Erwin Van der Zee Paul Oosterveld Marinus A.J.

Van Waas

Effect of GBR and xture installation on gingiva and bone levels at adjacent teeth

Authors afliation: Erwin Van der Zee, Paul Oosterveld, Marinus A.J. Van Waas, Clinic for Oral Implantology, Department of Oral Function, Academic Centre for Dentistry (ACTA), 1066, Amsterdam, the Netherlands Correspondence to: Dr Erwin Van der Zee Clinic for Oral Implantology Department of Oral Function Academic Centre for Dentistry (ACTA) Louwesweg 1 NL-1066 EA Amsterdam The Netherlands Tel.: 31 2051 88412 Fax: 31 2051 88414 e-mail: e.vd.zee@ACTA.NL

Key words: adjacent teeth, bone resorption, GBR, gingival recession, randomised controlled trial Abstract: Guided bone regeneration (GBR) is frequently used in oral implantology. It is unclear to what extent GBR affects the periodontium of adjacent teeth. Therefore, the present study quanties changes in the proximal gingiva and bone levels at these teeth in 30 patients. Staged surgery involved a standard GBR treatment, randomly using resorbable membranes with a bone substitute or non-resorbable membranes with or without a bone substitute, followed by xture installation at 6 months and abutment connection a further 6 months later. The data were sampled at each surgery and analysed using MANOVA. Twelve months after GBR, there was on average a small but statistically signicant amount of proximal gingival recession (0.75 mm) and bone resorption (0.34 mm) observed, of which 50% was the result of GBR surgery. No signicant differences were found between the different GBR treatment modalities. It is concluded that GBR treatment may have a small negative effect on the levels of the free gingival margin and alveolar bone at adjacent teeth, which is in most patients not clinically relevant.

Date: Accepted 15 May 2003 To cite this article: Van der Zee E, Oosterveld P, Van Waas MAJ. Effect of GBR and xture installation on gingiva and bone levels at adjacent teeth. Clin. Oral Impl. Res. 15, 2004; 6265

Copyright r Blackwell Munksgaard 2004

Guided bone regeneration (GBR) is a surgical treatment modality in oral implantology derived from the guided tissue regeneration (GTR) principle in periodontology (reviewed by Karring et al. 1997). In short, a membrane is used to create a space around a local bone defect or dehisced or fenestrated implant surface to enable new bone formation and separate it from the gingiva in order to prevent soft connective tissue ingrowth (Dahlin et al. 1988; Nyman & Lang 1994). It is unclear if and to what extent the periodontium of adjacent teeth is affected by GBR. On one hand, it may be stabilised or additionally improved, and on the other hand, it may be compromised by the trauma of performing (multiple) surgical interventions, which may result in gingival recession and bone resorption. Only a few data are published about

the negative effects of GTR on adjacent teeth (Bra gger et al. 1992; Nygaard & Ostby 1996), and no information is available in this respect concerning GBR. The aim of the present study was to monitor changes in the gingiva and bone level at adjacent teeth following each surgical step in a GBR treatment and to evaluate in a randomised controlled clinical trial whether different GBR treatment modalities using resorbable membranes with a bone substitute or non-resorbable membranes with or without a bone substitute affect these parameters.

Materials and methods

The study sample consisted of 30 patients who applied for implants at the Clinic for

62

Van der Zee et al . Effect of GBR on adjacent gingiva and bone levels

Oral Implantology of ACTA (Academic Centre for Dentistry, Amsterdam) for replacement of one or more missing teeth in the maxilla (front and/or pre-molar region). Based on a thorough intra-oral, periodontal and radiological screening, the inclusion criteria were local bone defect class III according to Seiberts classication (Seibert 1983; combination buccolingual and apicocoronal loss of tissue resulting in loss of normal height and width) preventing proper implant positioning with primary stability, adjacent teeth with probable pocket depths smaller than 4 mm and good oral hygiene. Exclusion criteria were periodontal disease and smoking. The study was approved by the Medical Ethical Committee of the faculty (Academic Medical Centre (AMC), Amsterdam). After informed consent was obtained, the patients were allocated using a balancing procedure (Zielhuis et al. 1990), which was aimed at an equal distribution of the patients over three treatment groups regarding the administered balancing criteria of gender, age (until 39 years of age/40 years and older) and region of the maxilla (front/ pre-molar). One group was treated with a bovine xenograft (BioOsss, Geistlich AG, Wolhusen, Switzerland) covered with a fast resorbable polylactide/polyglycolide membrane (Resoluts, W.L. Gore and Ass., Flagstaff, AZ, USA), one group was treated with a bovine xenograft (BioOsss) covered with a non-resorbable titanium-reinforced (TR) expanded-polytetrauoroethylene (e-PTFE) membrane (Gore-Texs, W.L. Gore and Ass., Flagstaff), and one group was treated with a non-resorbable TR ePTFE membrane (Gore-Texs) covering a blood cloth only. A three-stage surgical protocol was used involving GBR as stage one, xture installation as stage two and abutment connection surgery as stage three, each with a 6month healing interval. Flap design for both installation and removal of the membranes involved a full thickness mucoperiosteal ap with the incision slightly palatal to the top of the crest, followed by a sulcular incision vestibular of the adjacent teeth stopping just before the interdental papilla, to be continued by two vestibular vertical releasing incisions. All surgeries were performed by one surgeon well experienced with GBR and GTR techniques. A standard GBR protocol was performed involving

preparation of the membrane according to a prepared trial mould in order to cover the defect properly, if possible, with a minimum of 3 mm overlap, keeping 12 mm distance between the membrane and the neck of the adjacent teeth. In order to produce a wound surface with a blood cloth, multiple perforations of the cortex were made with a round drill under irrigation with sterile saline. Maintenance of space underneath the membrane was secured by bending the titanium skeleton in case of the reinforced membranes. Before application, the xenograft was mixed with locally obtained full blood. After proper adaptation, the membranes were immobilised by tacking two spikes (Frios, Friatec, Mannheim, Germany) vestibular (apically) and one spike palatal. The ap was mobilised by a thorough releasing incision of the periosteum at the vestibular base and primary wound closure was obtained with minimal tension, by vertical mattress sutures with e-PTFE (CV-5, Gore Tex, W.L. Gore and Ass. Inc., Flagstaff). The GBR protocol was essentially similar as described by Jovanovic & Nevins (1995). All other surgical steps were performed according to the nemark system staged protocol of the Bra (Adell et al. 1981) using Mk II self-tap nemark system, Nobel Biocare xtures (Bra AB, Gothenburg, Sweden). Patients were observed frequently with a minimum interval of 1, 2, 4, 8 and 16 weeks following each surgery. Temporary partial prostheses were not allowed until removal of the sutures 2 weeks following GBR or xture installation surgery, and all prostheses were carefully adjusted and relined with soft tissue conditioner to avoid pressure at the surgical area. Amoxicillin was prescribed for 1 week (3 500 mg) and Corsodyls mouthrinse (2 10 ml) for 2 weeks postoperatively. Periodontal probe recordings of the gingiva and alveolar bone levels were made just prior to and during each surgical session, respectively, relative to the cemento-enamel junction (CEJ) at the midproximal site of both teeth adjacent to the edentulous space. These recordings were performed with the same surgical set including a Williams probe employing no standardisation for recordings and estimated clinically to 0.5 mm by the same periodontist. The data were analysed with multivariate analysis of variance (MANOVA). In

the analyses, the bone and gingiva measurements are the dependent variables; site (left or right adjacent tooth) and staged surgery are within-subject factors. The effect of different GBR treatments was investigated by running the MANOVA with a between-subjects factor (a 0.05). Z2 was used to evaluate the size of the effects. Pairwise comparison between the two intervals - (interval 1: 6 months after GBR surgery; interval 2: 6 months after xture installation) was performed to test the effect of GBR surgery and the xture installation procedure separately.

Results
Just prior to the GBR treatment, the interproximal gingival margin at teeth adjacent to the edentulous space was situated on average at the level of the CEJ, and the bone margin 3.1 mm below the CEJ. The mean changes in these levels occurring between GBR, xture installation, and abutment connection are shown in Table 1. The data of one patient were partially missing during evaluation, and therefore this patient was excluded from statistical analysis. Overall MANOVA indicates a highly statistically signicant effect of the staged surgery on the gingiva and bone levels (F2,27 11.76, P 0.00). This effect differs between both tissues (F2,27 3.56, P 0.04): the amount of gingival recession is on average higher than the decrease in bone level. There is no difference between the two proximal sites (S1 and S2) of the left and right adjacent tooth (F1,28 0.00, P 0.95), and no interaction between surgery and site (F2,27 0.77, P 0.47). Pairwise comparison showed a statistically signicant effect on both tissues at interval 1 (P 0.02) and interval 2 (P 0.00). There is also a statistically signicant within-subjects effect of the staged surgery on the gingiva for the total interval (F2,27 14.69, P 0.00) and for both separate intervals (interval 1: P 0.01; interval 2: P 0.00) as well as for bone levels (F2,27 3.92, P 0.03). The observed bone resorption at the separate intervals was not signicant (P 0.4, 0.07, respectively). Comparing the results of the three different GBR treatment modalities, no differences were found: neither between the gingiva

63 |

Clin. Oral Impl. Res. 15, 2004 / 6265

Van der Zee et al . Effect of GBR on adjacent gingiva and bone levels

Table 1. Descriptive statistics of changes in gingiva and bone level between GBR surgery and xture installation (interval 1), between xture installation and abutment connection (interval 2), and between GBR and abutment connection (total) in mm
Mean Gingiva level Interval 1 Interval 2 Total Bone level Interval 1 Interval 2 Total SD P Minimum Q1 Median Q3 Maximum

0.34 0.41 0.75 0.11 0.23 0.34

0.70 0.50 0.80 0.83 0.68 0.73

0.01 0.00 0.00 NS NS 0.03

2.75 1.50 2.75 1.75 2.00 2.00

0.75 0.75 1.25 0.50 0.75 0.75

0.38 0.50 0.75 0.25 0.25 0.50

0.06 0.25 0.25 0.25 0.13 0.00

0.75 0.75 1.50 1.75 1.25 1.50

(F2,26 0.37, P 0.70) nor between the bone levels (F2,26 0.95, P 0.40). Z2 showed that only 3% of the observed variance in gingiva level and only 7% of the observed variance in bone level were due to the GBR treatment modality. However, MANOVA indicated highly signicant differences for both gingiva (F2,25 15.64, P 0.00) and bone levels (F2,25 4.02, P 0.03) between the staged surgeries for each group. For both gingiva and bone levels, there were no interactions between the groups and staged surgeries (F4,52 1.03, P 0.40, and F4,52 1.24, P 0.31, respectively).

Discussion
The hypothesis of the present study was that GBR treatment could be performed without substantial damage to healthy adjacent teeth. Twelve months after GBR and 6 months after xture installation, a statistically signicant but small amount of both gingival recession and bone resorption is found at the interproximal sites of adjacent teeth bounding the edentulous space. Considering the overall effect on gingiva and bone levels, each surgical step has a comparable effect. The gingival recession is found to a similar degree of on average 0.4 mm following each type of surgery. Thus, both the instalment of the GBR material and the instalment of the xture have a comparable small, but statistically signicant, detrimental effect on the level of the free gingival margin. The inuence on the interproximal bone level of the adjacent teeth was too small for a signicant effect by each surgical step alone, but an overall signicant decrease in bone height of on average 0.34 mm was noticed following both the surgeries. The study shows that the small negative effect of repeated surgery on the mean

gingiva level was signicantly higher than on the mean bone level, which suggests that the former is more sensitive to ap surgery than the latter. This is an important aspect for the aesthetic result of the treatment, which is primarily associated with the margin of the gingiva. In this clinical study with 29 subjects, no differences in proximal gingival recession and bone resorption were found between the three different GBR treatment modalities. Thus, the type of membrane (fast resorbable or non-resorbable titanium-reinforced) and the use of a xenograft (blood cloth only versus BioOsss) seem not to have affected the amount of gingival recession and bone resorption at adjacent teeth. In this respect, it is important to emphasise that the same ap design and reection was used at rst- and second-stage surgery for each GBR group. Thus, no information is obtained about the amount of bone resorption and gingival recession found after onestage procedures, using only resorbable GBR materials and simultaneous xture installation. The effect of surgery on average gingiva and bone levels was similar and still statistically signicant for each GBR group of 10 patients in spite of the limited effect. The nding that gingival recession and bone resorption occur following GBR seems in contradiction with the aim of a regenerative technique. However, it is interesting to note that even in GTR studies aiming at (partial) repair of gingival recessions or infrabony pockets (Karring et al. 1997; Laurell & Gottlow 1998) improved attachment levels are often found after 612 months in combination with gingival recessions around the neck of the treated teeth (Lekovic et al. 1989; Bra gger et al. 1992; Nygaard-Ostby et al. 1996). We are not aware of any GBR studies investigating this aspect at adjacent teeth. These teeth

are not planned to participate actively in a regenerative treatment, but merely happen to be in the close vicinity of the bony defect in the edentulous zone. The observed mean bone resorption of 0.34 mm and gingival recession of 0.75 mm (of which only on average 50% is directly related to the GBR stage) are clinically not relevant for most patients. Although a quartile experienced no gingival recession following GBR and only 0.25 mm recession following the xture installation surgery, another 25% of the patients showed a mean of 0.75 mm gingival recession after each surgical step and a mean of 1.25 mm after both the surgical interventions. For patients in the latter group, the observed effect may be clinically relevant if present in combination with a high smile line, especially as only partial edentulous areas in the aesthetically critical zone of the frontal/pre-molar region of the maxilla were included in the study. It should be mentioned, however, that improvement of the interdental papilla during abutment (micro)surgery with the possibility of papilla regeneration techniques as well as the positive effects of a welldesigned superstructure on the interdental area (Tarnow et al. 1992; Blatz et al. 1999) may (partially) counteract the observed small detrimental effect on the level of the interdental gingiva as well as on the overall aesthetic appearance. In this respect, it is also important to emphasise the large improvement on aesthetics by regenerating bone in the defect zone with GBR, and thus making proper installation of implants and superstructure in the edentulous space possible in the rst place. It is concluded that GBR treatment may have a small negative effect on the levels of the free gingival margin and alveolar bone at adjacent teeth. Although it would be ideal if future modications of the GBR technique would exclude this, in most patients the

64 |

Clin. Oral Impl. Res. 15, 2004 / 6265

Van der Zee et al . Effect of GBR on adjacent gingiva and bone levels

clinical relevance of the presently found adverse effect is low.

pas du tout important du point de vue clinique chez la plupart des patients.

Acknowledgements: All members of the staff of the Clinic of Oral Implantology of ACTA are gratefully acknowledged for their interest in the present study. We specially thank Wieneke Middeldorp, Neeltje Visser and Jacqueline de Cleen for their assistance during clinical treatment sessions, and Hanneke Comis, Francis Niessen and Tineke Fortuin for their support in secretarial work.

Zusammenfassung
Der Einuss der GBR und der Implantatplatzierung auf die Gingiva und die Knochenho he am benachbarten Za hnen Die gesteuerte Knochenregeneration (GBR) wird in der oralen Implantologie ha ug angewendet. E ist unklar, in welchem Masse die GBR das Parodont der Nachbarza hne beeintra chtigt. Daher quantiziert die vorliegende Studie bei 30 Patienten die Vera nderungen in der approximalen Gingiva und im Knochenniveau an solchen Za hne. Die gestaffelten chirurgischen Massnahmen bestanden zuerst aus einer Standard GBR Behandlung, bei welcher zufa llig eine resorbierbare Membran mit einem Knochenersatzmaterial oder eine nicht resorbierbare Membran mit oder ohne Knochenersatzmaterial verwendet wurde. 6 Monate spa ter folgte darauf die Implantatplatzierung und wiederum 6 Monate spa ter wurde die Distanzoperation durchgefu hrt. Bei jedem chirurgischen Eingriff wurden Daten aufgenommen und mittels Manova analysiert. 12 Monate nach der GBR konnte im Mittel eine kleine aber statistisch signikante Rezession der approximalen Gingiva (0.75mm) und eine Knochenresorption (0.34mm) beobachtet werden. 50% der Verluste waren auf die GBR Chirurgie zuru ckzufu hren. Zwischen den verschiedenen GBR Verfahren konnten keine statistisch signikanten Unterschiede gefunden werden. Es wird die Schlussfolgerung gezogen, dass die GBR Behandlung einen kleinen negativen Einuss auf die Ho he des Randes der freien Gingiva und auf den Alveolarknochen an benachbarten Za hnen haben ko nnte. Bei den meisten Patienten ist dies aber klinisch nicht relevant.

sume Re
ge ne ration osseuse guide e (GBR) est fre quemLa re e en implantologie buccale. Limpact de ment utilise la GBR sur le parodonte des dents adjacentes nest tabli. Cest pourquoi, le tude pre pas clairement e sente quantie les variations de la gencive proximale et des niveaux osseux de ces dents chez 30 patients. tape a inclus un traitement GBR Une chirurgie par e standard, au hasard une utilisation de membrane sorbable avec un substitut osseux ou de membrane re sorbable avec ou sans substitut osseux, suivi non-re ` six mois et de la du placement de limplant a es connexion du pilier six mois plus tard. Les donne ` chaque chirurgie et analyse te recueillies a es en ont e `s la GBR, il y utilisant le Manova. Douze mois apre avait en moyenne une faible mais statistiquement de re cession gingivale intersignicative quantite sorption osseuse (0,34 proximale (0,75 mm) et de re mm) desquelles 50% provenait de la chirurgie GBR. rence signicative na e te trouve e entre Aucune diffe rents modalite s de traitement GBR. Le les diffe traitement GBR pourrait donc avoir un petit effet gatif sur les niveaux de la gencive marginale libre ne olaire des dents adjacentes mais qui nest et los alve

donde la GBR afecta al periodonto y a los dientes vecinos. Por ello, el presente estudio cuantica los a y hueso proximal cambios en los niveles de la enc a por fases en estos dientes en 30 pacientes. La cirug ndar, usando un tratamiento de GBR esta incluyo aleatoriamente membranas reabsorbibles con un seo o membranas no reabsorbibles con o sustituto o seo, seguida por la instalacio n de la sin sustituto o n a los 6 meses y la conexio n del pilar otros 6 jacio s. Los datos se tomaron en cada cirug a meses despue y se analizaron usando Manova. Doce meses tras la a existio de media una pequen scirug a pero estad n gingival ticamente signicativa cantidad de recesio n o sea (0.34 mm) de la proximal (0.75) y de reabsorcio a GBR. No se cual el 50 fue el resultado de la cirug encontraron diferencias signicativas entre las diferentes modalidades de tratamiento de GBR. Se concluye que el tratamiento de GBR puede tener un pequen o efecto negativo sobre los niveles del margen libre gingival y del hueso alveolar en los a de los dientes adyacentes que en la mayor pacientes no es relevante.

Resumen
n o sea guiada (GBR) es usada frecuenLa regeneracio claro hasta a oral. No esta temente en implantolog

References
nemark, Adell, R., Lekholm, U., Rockler, B. & Bra P.I. (1981) A 15 year study of osseointegrated implants in the treatment of the edentulous jaw. International Journal of Oral Surgery 10: 387 416. Blatz, M.B., Hurzeler, M.B. & Strub, J.R. (1999) Reconstruction of the lost interproximal papilla. Presentation of surgical and nonsurgical approaches. International Journal of Periodontics and Restorative Dentistry 4: 395406. Bra gger, U., Ha mmerle, C.H.F., Mombelli, A., Bu rgin, W. & Lang, N.P. (1992) Remodelling of periodontal tissues adjacent to sites treated according to the principles of guided tissue regeneration (GTR). Journal of Clinical Periodontology 19: 615624. Dahlin, C., Linde, A., Gottlow, J. & Nyman, S. (1988) Healing of bone defects by guided tissue regeneration. Plastic and Reconstructive Surgery 81: 672676. Jovanovic, S.A. & Nevins, M. (1995) Bone formation utilizing titanium-reinforced barrier membranes. International Journal of Periodontics and Restorative Dentistry 1: 5669. Karring, T., Cortellini, P. & Lindhe, J. (1997) Regenerative periodontal therapy. In: Lindhe, J., Karring, T. & Lang, N.P., eds. Clinical periodontology and implant dentistry, 597646. Copenhagen: Munksgaard. Laurell, L. & Gottlow, J. (1998) Guided tissue regeneration update. International Dental Journal 48: 386398. Lekovic, V., Kenney, E.B., Kovacevic, K. & Carranza, F.A. (1989) Evaluation of guided tissue regeneration in class II furcation defects. A clinical re-entry study. Journal of Periodontology 60: 694698. Nygaard-Ostby, P., Tellefsen, G., Sigurdsson, T.J., Zimmerman, G.J. & Wikesjo , U.M. (1996) Periodontal healing following reconstructive surgery: effect of guided tissue regeneration. Journal of Clinical Periodontology 23: 10731079. Nyman, S.R. & Lang, N.P. (1994) Guided tissue regeneration and dental implants. Periodontology 2000 4: 109118. Seibert, J.S. (1983) Reconstruction of deformed, partially edentulous ridges using full thickness onlay grafts. Part I, technique and wound healing. Compendium of Continuing Education in Dentistry 4: 437453. Tarnow, D.P., Wagner, A.W. & Fletcher, P. (1992) The effect of the distance from the contact point to the crest of bone on the presence or absence of the interproximal dental papilla. Journal of Periodontology 63: 995996. Zielhuis, G.A., Straatman, H., Vant Hof-Grootenboer, A.E., Van Lier, H.J., Rach, G.H. & Van den Broek, P. (1990) The choice of a balanced allocation method for a clinical trial in otitis media with effusion. Statistics in Medicine 9: 237246.

65 |

Clin. Oral Impl. Res. 15, 2004 / 6265