Matthew Vierra Lian Ping Mau Guy Huynh-Ba John Schoolfield David L. Cochran

Authors’ affiliations: Matthew Vierra, Guy Huynh-Ba, John Schoolfield, David L. Cochran, Department of Periodontics, University of Texas Health Science Center at San Antonio (UTHSCSA), San Antonio, TX, USA Lian Ping Mau, Department of Dentistry, Chi Mei Medical Center, Tainan City, Taiwan

A lateral ridge augmentation study to evaluate a synthetic membrane for guided bone regeneration: an experiment in the canine mandible

Key words: animal study, barrier membrane, defect, dental implants, guided bone

regeneration Abstract Objectives: To evaluate guided bone regeneration outcomes in defects protected with an in situ formed polyethylene glycol (PEG) hydrogel membrane as compared to a non-cross-linked collagen

Corresponding author: David L. Cochran, DDS, MS, PhD Department of Periodontics The University of Texas Health Science Center at San Antonio 7703 Floyd Curl Drive, MSC 7894 San Antonio, TX 78229-3900 USA Tel.: +1 210 567 3604 Fax: +1 210 567 3643 e-mail: [email protected]

membrane (CM). Material and methods: Four mandibular alveolar ridge defects were created in eight hound dogs. Regenerative procedures were randomly allocated to one of four groups consisting of freeze-dried bone allograft, which is referred to in this study as freeze-dried bone xenograft (FDBX) + PEG, autogenous bone (AB) + PEG, AB + CM, and AB alone. After 8 weeks, titanium dental implants were placed into augmented sites. After 8 weeks of allowed time for osseointegration, the animals were sacrificed to harvest block specimens for bone-to-implant contact (BIC) and ridge width histomorphometric analysis. Results: Polyethylene glycol membranes had an exposure rate of 50% as compared to 12.5% for sites grafted with CM. Regenerative outcomes with respect to implant placement were least favorable for FDBX + PEG which had implants placed in 37.5% of augmented sites compared to 100% implant placement for all other groups. No statistically significant differences were noted between groups for ridge width measurements in implant and non-implant histologic sections (P > 0.05). Buccal BIC (%) values between treatment groups also failed to reach statistical significant difference (FDBX + PEG [60.2  9.4]; AB + PEG [58.8  8.5]; AB + CM [57.9  12.8]; AB [61.0  10.2]). Conclusion: When used in conjunction with FDBX, PEG had unpredictable bone formation and in most cases negatively impacted future implant placement.

Date: Accepted 16 September 2014 To cite this article: Vierra M, Mau LP, Huynh-Ba G, Schoolfield J, Cochran DL. A lateral ridge augmentation study to evaluate a synthetic membrane for guided bone regeneration: an experiment in the canine mandible. Clin. Oral Impl. Res. 00, 2014, 1–10 doi: 10.1111/clr.12517

Guided bone regeneration (GBR) procedures have enjoyed predictable success rates in localized horizontal alveolar ridge defects (H€ammerle & Karring 1998). Additionally, dental implants placed into defects augmented with GBR techniques have been shown to have survival rates comparable to those encountered with native bone (Donos et al. 2008). Non-resorbable membranes have traditionally been the mainstay for GBR procedures in this respect. However, their technique sensitive nature and inherently high complication rates have led to new avenues of research aimed at developing natural and synthetic resorbable membranes which are more benevolent in their soft tissue interactions (Simion et al. 1994; Zitzmann et al. 1997; McAllister & Haghighat 2007).

© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Introduction of collagen membranes for purposes of GBR illustrates how biologic materials have been designed to overcome limitations inherent to non-resorbable membranes. As a limitation of their own, collagen membranes may be deficient with regard to volume stability, an important requirement for barrier function in GBR procedures (Hardwick et al. 1995). A study investigating biodegradation of natural and synthetic membranes showed a decrease in material thickness of more that 50% for a non-crosslinked collagen membrane at 4 weeks following implantation (Herten et al. 2009). The development of synthetic polymers for use as barrier membranes in GBR has several advantages over natural collagen-based materials. Due to the fact that donor tissue is not required, these materials can be prepared in

1

Vierra et al  Lateral ridge augmentation

large quantities without supply limitations where chemical modification allows for membranes to be engineered with different mechanical and physical properties (Hutmacher et al. 1996). Polyethylene glycol (PEG) hydrogel is one such synthetic polymer that has been investigated as a barrier membrane for use in GBR procedures with promising results. Polyethylene glycol is a highly biocompatible polymer (Working et al. 1997) that has been widely researched for use in pharmacology, drug delivery, and wound covering (Zalipsky & Harris 1997). PEG molecules are capable of in situ hydrogel formation through cross-linking reactions that are supported at physiologic pH and readily degradable through a non-acidic hydrolytic process (Elbert et al. 2001). Modifications to pH of the PEG hydrogel matrix have been found to influence amount of new bone formation and PEG degradation time in mini-pigs, with a pH of 8.7 supporting the greatest amount of new bone formation (Thoma et al. 2014). Furthermore, a histologic study examining degradation and cell-occlusive properties of PEG in a rat model revealed that tissue in-growth into PEG hydrogels was delayed for greater than 4 months (Wechsler et al. 2008). The ability of PEG membranes to support bone formation has also been evaluated histologically in proof of principle experiments. A study analyzing prepared calvarial defects in rabbits showed that PEG membranes were as effective as non-resorbable membranes with respect to the amount of newly formed bone present within test sites at 4 weeks (Jung et al. 2006). PEG membranes were also proven to be successful with regard to space maintenance by preventing soft tissue collapse in mandibular defects created in minipigs, outperforming polylactic acid membrane and non-membrane groups in percent newly formed bone and inhibition of tissue in-growth (Thoma et al. 2009). Additionally, a pre-clinical dog study evaluating PEG for GBR and simultaneous implant placement showed similarities in new bone formation (31–38%), BIC (71–82%), and soft tissue interaction to a non-crosslinked collagen membrane (Jung et al. 2009b). Finally, a randomized controlled clinical trial in humans studying delayed or late implant placement with simultaneously preformed GBR showed that PEG had similar results for percent defect fill (PEG [94.9%]; collagen membrane (CM) [96.4%]) compared to a noncross-linked collagen membrane (Jung et al. 2009a). Clinical and pre-clinical studies have shown PEG to be as effective as a non-crosslinked collagen membrane for augmenting

2 |

Clin. Oral Impl. Res. 0, 2014 / 1–10

defects using GBR. However, it has also been suggested that a tendency for PEG membranes to rupture following polymerization may lead to soft tissue proliferation between graft particles and implant surfaces resulting in reduced BIC values (Zambon et al. 2012). Some questions still remain as to the ability of PEG to maintain barrier functions over time in long span lateral ridge defects up to 20 mm. Additionally, there is a need to understand how different grafting materials used in conjunction with PEG for GBR procedures compare to autogenous bone in terms of new bone formation determined histologically in clinical application. The performance of PEG has been studied as it pertains to implant osseointegration but has not yet been evaluated for efficacy in a standalone GBR procedure without implant placement. In light of previous PEG studies, it is our hypothesis that implementation of PEG in conjunction with either freeze-dried bone allograft, which will be referred to as freezedried bone xenograft (FDBX) in this study, or autogenous bone for purposes of GBR will result in similar amounts of new bone formation as compared to a non-cross-linked collagen membrane.

Material and methods Animals

Eight male hound dogs approximately 2 years of age and weighing 32–35 kg were quarantined prior to the start of enrollment in the study. The dogs were maintained on a soft diet throughout the experiment and socialization with husbandry staff and other dogs was

allowed when appropriate. Present study protocols pertaining to humane animal management were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Texas Health Science Center at San Antonio (UTHSCSA). Study design

A total of three surgeries were performed over the course of this study, which was 7 months in duration. Surgical extractions of second and third premolars (PM2-PM3) along with the first molar (M1) were accomplished bilaterally yielding a total of four treatment sites per animal (Fig. 1). Initial defects were prepared to approximate dimensions of 20 mm in length, 10 mm in height, and 5 mm in depth. After 12 weeks of healing, initial defects were refreshed to approximate dimensions of 15 mm in length, 8 mm in height, and 7 mm in depth. Four different GBR modalities were randomly assigned to the four treatment sites for each animal and consisted of the following (Fig. 2): FDBX + PEG (test group) AB + PEG (test group) AB + CM (+ control group) AB (- control group) At 8 weeks following GBR procedures, titanium dental implants were placed and obturated with closure screws. An additional 8 weeks of healing were allowed prior to animal sacrifice. Pre- and postoperative medication

All surgical procedures undertaken for this study were performed under sterile

(a)

(b)

Fig. 1. Buccal plate thinning, tooth sectioning (a), and extraction of PM2, PM3, and M1 with subsequent initial defect creation (b) to approximate dimensions of 20 mm long, 10 mm high, and 5 mm deep.

© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Vierra et al  Lateral ridge augmentation

(a)

(b)

(c)

(d)

Fig. 2. Four different treatment allocations are shown consisting of test group freeze-dried bone xenograft (FDBX) + polyethylene glycol (PEG) (a), test group autogenous bone (AB) + PEG (b), positive (+) control group AB + CM (c), and negative ( ) control group AB with no membrane (d). Note the adaptation to physiologic contour without overbulking of graft material.

conditions with aseptic surgical technique. All surgical procedures were performed under general anesthesia with preoperative administration of intravenous propofol (6 mg/kg, PropoFloâ, Abbott Laboratories, North Chicago, IL, USA), three ml of subcutaneous benzathine-penicillin + procainepenicillin G (300,000 U/ml, Pen Bp-48â, Pfizer Inc., Lee’s Summit, MO, USA), continued every 48 h for 7–10 days, and buprenorphine (0.01 mg/kg, Buprenexâ, Reckitt Benckiser Pharmaceuticals Inc., Richmond, VA, USA), continued twice per day for 48 h. The dogs were placed on a heating pad, intubated, and given isoflurane 1.5–3% (AErraneâ, Ohmeda, Liberty Corner, NJ, USA) while being monitored with an electrocardiogram. Maxillary and mandibular teeth were debrided of plaque and calculus and disinfected with providone-iodine solution/1% titratable iodine (Purdue Products L.P., Stanford, CT, USA) prior to the start of each surgery. Oral hygiene procedures were accomplished during induction anesthesia (propofol, 6 mg/kg) for suture removal using gauze soaked in 0.12% chlorhexidine gluconate (3M, St. Paul, MN, USA) at 7–10 days postoperatively. Local anesthetic (Lidocaine HCL 2% lidocaine with epinephrine 1 : 100,000, Henry Schein Inc., Port Washington, NY, USA) was administered to buccal and lingual areas of the surgical site.

Surgery 1 (Extractions)

Buccal and lingual mucoperiosteal flaps were reflected. Prior to sectioning and extraction, the buccal plates of PM2, PM3, and M1 were thinned. Initial defects were created under copious sterile saline irrigation with a Goldie burr (Brasseler USA, Savannah, GA, USA) by removing the buccal plate, inter-radicular septa, and medullary bone approximating extraction sockets (Fig. 1). Wound closure was accomplished using interrupted sutures (VicrylTM, Ethicon Inc., Somerville, NJ, USA). Surgery 2 (Defect modification and GBR)

At 12 weeks following tooth extractions, buccal and lingual mucoperiosteal flaps were reflected. Using a Goldie burr under copious sterile saline irrigation, the existing defects were refreshed. Defect height and depth were measured with a UNC 15 periodontal probe (Hu-Friedy, Chicago, IL, USA), and defect length was measured with a boley gauge (Miltexâ, Integra, York, PA, USA). The bucco-lingual dimension of the lingual plate was measured with a boley gauge at an area of consistent thickness approximately 5 mm below the lingual crest. Measurements were rounded to the nearest mm and recorded. Surgical sites were irrigated with sterile saline prior to grafting procedures. Autogenous bone was harvested from the posterior ramus and the inferior border of the mandible at a distance of at least 5 mm from

© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

the defect margins. Harvested bone was cortical in nature and particulated with double action rongeurs (Hu-Friedy) to a size of 1– 3 mm before being hydrated in the animal’s own blood. Mineralized ground cortical FDBX (Straumann AlloGraft GC, Straumann, Medford, MA, USA) was also hydrated using the subject’s blood for approximately 30 min prior to being placed into defect sites. AB and FDBX were placed into corresponding randomly assigned defects so as to recreate a natural physiologic contour without over filling. The use of PEG at designated treatment sites was accomplished by following the “Surgical Guidelines and Product Handling” recommendations published by Straumann (©Institute Straumann AG, 2011). Further recommendations were also followed as set forth by the MembraGelâ advisory board (December, 2010). PEG application kits (MembraGelâ, Institute Straumann AG, Waldenburg, Switzerland) were allowed to acclimatize to room temperature for 30 min prior to application and only one kit was used for each corresponding graft site. Defect sites receiving PEG were covered to an extent of 1–2 mm past the defect margins and up to the alveolar ridge crest without transitioning over it. Finally, the stability of the solidified PEG membrane was assessed by trying to displace it with light force from a periodontal probe. If the membrane became dislodged or was inconsistent in color, it was removed completely along with any grafting material and the process was repeated. For test sites receiving CM (BioGideâ, Geistlich Biomaterials, Wolhusen, Switzerland), the dimensions of the graft site were measured and the membrane was trimmed so that 1–2 mm overlapped the defect margins. The membrane was carried over the alveolar crest and onto the lingual plate approximately 3 mm to aid in stabilization. All mucoperiosteal flaps were deemed passive and free of muscular involvement in complete coverage of the grafting materials and no vertical or periosteal releasing incisions were made. Closure of the surgical site was accomplished with interrupted and 1–2 internal horizontal mattress sutures (W.L. Gore and Associates Inc., Flagstaff, AZ, USA). Surgery 3 (Implant placement)

At 8 weeks following GBR procedures, previously grafted sites were evaluated and dental implants were placed if possible. Buccal and lingual mucoperiosteal flaps were reflected and the implant was placed in the middle of the defect area with the lingual aspect of the implant adjacent to the remaining native

3 |

Clin. Oral Impl. Res. 0, 2014 / 1–10

Vierra et al  Lateral ridge augmentation

bone present at the time of defect creation. Alveoloplasty was not conducted prior to implant placement. Bony dehiscences present after implant placement were measured with a UNC 15 periodontal probe, rounded to the nearest 0.5 mm and recorded. Placement of one implant (Bone Levelâ Roxolidâ SLActiveâ, 3.3 mm diameter, 8 mm length, Institute Straumann AG, Waldenburg, Switzerland) for each grafted site was attempted. Implants were placed and obturated with closure screws prior to mucoperiosteal flap repositioning with interrupted sutures (W.L. Gore and Associates Inc., Flagstaff, AZ, USA). Animal sacrifice

At 8 weeks following implant placement procedures, all animals were sacrificed using intravenous propofol (6 mg/kg) and Euthasolâ (1 ml/10 lbs, Virbac Corp., St. Louis, MO, USA). Mandibular block-resection was accomplished along with removal of the inferior border of the mandible to aid in the diffusion of fixative throughout the specimen. The recovered block sections were placed in a solution of 10% neutral buffered formalin (StatLab, McKinney, TX, USA) prior to histologic preparation and analysis. Histologic preparation

Block specimens were infiltrated over a period of 2 weeks and hardened with light curing methacrylate (Technovit 7200 VLC, Kulzer & Co. GmbH, Wehrheim, Germany). Radiographs were made of the embedded block sections to aid in sectioning implants equally through the center in a bucco-lingual plane, which provided two specimens. Two additional specimens were obtained by sectioning sites mesial and distal to the implant in a bucco-lingual plane for a total of four specimens per augmentation site. Initial cuts were made using an irrigated diamond saw at low speed (Exakt Technologies Inc., Wehrheim, Germany). The sections were then fixed to plastic slides with acrylic cement (Exakt Technologies Inc.), ground, and polished using an irrigated rotary grinder (Exakt Technologies Inc.). Final thickness of the ground sections was 20–30 lm. Histologic slides were then stained with Sanderson’s Rapid Bone Stain (Dorn & Hart, Villa Park, IL, USA) and counterstained with acid fuchsin/1% (Poly Scientific, Bay Shore, NY, USA). Outcome measurements

The primary outcome parameter for this study is histologic ridge width measurement. Secondary outcome parameters include BIC and qualitative GBR assessment.

4 |

Clin. Oral Impl. Res. 0, 2014 / 1–10

Primary outcome (ridge width measurement)

Histometric quantification was accomplished using a light microscope (Vanox-T AH-2, Olympus, Tokyo, Japan) at 19 and 49 magnifications. Pictures were taken with an attached camera (Olympus DP71, Olympus) and CellSensâ software (Olympus Corp., Tokyo, Japan) was used for image analysis. The following histologic landmarks were identified: (Fig. 3a) apical extent of implant platform bevel (IB) and point of first bone-toimplant contact on the buccal implant surface (fBIC). Horizontal ridge width measurements were made (Fig. 3b) at fBIC (I-RWf0) and at 1 (I-RWf1), 3 (I-RWf3), and 5 (I-RWf5) mm apical to fBIC. The following measurements for nonimplant sites were recorded: (Fig. 3c) buccolingual horizontal ridge width at 1 mm (RW1), 3 mm (RW3), 5 mm (RW5), and 7 (RW7) mm apical to the ridge crest. Measurements were rounded to the nearest 0.1 mm.

(a)

(b)

Secondary outcome (BIC)

The following measurements for implant sites were recorded: fBIC to IB (fBIC-IB) (Fig. 3a), total implant length (IB to implant apex) buccal and lingual, and BIC buccal and lingual. BIC (%) on the buccal (BIC%[B]) and lingual (BIC%[L]) was calculated from respective measurements (BIC/total implant length 9 100). Measurements were rounded to the nearest 0.1 mm.

(c)

Secondary outcome (qualitative GBR assessment)

Quality of new bone formation at grafted sites was assessed and based on the ability to place a dental implant ideally within the bucco-lingual confines of the alveolar ridge. The GBR outcome scores used for this description are as follows (Fig. 4): 0: Implant placement not possible 1: Implant placement not ideal (buccal plate < 1 mm) 2: Implant placement ideal (buccal plate ≥ 1 mm)

Statistical analysis

The analysis of collected data was accomplished using SPSSâ (IBM, Armonk, NY, USA) version 19 statistical software. Data points for duplicate histologic slides were averaged to assess precision. Treatment response bias due to the location of test sites within the mandible was controlled for in the randomization process. Data analysis was performed using mixedmodel analysis of variance (ANOVA) to assess the treatment effect after controlling for any random variance attributable to the

Fig. 3. (a) Histologic section showing fBIC-IB which is measured from the apical extent of the implant platform bevel (IB) to the point of first bone-to-implant contact on the buccal implant surface (fBIC). Original magnification 940. (b) Histologic slide showing implant horizontal ridge width measurements at fBIC (I-RWf0) and at 1 (I-RWf1), 3 (I-RWf3), and 5 (I-RWf5) mm apical to fBIC and (c) non-implant horizontal ridge width measurements at 1 mm (RW1), 3 mm (RW3), 5 mm (RW5), and 7 (RW7) mm apical to the ridge crest (RC). Original magnification 910.

dog effect. Mixed-model ANOVAs were performed for BIC values, implant ridge width (I-RWf0-f5), and non-implant ridge width (RW1-7) data to identify parameters for which significant mean differences were observable

© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Vierra et al  Lateral ridge augmentation

test. Cochran’s Q test was also significant (P = 0.004) for the comparison of implant placement, with fewer implants placed for FDBX + PEG (37%) compared to all other groups (100%). Soft tissue encapsulation of FDBX at failed graft sites was confirmed histologically. Bone densities recorded during implant placement were characterized as type 2 or type 3 and implant insertion torques ranged from 5 to >35 Ncm. Bone types and torque values were equally represented in the four treatment groups.

(a)

(b) Descriptive histology FDBX + PEG

The histologic presentation of bony dehiscences present on the buccal surfaces of implants grafted with FDBX + PEG were all 0.05). Three membrane exposures were noted for FDBX + PEG, five for AB + PEG, and one for AB + CM. Membrane dehiscences observed in AB + PEG tended to be larger compared to exposures in FDBX + PEG. Membrane exposure did not appear to be correlated with lower GBR outcome scores (Fig. 5), and no attempt was made to remove membranes that had become exposed. All wound dehiscence complications resolved without further complication by 21 days following GBR procedures. Treatments differed significantly (P = 0.021) using Cochran’s Q test with respect to the presentation of wound dehiscence in AB + PEG and AB groups. Five membrane dehiscences were noted in the AB + PEG group as compared to no membrane dehiscences in the AB group. GBR outcome scores were significantly higher (P = 0.006) for the AB group as opposed to the FDBX + PEG group using the Friedman

© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Bony dehiscences on the buccal surfaces of implants grafted with AB + PEG were 1 mm (Fig. 8). However, three sites presented with BIC at the apical extent of the IB. Levels of lamellar bone formation at sites grafted with AB + CM were comparable to all other groups. As with the previously mentioned groups, connective tissue infiltration was randomly observed along buccal surfaces of alveolar bone for both implant and nonimplant sites. AB

Buccal dehiscences noted for implant sites grafted with AB were 0.05). Mean fBIC measurements (Table 3) were lowest for AB (0.5  0.7) and highest for AB + PEG (0.7  0.5). In the current data set, values closer to zero correspond to a more coronal position of buccal bone on the implant surface. Observed differences between groups (P > 0.05) in this respect were trivial and also statistically not significant (Table 3). Similar non-significant group differences were found for BIC%[B] (P > 0.05). Values for BIC%[B] ranged from 57.9% to 61.0% and were comparable to values obtained for BIC%[L] which ranged from 55.7% to 61.4% (Table 3).

c

Discussion

Fig. 5. Membrane dehiscence noted in the autogenous bone (AB) + polyethylene glycol (PEG) (a), freeze-dried bone xenograft (FDBX) + PEG (d), and AB + CM (g) groups. Corresponding sites 2 months following guided bone regeneration (GBR) procedures for AB + PEG (b), FDBX + PEG (e), and AB + CM (h) with implant GBR outcome scores of 2 for AB + PEG (c), FDBX + PEG (f), and AB + CM (i). A lack of correlation can be noted between membrane dehiscence and GBR outcome score for the examples provided.

(a)

(b)

Fig. 6. Ground sections of both implant placement (a) and non-implant placement (b) sites for the freeze-dried bone xenograft (FDBX) + polyethylene glycol (PEG) test group with horizontal ridge width areas corresponding to I-RWf0f5 and RW1-7 demarcated in yellow. Figures a & b represent adjacent sites within the same grafted defect. Original magnification 910.

sites showed BIC at the apical extent of the implant platform bevel. Levels of lamellar bone were comparable to all other groups and connective tissue infiltration was randomly observed along buccal surfaces of alveolar bone for both implant and nonimplant sites. Four samples appeared to have new bone formation that exceeded normal physiologic contour in the lateral dimension.

6 |

Clin. Oral Impl. Res. 0, 2014 / 1–10

Histomorphometric analysis

Histologic slides with sections taken through the implant were unable to be made for the AB + PEG group in one animal (n = 7). Histomorphometric data were collected for three implant sites (n = 3) in the FDBX + PEG group. For non-implant histologic sections, FDBX + PEG and AB + PEG each had data that were unable to be obtained for one animal (n = 7).

The aims of the present animal study were to evaluate and contrast the clinical and histologic performance characteristics of PEG membranes with non-cross-linked collagen membranes in regenerating lateral ridge defects using the principles of GBR. Another goal was to determine any effect that grafting materials may have on regenerative outcomes with and without implant placement into grafted sites. The use of dogs as test subjects in the current experiment is based on their acceptance as a beneficial conduit for modeling alveolar ridge defects to be treated with experimental therapies or products (Dumitrescu 2012). Although the present study did not show statistically significant differences in histomorphometric data pertaining to BIC and horizontal ridge width measurements, significant differences were observed with respect to clinical performance in PEG membrane groups which are not representative of prior research. The FDBX + PEG group had the least favorable results with respect to GBR outcome score and implant placement success. Membrane exposure, which has been shown to result in less new bone formation as compared to non-dehisced graft sites (Machtei 2001), may only partially account for the inadequate augmentations observed. One possible explanation has to do with graft material effect. Canine studies that best describe possible cross-species incompatibility of human allograft materials are limited to experiments using human-derived demineralized FDBX (Becker et al. 1995; Park et al. 2008). The

© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Vierra et al  Lateral ridge augmentation

(a)

(b)

Fig. 7. Ground sections of both implant placement (a) and non-implant placement (b) sites for the autogenous bone (AB) + polyethylene glycol (PEG) test group with horizontal ridge width areas corresponding to I-RWf0-f5 and RW1-7 demarcated in yellow. Figures a & b represent adjacent sites within the same grafted defect. Original magnification 910.

(a)

(b)

Fig. 8. Ground sections of both implant placement (a) and non-implant placement (b) sites for the autogenous bone (AB) + CM positive (+) control with horizontal ridge width areas corresponding to I-RWf0-f5 and RW1-7 demarcated in yellow. Figures a & b represent adjacent sites within the same grafted defect. Original magnification 910.

(a)

(b)

Fig. 9. Ground sections of both implant placement (a) and non-implant placement (b) sites for the autogenous bone (AB) negative ( ) control with horizontal ridge width areas corresponding to I-RWf0-f5 and RW1-7 demarcated in yellow. Figures a & b represent adjacent sites within the same grafted defect. Original magnification 910.

results of these studies did not suggest the presence of a deleterious effect associated with human allograft. Other non-human xenogenic grafting materials such as equine and bovine bone have been used in canine studies and were shown to have good biocompatibility (Schwarz et al. 2010). It appears more likely that any graft material effect would be

the result of inherent advantages present in vital autogenous bone as compared to nonliving, transplanted substitutes. Supporting this stance is a review of literature pertaining to lateral ridge augmentation techniques, which concludes that more favorable outcomes are achieved in cases where autogenous bone is used (Jensen & Terheyden

© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

2009). Additionally, the incorporation of a CM + FDBX group may have helped determine to what extent either FDBX or PEG negatively affected GBR outcomes. Polyethylene glycol membrane groups experienced exposure rates of 50% compared to 12.5% for the CM group. Most exposures were found in AB + PEG. These tended to be large (>10 mm) and resulted in partial or complete membrane exfoliation. This statistic does not correlate with published observations from two previous dog studies using PEG. In one study (Schwarz et al. 2012), saddle-type defects 10 mm in length and 8 mm in height were grafted with PEG and no instances of membrane exposures were noted. An 18% exposure rate was noted for PEG in another dog study (Jung et al. 2009b). These dehiscences seemed to be limited to slight uncovering of the implant closure screw. A human randomized controlled clinical trial (Jung et al. 2009a) had higher instances of PEG dehiscence (31%) that were described as small to moderate. Reports in the literature for both resorbable and non-resorbable membrane exposures in dogs range from 20% to 60% (Buser et al. 1995; von Arx et al. 2001; Oh et al. 2003), which would rank the current experience as high. High incidences of PEG exposure observed in the present study may also have to do with performance characteristics of PEG itself. PEG membranes have been shown to maintain cell-occlusive properties for up to 4 months after implantation (Wechsler et al. 2008). These results have been substantiated through additional studies in minipigs (Thoma et al. 2009) and rabbits (Jung et al. 2006). A potentially adverse effect relating to prolonged degradation of PEG is that trans-membrane angiogenesis can be inhibited for up to 16 weeks, whereas more readily resorbed collagen membranes show vascularization at 2 weeks post-implantation (Herten et al. 2009). Pericytes approximating blood vessels within connective tissue have been reported to give rise to osteogenic cells (Rickard et al. 1996), which could be beneficial for GBR if they assume an intimate association with a membrane or graft material. Although membrane permeability to cells or fluids has been shown to be an unnecessary prerequisite for new bone formation in the rat model (Zellin & Linde 1996), it is, however, beneficial in terms of wound stabilization, soft tissue integration, and prevention of soft tissue ingrowth (H€ammerle & Karring 1998). An impediment to transgression of nutritive components within blood across a membrane and to plasmosis in general may negatively

7 |

Clin. Oral Impl. Res. 0, 2014 / 1–10

Vierra et al  Lateral ridge augmentation

Table 1. Horizontal ridge width measurements at non-implant sites for RW1-7 in millimeters Group

n

RW1

FDBX + PEG AB + PEG AB + CM AB

7 7 8 8

4.6 5.9 5.6 5.7

RW3

   

1.1 1.4 0.6 1.7

7.4 8.7 8.2 8.4

RW5

   

1.1 1.9 1.2 1.8

8.9 10.5 9.5 9.9

RW7    

1.4 2.0 1.2 1.8

10.3 11.6 10.6 11.1

   

1.6 2.0 1.1 1.6

Values are mean  SD. RW1-7, total ridge width for non-implant sites at 1, 3, 5, and 7 mm apical to ridge crest. AB, autogenous bone; CM, collagen membrane; FDBX, freeze-dried bone xenograft; PEG, polyethylene glycol. *P < 0.05 using one-way ANOVA.

Table 2. Horizontal ridge width measurements at implant sites for I-RWf0-f5 in millimeters Group

n

I-RWf0

FDBX + PEG AB + PEG AB + CM AB

3 7 8 8

0.3 0.2 0.4 0.4

   

I-RWf1

0.2 0.1 0.2 0.6

0.9 1.2 1.2 1.3

   

I-RWf3

0.4 0.5 0.6 0.8

2.6 1.9 2.0 2.3

   

1.5 0.5 0.9 0.8

I-RWf5 3.0 2.9 2.9 3.0

   

1.9 0.6 1.4 1.0

Values are mean  SD. I-RWf0-f5, buccal ridge width for implant sites at fBIC and 1, 3, and 5 mm apical to fBIC. AB, autogenous bone; CM, collagen membrane; FDBX, freeze-dried bone xenograft; PEG, polyethylene glycol. *P < 0.05 using one-way ANOVA.

Table 3. Measurements at implant sites for fBIC in millimeters, BIC%[B], and BIC%[L] in % Group

n

fBIC-IB

FDBX + PEG AB + PEG AB + CM AB

3 7 8 8

0.6 0.7 0.7 0.5

   

0.3 0.5 0.6 0.7

BIC%[B] 60.2 58.8 57.9 61.0

   

9.4 8.5 12.8 10.2

BIC%[L] 61.4 58.4 55.8 59.1

   

7.1 18.8 11.4 16.9

Values are mean  SD. fBIC-IB, apical extent of implant platform bevel (IB) to point of first bone-toimplant contact on buccal (fBIC); BIC%[B], bone-to-implant contact (%) buccal; BIC%[L], bone-toimplant contact (%) lingual. AB, autogenous bone; CM, collagen membrane; FDBX, freeze-dried bone xenograft, PEG, polyethylene glycol. *P < 0.05 using one-way ANOVA.

impact the ability of flaps covering large defect expanses to survive. In addition to delayed tissue integration, PEG membranes have been observed to swell significantly as the result of osmotic forces (Jung et al. 2006) and display marginally greater inflammatory reactions as compared to collagen membranes (Jung et al. 2009b). The large defect size chosen for the current study, absence of tissue integration, propensity for membrane enlargement, and PEG associated inflammatory responses may have contributed to the observed mechanical destabilization and rupture of PEG membranes, resulting in loss of space maintenance and barrier function. In correlating the clinical observations of FDBX + PEG with histologic data, it should be mentioned that the sample size was smaller (n = 3) due to the fact that only three implants were placed. While the remaining five sites in this group did not have sufficient bone volume for implant placement, the GBR outcomes for these three implants were

8 |

Clin. Oral Impl. Res. 0, 2014 / 1–10

similar to other treatment groups when using the GBR outcome scores. This may account for the insignificant differences observed with respect to buccal ridge width measurements at implant sites. However, no statistically significant differences were found between groups with respect to ridge width at non-implant sites where sample sizes were roughly equivalent. This may be the result of high sample variance observed in the present study or perhaps the dogs themselves were excellent wound healers. It is also equally probable that additional bone formation occurred after failed graft sites were degranulated at implant placement and allowed to heal for an additional 8 weeks. The acute nature of the wounds created in the present study by defect refreshing could have contributed to bone fill over the additional healing time by increasing osteogenic potential in graft sites through activation of innate biological responses to injury (Schenk 1992).

With consideration to the role that regenerative membranes play in augmentation procedures, it should be mentioned that the AB + PEG group had equivalent GBR outcome scores with AB + CM despite having five times as many membrane exposures. Additionally, there were no statistically significant differences between AB, AB + PEG, and AB + CM with respect to GBR outcome score or RW1-7 and I-RWf0-f5 ridge width measurements. Thus, within the limitations of this study, it appears that the use of either PEG or CM as a barrier membrane was not noticeably advantageous in GBR. The target defect dimensions prescribed for this study (15 mm length 9 8 mm height 9 7 mm depth) were for implementation of the critical size defect model, which is necessary for understanding the true amount of new bone formation that can be supported with the use of a barrier membrane (Wikesj€ o & Selvig 1999). Although it seems unlikely that these dimensions would produce a non-critical defect, the absence of a group with untreated defects prevents any definitive conclusions. Similarities observed between AB + PEG and AB + CM, given the difference in membrane exposure rates, may also highlight the ability of autogenous bone to transcend adverse membrane healing outcomes. The AB + CM group, having only one exposure, performed the best in this respect. This finding would be in agreement with previous clinical studies that have found non-cross-linked BioGideâ membranes to be highly biocompatible and resistant to wound dehiscence and membrane exposure (Becker et al. 2009). In evaluating histomorphometric data for ridge width measurements with and without implant placement, no statistically significant differences were observed across all four treatment groups. The same was true for data pertaining to BIC values, which ranged from 57.9% to 61.0% on buccal implant surfaces and 55.7–61.4% on lingual implant surfaces. These values fall within the wide range (29–77%) of what has been reported in the literature for dog studies (Kohal et al. 1999; Botticelli et al. 2004). The SLActiveâ surface on the dental implants used in the current study has been reported to enhance BIC values as well as promote bone regeneration in dehiscence type defects (Schwarz et al. 2008). The SLActiveâ dental implants used in the current study failed to reach reported BIC values (82%) reported in a previous dog study utilizing the same SLActiveâ surface (Schwarz et al. 2007). This observed difference in BIC (%) might be attributed to inequalities in study design,

© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Vierra et al  Lateral ridge augmentation

specifically defect size and time allowed for osseointegration. Finally, failure to observe any statistically significant differences between treatment groups with respect to the histologic data presented above may have resulted from being underpowered. Due to the fact that the present experiment is a pilot study, expected group variances for the chosen treatments were unknown. A post hoc power analysis conducted on ridge with at RW1 with an overall standard deviation of 1.25 revealed that 20 dogs would be necessary to detect mean differences by ANOVA (P < 0.05) with a power of 80%. In conclusion, the use of FDBX in conjunction with PEG resulted in inadequate new bone formation that in most cases negatively

impacted future implant placement. Higher membrane exposure rates observed at PEG sites as well as the overall effect that CM and PEG membranes had on new bone formation were negligible in cases where autogenous bone was used as a graft material. Given the likelihood that the present study is underpowered, careful consideration should be given for extrapolating presented results to clinical settings.

Acknowledgements: The authors would like to thank and acknowledge the assistance of Sonja A. Bustamante for her expertise in preparation of histological specimins. We would also like to thank the staff at laboratory animal resources

UTHSCSA for their professionalism in managing the dog surgeries.

Source of funding This study was funded through a grant from the International Team for Implantology (ITI). Materials were generously provided for by Institut Straumann AG and Geistlich Biomaterials.

Conflict of interests The authors report no conflict of interests related to the present study.

References von Arx, T., Cochran, D.L., Hermann, J.S., Schenk, R.K. & Buser, D. (2001) Lateral ridge augmentation and implant placement: an experimental study evaluating implant osseointegration in different augmentation materials in the canine mandible. International Journal of Oral and Maxillofacial Implants 16: 343–354. Becker, J., Al-Nawas, B., Klein, M.O., Schliephake, H., Terheyden, H. & Schwarz, F. (2009) Use of a new cross-linked collagen membrane for the treatment of dehiscence-type defects at titanium implants: a prospective, randomized controlled double-blinded clinical multicenter study. Clinical Oral Implants Research 20: 742–749. Becker, W., Schenk, R.K., Higuchi, K., Leckholm, U. & Becker, B.E. (1995) Variations in bone regeneration adjacent to implants augmented with barrier membranes alone or with demineralized freeze-dried bone or autologous grafts: a study in dogs. International Journal of Oral and Maxillofacial Implants 10: 143–154. Botticelli, D., Berglundh, T. & Linde, J. (2004) The influence of a biomaterial on the closure of a marginal hard tissue defect adjacent to implants. An experimental study in the dog. Clinical Oral Implants Research 15: 285–292. Buser, D., Ruskin, J., Higginbottom, F., Hardwick, R., Dahlin, C. & Schenk, R.K. (1995) Osseointegration of titanium implants in membrane-protected defects: a histologic study in the canine mandible. International Journal of Oral and Maxillofacial Implants 10: 666–681. Donos, N., Mardas, N. & Chadha, V. (2008) Clinical outcomes of implants following lateral bone augmentation: systematic assessment of available options (barrier membranes, bone grafts, split osteotomy). Journal of Clinical Periodontology 35: 173–202. Dumitrescu, A.L. (2012) Animal models in periodontal research. In: Dumitrescu, A.L., ed. Understanding Periodontal Research, 655–771. New York, NY: Springer. Elbert, D.L., Pratt, A.B., Lutolf, M.P., Halstenberg, S. & Hubbell, J.A. (2001) Protein delivery from

materials formed by self-selective conjugate addition reactions. Journal of Controlled Release 76: 11–25. H€ammerle, C.H. & Karring, T. (1998) Guided bone regeneration at oral implant sites. Periodontology 2000 17: 151–175. Hardwick, R., Hayes, B.K. & Flynn, C. (1995) Devices for dentoalveolar regeneration: an up-todate literature review. Journal of Periodontology 66: 495–505. Herten, M., Jung, R.E., Ferrari, D., Rothamel, D., Golubovic, V., Molenberg, A., H€ammerle, C.H., Becker, J. & Schwarz, F. (2009) Biodegradation of different synthetic hydrogels made of polyethylene glycol hydrogel/RGD-peptide modifications: an immunohistochemical study in rats. Clinical Oral Implants Research 20: 116–125. Hutmacher, D., H€ urzeler, M.B. & Schliephake, H. (1996) A review of material properties of biodegradable and bioresorbable polymers and devices for GTR and GBR applications. International Journal of Oral and Maxillofacial Implants 11: 667–678. Jensen, S.S. & Terheyden, H. (2009) Bone augmentation in localized defects in the alveolar ridge: clinical results with different bone grafts and bone-substitute materials. International Journal of Oral and Maxillofacial Implants 24: 218–236. Jung, R.E., H€alg, G.A., Thoma, D.S. & H€ammerle, C.H. (2009a) A randomized, controlled clinical trial to evaluate a new membrane for guided bone regeneration around dental implants. Clinical Oral Implants Research 20: 162–168. Jung, R.E., Lecloux, G., Rompen, E., Ramel, C.F., Buser, D. & H€ammerle, C.H. (2009b) A feasibility study evaluating an in situ formed synthetic biodegradable membrane for guided bone regeneration in dogs. Clinical Oral Implants Research 20: 151–161. Jung, R.E., Zwahlen, R., Weber, F.E., Molenberg, A., van Lenthe, G.H. & H€ammerle, C.H. (2006) Evaluation of an in situ formed synthetic hydrogel as a biodegradable membrane for guided bone regeneration. Clinical Oral Implants Research 17: 426–433.

© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Kohal, R.J., Trejo, P.M., Wirsching, C., H€ urzeler, M.B. & Caffesse, R.G. (1999) Comparison of bioabsorbable and bioinert membranes for guided bone regeneration around non-submerged implants. An experimental study in the mongrel dog. Clinical Oral Implants Research 10: 226–237. Machtei, E.E. (2001) The effect of membrane exposure on the outcome of regenerative procedures in humans: a meta-analysis. Journal of Periodontology 72: 512–516. McAllister, B.S. & Haghighat, K. (2007) Bone augmentation techniques. Journal of Periodontology 78: 377–396. MembraGelâ advisory board (©Institute Straumann AG December, 2010). Oh, T.J., Meraw, S.J., Lee, E.J., Giannobile, W.V. & Wang, H.L. (2003) Comparative analysis of collagen membranes for the treatment of implant dehiscence defects. Clinical Oral Implants Research 14: 80–90. Park, J.H., Kwon, Y.H., Park, J.B., Chung, J.H., Shin, S.I. & Herr, Y. (2008) The effect of human demineralized freeze-dried xenograft on vertical bone formation in beagle dogs. The Journal of the Korean Academy of Periodontology 38: 75–82. Rickard, D.J., Kassem, M., Hefferan, T.E., Sarkar, G., Spelsberg, T.C. & Riggs, B.L. (1996) Isolation and characterization of osteoblast precursor cells from human bone marrow. Journal of Bone and Mineral Research 11: 312–324. Schenk, R.K. (1992) Biology of fracture repair. In: Browner, B.D., Jupiter, J.B., Levine, A.M. & Traften, P.G., eds. Skeletal Trauma, 31–75. Philadelphia, PA: Saunders. Schwarz, F., Ferrari, D., Balic, E., Buser, D., Becker, J. & Sager, M. (2010) Lateral ridge augmentation using equine- and bovine-derived cancellous bone blocks: a feasibility study in dogs. Clinical Oral Implants Research 21: 904–912. Schwarz, F., Herten, M., Sager, M., Wieland, M., Dard, M. & Becker, J. (2007) Bone regeneration in dehiscence-type defects at chemically modified (SLActive) and conventional SLA titanium

9 |

Clin. Oral Impl. Res. 0, 2014 / 1–10

Vierra et al  Lateral ridge augmentation

Implants: a pilot study in dogs. Journal of Clinical Periodontology 34: 78–86. Schwarz, F., Mihatovic, I., Golubovic, V., Hegewald, A. & Becker, J. (2012) Influence of two barrier membranes on staged guided bone regeneration and osseointegration of titanium implants in dogs: part I. Augmentation using bone graft substitutes and autogenous bone. Clinical Oral Implants Research 23: 83–89. Schwarz, F., Sager, M., Ferrari, D., Herten, M., Wieland, M. & Becker, J. (2008) Bone regeneration in dehiscence-type defects at non-submerged and submerged chemically modified (SLActive) and conventional SLA titanium implants: an immunohistochemical study in dogs. Journal of Clinical Periodontology 35: 64–75. Simion, M., Baldoni, M., Rossi, P. & Zaffe, D. (1994) A comparative study of the effectiveness of e-PTFE membranes with and without early exposure during the healing period. The International Journal of Periodontics & Restorative Dentistry 14: 166–180. Straumannâ MembraGelâ “Surgical guidelines and handling” (©Institute Straumann AG, 2011).

10 |

Clin. Oral Impl. Res. 0, 2014 / 1–10

Thoma, D.S., Halg, G.A., Dard, M.M., Seibl, R., H€ammerle, C.H. & Jung, R.E. (2009) Evaluation of a new biodegradable membrane to prevent gingival ingrowth into mandibular bone defects in minipigs. Clinical Oral Implants Research 20: 7– 16. Thoma, D.S., Schneider, D., Mir-Mari, J., H€ammerle, C.H., Gemperli, A.C., Molenberg, A., Dard, M. & Jung, R.E. (2014) Biodegradation and bone formation of various polyethylene glycol hydrogels in acute and chronic sites in mini-pigs. Clinical Oral Implants Research 25: 511–521. Wechsler, S., Fehr, D., Molenberg, A., Raeber, G., Schense, J.C. & Weber, F.E. (2008) A novel, tissue occlusive poly (ethylene glycol) hydrogel material. Journal of Biomedical Materials Research 85: 285–292. Wikesj€ o, U.M. & Selvig, K.A. (1999) Periodontal wound healing and regeneration. Periodontology 2000 19: 21–39. Working, P.K., Newman, M.S., Johnson, J. & Cornacoff, J.B. (1997) Safety of poly(ethylene glycol) and poly(ethylene glycol) derivates. In: Harris, J.M. & Zalipsky, S., eds. Poly(Ethylene

Glycol): Chemistry and Biological Applications, 45–57. Washington, DC: American Chemical Society. Zalipsky, S. & Harris, J.M. (1997) Introduction to chemistry and biological applications of poly(ethylene glycol). In: Harris, J.M. & Zalipsky, S., eds. Poly(Ethylene Glycol): Chemistry and Biological Applications, 1–13. Washington, DC: American Chemical Society. Zambon, R., Mardas, N., Horvath, A., Petrie, A., Dard, M. & Donos, N. (2012) The effect of loading in regenerated bone in dehiscence defects following a combined approach of bone grafting and GBR. Clinical Oral Implants Research 23: 591–601. Zellin, G. & Linde, A. (1996) Effects of different osteopromotive membrane porosities on experimental bone neogenesis in rats. Biomaterials 17: 695– 702. Zitzmann, N.U., Naef, R. & Sch€arer, P. (1997) Resorbable versus nonresorbable membranes in combination with Bio-Oss for guided bone regeneration. International Journal of Oral and Maxillofacial Implants 12: 844–852.

© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

A lateral ridge augmentation study to evaluate a synthetic membrane for guided bone regeneration: an experiment in the canine mandible.

To evaluate guided bone regeneration outcomes in defects protected with an in situ formed polyethylene glycol (PEG) hydrogel membrane as compared to a...
2MB Sizes 0 Downloads 13 Views