronal part will integrate before the fibrous encapsulation starting at the tip can reach this area. Thus it is anchored in the bone although its apical part is surrounded by a granuloma or abscess. 126

137 GENERAL CONCLUSIONS 127

138 128

139 CHAPTER IX. General conclusions The vast number of consecutive patients and the homogeneity of the treatment hardware and software (chapter II) allow identifying a number of systemic and local factors which interfere with the osseointegration process in early stage, when the majority of the implants were of machined surface. Some previously identified factors for failure such as smoking, Crohn s disease, osteoporosis, vicinity with the natural dentition were all clearly confirmed. As a consequence, the indication for the use of oral implants should sometimes be reconsidered when alternative prosthetic treatments are available in the presence of interfering systemic or local factors. When an improved implant surface such as TiUnite was evaluated (chapter III), the early failure rate was so small (1.9%), that the analysis of interfering factors became difficult. It appears that poor bone quality did not influence the outcome as accertained by the Fisher test for osteoporosis, diameter. On the other hand gastric problems, Crohn, diabetes type I and radical hysterectomy seems to increase the incidence of early implant losses. Local factors seem predominant variables for the predictability of osseointegration like bone quality and radiotherapy on the late implant losses (chapter IV). Implant parameters such as wide diameter torque force and PTV are also relevant. Smoking and systemic health factors seem not prominent players in the etiology of late implant failures. Since bone quality affects the outcome in some instances (chapter II and chapter IV) it was logical to define whether preoperative assessment as is routinely done on radiographs of the jaws is a valid and discriminating method. Chapter IV evaluated this subjective assessment by relating it to several biomechanical measuring devices like Osseocare, Periotest and Osstell. It seems that these measurements related to the categories of the Lekholm & Zarb index. 129

140 An improved implant surface such as TiUnite also offers a better prognosis when a failed implant has to be replaced (chapter VI). The higher incidence of retrograde peri-implantitis and lower incidence of failure of TiUnite implants when compared to machined implants, suggests an accelerated healing and bone to implant contact for these implants. Retrograde peri-implantitis is provoked by remaining scar or granulomatous tissue at the recipient site: endodontic pathology of extracted tooth (scar tissue-impacted tooth) or possible endodontic pathology from a neighboring tooth (chapter VII). 130

141 GENERAL REFERENCES

142 132

143 General references Adell, R., Eriksson, B., Lekholm, U., Brånemark, PI. & Jemt, T. (1990) Long-term follow-up study of osseointegrated implants in the treatment of totally edentulous jaws. International Journal of Oral and Maxillofacial Implants 5, Albrektsson, T., Johansson, C., Lundgren, A., Sul, Y. & Gottlow, J. (2001) Experimental studies on oxidized implants. A histomorphometrical and biochemical analysis. Applied Osseointegration Research 1, Albrektsson, T. & Wennerberg, A. (2004) Oral implant surfaces: Part 2--review focusing on clinical knowledge of different surfaces. International Journal of Prosthodotics 17, Bain, C. A. & Moy, P. K. (1993) The association between the failure of dental implants and cigarette smoking. International Journal of Oral & Maxillofacial Implants 8: Bain, C. A. (1996) Smoking and implant failure--benefits of a smoking cessation protocol. International Journal of Oral and Maxillofacial Implants 11, Bain, C. A. (2003) Implant installation in the smoking patient. Periodontology , Becker W., Hujoel P.P., Becker B.E. &Willingham H. (2000) Osteoporosis and implant failure: an exploratory case-control study. Journal of Periodontology 4, Bhaskar, S. (1966) Periapical lesions - types, incidence and clinical features. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology & Endodontics 21, Brånemark, P-I., Adell, R., Breine, U., Hansson, B.O., Lindstrom, J. & Ohlsson, Ǻ (1969). Inta-osseos anchorage of dental prostheses. I. Experimental studies. Scandinavian Journal of Plastic and Reconstructive Surgery 3, Brånemark, P. I., Zarb, G.A. & Albrektsson, T. (1985) Tissue integrated prostheses. Osseointegration in clinical dentistry volume 13 p. 211 Chicago, Quintessence Publishing Co Inc. Buser, D., Schenk, R., Steinemann, S., Fiorellini, J., Fox, C. & Stich, H. (1991) Influence of surface characteristics on bone integration of titanium implants. A histomorphometric study in miniature pigs. Journal of Biomedical Materials Research 25, Calandriello, R., Tomatis, M., Vallone, R., Rangert, B. & Gottlow, J. (2003) Immediate occlusal loading of single lower molars using Brånemark System Wide-Platform TiUnite implants: an interim report of a prospective openended clinical multicenter study. Clinical Implant Dentistry and Related Research 1, Carlsson, L., Rostlund, T., Albrektsson, B. & Albrektsson, T. (1988) Removal torques for polished and rough titanium implants. International Journal of Oral & Maxillofacial Implants 3, Cochran, D.L., Buser, D., ten Bruggenkate, C.M., Weingart, D., Taylor, T.M., Bernard, J.P., Peters, F. & Simpson, J.P. (2002) The use of reduced 133

144 healing times on ITI implants with a sandblasted and acid-etched (SLA) surface: early results from clinical trials on ITI SLA implants. Clinical Oral Implants Research 13, Ekelund, J.A., Lindquist, L.W., Carlsson, G.E. & Jemt, T. (2003) Implant treatment in the edentulous mandible: a prospective study on Brånemark system implants over more than 20 years. International Journal of Prosthodontics 16, el Askary, A.S., Meffert, R.M. & Griffin, T. (1999) Why do dental implants fail? Part I. Implant Dentistry 8, Engquist, B., Bergendal, T., Kallus, T. & Linden, U. (1988) A retrospective multicenter evaluation of osseointegrated implants supporting overdentures. International Journal of Oral and Maxillofacial Implants 3, Esposito, M., Hirsch, J.M., Lekholm, U. & Thomsen, P. (1998) Biological factors contributing to failures of osseointegrated oral implants. (II). Etiopathogenesis. Eurupean Journal of Oral Sciences 3, Esposito, M., Thomsen, P., Ericson, L.E. & Lekholm U. (1999) Histopathologic observations on early oral implant failures. International Journal of Oral and Maxillofacial Implants 6, Friberg, B., Jemt, T. & Lekholm, U. (1991) Early failures in 4.,641 consecutively placed Brånemark dental implants, a study from stage 1 surgery to the connection of completed prostheses. International Journal of Oral and Maxillofacial Implants. 6, Friberg, B., Sennerby, L., Roos, J. & Lekholm, U. (1995) Identification of bone quality in conjunction with insertion of titanium implants. A pilot study in jaw autopsy specimens. Clinical Oral Implants Research. 6, Friberg, B., Sennerby, L., Gröndahl, K., Bergstrom, C., Back, T. & Lekholm, U. (1999a) On cutting torque measurements during implant placement, a 3-year clinical prospective study. Clinical Implant Dentistry and Related Research. 1, Friberg, B., Sennerby, L., Meredith, N. & Lekholm, U. (1999b) A comparison between cutting torque and resonance frequency measurements of maxillary implants. A 20-month clinical study. International Journal of Oral and Maxillofacial Surgery. 28, Friberg, B. & Billström, C. (2002) Preliminary results of a prospective multicenter clinical study on TiUnite implants. Applied Osseointegration Research 3, Gottlow, J., Henry, P., Tan, A., Allan, B., Johansson, C. & Hall, J. (2000) Biomechanical and histologic evaluation of the TiUnite and Osseotite implant surfaces in dogs. Applied Osseointegration Research 1, Green, T.L., Walton, R.E., Taylor, J.K. & Merrell, P. (1997) Radiographic and histologic periapical findings of root canal treated teeth in cadaver. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology & Endodontics 83, Hall, J. & Lausmaa, J. (2000) Properties of a new porous oxide surface on titanium implants. Applied Osseointegration Research 1,5-8. Helkimo, E., Carlsson, G.E. & Helkimo, M. (1977). Bite force and state of dentition. Acta Odontol Scand 35,

145 Henry, P., Tan, A., Allan, B., Hall, J. & Johansson, C. (2000) Removal torque comparison of TiUnite and turned implants in the greyhound dog mandible..appllied Osseointegration Research 1, Homolka P., Beer A., Birkfellner W., Nowotny R., Gahleitner A., Tschabitscher M. & Bergmann H (2002) Bone mineral density measurement with dental quantitative CT prior to dental implant placement in cadaver mandibles: pilot study. Radiology. 224, Huré, G., Donath, K., Lesourd, M., Chappard, D. & Basle, M. (1996) Does titanium surface treatment influence the bone-implant interface? SEM and histomorphometry in a 6-month sheep study. International Journal of Oral & Maxillofacial Implants 11, Ibanez, J.C., & Jalbout, Z.N. (2002) Immediate loading of osseotite implants: two-year results. Implant Dentistry 11, Ivanoff, C.J., Sennerby, L. & Lekholm, U. (1996) Influence of initial implant mobility on the integration of titanium implants. An experimental study in rabbits. Clinical Oral Implants Research 7, Ivanoff, C-J., Widmark, G., Johansson, C. & Wennerberg, A. (2003) Histological evaluation of the bone response to oxidised and turned titanium micro implants in human jawbone. International Journal of Oral & Maxillofacial Implants 3, Jacobs, R. (2003) Preoperative radiologic planning of implant surgery in compromised patients. Periodontology , Jaffin, R.A. & Berman, C.L. (1991) The excessive loss of Brånemark fixtures in type IV bone, a 5-year analysis. Journal of Periodontology. 62, 2-4. Johansson, B. Back, T. & Hirsch, J.M. (2004) Cutting torque measurements in conjunction with implant placement in grafted and nongrafted maxillas as an objective evaluation of bone density: a possible method for identifying early implant failures? Clinical Implant Dentistry and Related Research. 6,9-15. Johansson, P. & Strid, K. G. (1994) Assessment of bone quality from placement resistance during implant surgery. International Journal of Oral and Maxillofacial Implants. 9, Keller, E.E. (1997) Placement of dental implants in the irradiated mandible: a protocol without adjunctive hyperbaric oxygen. Journal of Oral & Maxillofacial Surgery 55, Kronström, M., Svensson, B., Erickson, E., Houston, L., Braham, P. & Persson, G.R. (2000) Humoral immunity host factors in subjects with failing or successful titanium dental implants. Journal of Clinical Periodontology 27, Kronström, M, Svenson, B, Hellman & M, Persson, GR. (2001) Early implant failures in patients treated with Brånemark System titanium dental implants: a retrospective study. International Journal of Oral and Maxillofacial Implants 16, Lazzara, R.J., Testori, T., Trisi, P., Porter, S.S. & Weinstein, R.L. (1999) A human histologic analysis of osseotite and machined surfaces using implants with 2 opposing surfaces. International Journal of Periodontics & Restorative Dentistry 19,

146 Lekholm, U. & Zarb, G. A. (1985) Patient selection and preparation. In: Brånemark, P-I., Zarb, G., Albrektsson, T. eds. Tissue integrated prosthesis: Osseointegration in clinical dentistry. p.199, Chicago, Quintessence publishing Co Inc. Lekholm, U., Gunn, J., Henry, P., Higuchi, K., Linden, U., Bergstrom, C. & van Steenberghe, D. (1999) Survival of the Brånemark implant in partially edentulous jaws: a 10-year prospective multicenter study. International Journal of Oral &Maxillofacial Implants 14, Lindquist, L.W., Carlsson, G.E. & Jemt, T. (1996) A prospective 15-year follow-up study of mandibular fixed prostheses supported by osseointegrated implants. Clinical results and marginal bone loss. Clinical Oral Implants Research 7, Marx, R.E. & Johnson, R.P. (1987) Studies in the radiobiology of osteoradionecrosis and their clinical significance. Oral Surgery Oral Medicine and Oral Patholology 64, Meredith, N. (1998) A review of nondestructive test methods and their application to measure the stability and osseointegration of bone anchored endosseous implants. Critical Reviews in Biomedical Engineering. 26, Mombelli, A. & Gionca, N. (2006) Systemic disease affecting osseointegration therapy. Clinical Oral Implants Research 17, Moy, P.K., Medina, D., Shetty, V. & Aghaloo, T. (2005) Dental implant failure rates and associated risk factors. International Journal of Oral & Maxillofacial Implants 20, Nevins, M. & Fiorellini, J. P. (1998) Placement of the maxillary posterior implants. In, Nevines M., Mellonig IT., editors. Implant therapy., clinical approaches and evidence of success, volume 2, p.153 Chicago, Quintessence. Olivé, J. & Aparicio, C. (1990) Periotest method as a measure of osseointegrated oral implant stability. International Journal of Oral and Maxillofacial Implants 4, Olsson, M., Urde, G., Andersen, J.B. & Sennerby, L. (2003) Early loading of maxillary fixed cross-arch dental prostheses supported by six or eight oxidized titanium implants: results after 1 year of loading, case series. Clinical Implant Dentistry & Related Research 1, Porter, J. A. & von Fraunhofer, J. A. (2005) Success or failure of dental implants? A literature review with treatment considerations. General Dentistry. 6, Quirynen, M., De Soete, M. & van Steenberghe, D. (2002) Infectious risks for oral implants: a review of the literature. Clinical Oral Implants Research 13, Quirynen, M., Gijbels, F. & Jacobs, R. (2003) An infected jawbone site compromising successful osseointegration. Periodontology , Rasmusson, L., Meredith, N., Kahnberg, K. E. & Sennerby, L. (1998) Stability assessments and histology of titanium implants placed simultaneously with autogenous onlay bone in the rabbit tibia. International Journal of Oral and Maxillofacial Surgery. 27, Reiser, G.M. & Nevins, M. (1995) The implant periapical lesion: etiology, prevention, and treatment. Compendium of Continuing Education in Dentistry 16,

147 Renouard, F. &, Nisand, D. (2006) Impact of implant length and diameter on survival rates Suppl Clinical Oral Implants Research 2, Rocci, A., Martignoni, M. & Gottlow, J. (2003) Immediate loading of Brånemark System TiUnite and machined-surface implants in the posterior mandible: a randomized open-ended clinical trial. Clinical Implant Dentistry and Related Research 5, Rompen, E., DaSilva, D., Lundgren, A.K., Gottlow, J. & Sennerby L. (2000) Stability measurements of a double-threaded titanium implant design with turned or oxidized surfaces. An experimental resonance frequency analysis study in the dog mandible. Applied Osseointegration Research 1, Rowe, A.H. & Binnie, W.H. (1974) Correlation between radiological and histological inflammatory changes following root canal treatment. Journal of the British Endodontics Society 7, Teerlinck, J., Quirynen, M., Darius, P. & van Steenberghe, D. (1991) Periotest, An objective clinical diagnosis of bone apposition toward implants. International Journal of Oral and Maxillofacial Implants. 1, Seltzer, S. (1999). Long-term radiographic and histological observations of endodontically treated teeth. Journal of Endodontics 25, Szmukler-Moncler, S., Salama, H., Reingewirtz, Y.& Dubruille, J. H. (1998) Timing of loading and effect of micromotion on bone-dental implant interface, review of experimental literature. Journal of Biomedical Materials Research. 43, Szmukler-Moncler, S., Piattelli, A., Favero, G.A. & Dubruille, J.H. (2000) Considerations preliminary to the application of early and immediate loading protocols in dental implantology. Clinical Oral Implants Research 11, Tricio, J., Laohapand, P., van Steenberghe, D., Quirynen, M.& Naert, I. (1995) Mechanical state assessment of the implant-bone continuum, a better understanding of the Periotest method. International Journal of Oral and Maxillofacial Implants. 101, van Steenberghe, D. (1989) A retrospective multicenter evaluation of the survival rate of osseointegrated fixtures supporting fixed partial prostheses in the treatment of partial edentulism. Journal of Prosthetic Dentistry 61, van Steenberghe, D, Lekholm, U, Bolender, C, Folmer, T, Henry, P, Herrmann, I., Higuchi, K., Laney, W., Linden, U. & Astrand, P. (1990) Applicability of osseointegrated oral implants in the rehabilitation of partial edentulism: a prospective multicenter study on 558 fixtures. International Journal of Oral and Maxillofacial Implants 53, van Steenberghe, D. & Quirynen, M. (1993) Reproducibility and detection threshold of peri-implant diagnostics. Advances in Dental Research. 7, van Steenberghe, D., Tricio, J., Naert, I. & Nys, M. (1995) Damping charachteristics of bone-to-implant interfaces. A clinical study with Periotest device. Clinical Oral Implants Research. 1, van Steenberghe, D., Yoshida, K., Papaioannou, W., Bollen, C., Reybroeck, G.I. & Quirynen, M. (1997) Complete nose coverage to prevent airborne contamination via nostrils is unnecessary. Clinical Oral Implants Research 8,

148 van Steenberghe, D., Quirynen, M., & Naert, I. (1999) Survival and success rates with oral endosseous implants. In: Lang, N.P., Attstrom, R. & Lindhe, J., eds. Proceedings of the 3de European Workshop on Periodontology, Berlin: Quintessence Publ. Co. Inc. van Steenberghe, D., Quirynen, M., Naert, I., Maffei, G. & Jacobs, R. (2001) Marginal bone loss around implants retaining hinging mandibular overdentures, at 4-, 8- and 12-years follow-up. Journal of Clinical Periodontolology 28, van Steenberghe, D., Jacobs, R., Desnyder, M., Maffei, G. & Quirynen, M. (2002) The relative impact of local and endogenous patient-related factors on implant failure up to the abutment stage. Clinical Oral Implants Research 13, van Steenberghe, D., Quirynen, M., Molly, L. & Jacobs, R. (2003) Impact of systemic diseases and medication on osseointegration. Periodontolology ,

149 CURRICULUM VITAE