Volume 72 • Number 11

Case Series

1616

Background: Space-maintaining capacity, cell dis-clusive potential, and stability over time are crucialfactors to achieving sufficient bone augmentation withmembrane barriers. The case series presented hereassessed a new collagen barrier used in bone aug-mentation. Clinically, the healing pattern, especiallyin cases of secondary healing, was studied.

Methods: Soft tissue healing was documented byphotographs, and the size of the dehiscences calcu-lated by image analysis. The measurements were per-formed on digitized photographs. During reentry, bar-rier remnants were dissected and histologicallyevaluated.

Results: The mean value for dehiscences was 35.5mm2; all dehiscences healed within 4 weeks after theexposure became evident. The difference was statis-tically significant between the week 2 and week 6 vis-its ( P = 0.008) for each previously exposed site. Thehistologic observation of barrier remnants revealeddirect apposition of fibrous and bone tissues on themembrane surface.

Conclusion: In cases of membrane exposure, gin-gival dehiscences always disappeared in the follow-ing weeks without affecting the healing process. His-tologic results showed barrier stability over a 6-monthperiod, promoting bone regeneration. J Periodontol2001;72:1616-1623.

KEY WORDSGuided bone regeneration; collagen; membranes,barrier; space maintenance.

The principle of guided bone regeneration (GBR)has been successfully proven in various controlledanimal studies and clinical trials.1-3 The healing pat-tern involves all steps of de novo bone formationincluding blood clot formation, invasion by the osteo-progenitor cells, their differentiation into osteoblasts,and apposition of connective tissue which finally min-eralizes to form woven bone, later remodeled intolamellar bone.4

In the process of guided bone regeneration, mem-brane barriers prevent cells of the gingival flap thatlack osteogenic potential from colonizing the regen-erating site. Bone grafts and substitutes stabilize theblood clot and prevent the collapse of barrier mem-branes or soft tissues into a non–self-maintainingalveolar ridge defect, thus creating space for boneformation.5-7

It has been shown that bone defects filled with bloodonly and effectively separated from the gingival softtissue by a barrier will generate new bone.8 However,soft tissue should cover the barrier to protect the areafrom infection and inflammation. Healing without expo-sure of the barrier membrane and the grafting mate-rial–if present–is a crucial factor for the outcome ofGBR.9 Expanded polytetrafluoroethylene (ePTFE) iscommon in bone regeneration as a non-resorbable,bioinert membrane material. After the 6 to 8 monthsneeded for osseous healing, this barrier must beremoved. In cases of incomplete soft tissue closure,the membrane becomes exposed and dehisced areasaccumulate oral microorganisms over time, makingpremature retrieval of the barrier necessary. Manyauthors have reported the unpredictability of resultsin cases associated with an inflammatory processcaused by such dehiscences.8,9 Bioabsorbable barriermembranes were developed to achieve improved softtissue healing and avoid membrane retrieval.10,11 His-tologic analysis of absorbable test membranes (lac-tid/glycolid polymers and polyglactin 910) versusePTFE barriers and periosteum as control membranesrevealed a significantly reduced period of functionalstability for the test membranes. The volume of newaugmented bone was significantly higher for the con-

Observations on a New Collagen Barrier Membrane in 16Consecutively Treated Patients. Clinical and HistologicalFindingsAnton Friedmann,* Frank Peter Strietzel,† Burghard Maretzki,* Sandu Pitaru,‡

and Jean-Pierre Bernimoulin*

* Department of Periodontology and Synoptic Dentistry, School ofDentistry, Charité, Humboldt University Berlin, Germany.

† Department of Oral Surgery and Dental Radiology.‡ Department of Oral Biology, School of Dental Medicine, Tel Aviv

University, Tel Aviv, Israel.

0327_IPC_AAP_553351 11/5/01 4:45 PM Page 1616

J Periodontol • November 2001

Case Series

Friedmann, Strietzel, Maretzki, Pitaru, Bernimoulin 1617

trols, especially for ePTFE,and was consistent with thehistological findings.12,13

Collagen membranesrepresent another type ofabsorbable barrier. Whilethe synthetic polymers arebroken down by hydrolysis,collagen is degraded byspecific collagenolytic en-zyme activity. Membranebarriers made of porcinecollagen type I and type III§

were tested and found to beeffective in bone regenera-tion procedures.14,15 How-ever, their functional stabil-ity over a 6-month periodand withstanding capacityagainst bacterial collage-nases in cases with de-hisced healing patternshave never been investi-gated. A low capacity toresist bacterial degradationmay cause a prematureexposure of the regenerat-ing tissue and jeopardizesuccessful augmentation.

A new collagen barrier�

that is resistant to animaland bacterial collagenasewhen prematurely exposedis currently available. The barrier consists of purifiedtype I collagen cross-linked by a non-toxic naturalagent.

The purpose of this study was to evaluate soft tis-sue healing in prematurely exposed barriers and totest the effect of premature exposure on the successof lateral ridge augmentation.

MATERIALS AND METHODSAfter approval of the study protocol by the EthicsCommittee of Charité, Humboldt University Berlin,16 partially edentulous patients were consecutivelyenrolled into the study. The participants were patientsreferred to the clinic for implant therapy or from thepool of recall patients in the Department of Peri-odontology. Each patient received printed informationconcerning the procedure and signed a written con-sent. Patients showing at least one edentulous areawith an insufficient amount of bone laterally either in

§ BioGide, Geistlich Biomaterials, Wohlhusen, Switzerland.� Ossix, ColBar R&D Ltd., Ramat Hasharon, Israel.

the maxilla or in the mandible and requiring implant-supported prosthetic rehabilitation were included. Themean age was 45 years, ranging from 22 to 65 years(Table 1). Exclusion criteria were systemic diseases(e.g., diabetes mellitus), pregnancy, lactation,untreated periodontitis, smoking, non-compliance,and poor oral hygiene. Patients included were ran-domly assigned to 1 of the 2 surgeons involved in theclinical study.

A 2-stage procedure based on lateral bone aug-mentation prior to implant installation was chosen.The surgical protocol followed the guidelines estab-lished by Buser et al. and Langer and Langer.16,17

Midcrestal incisions in the mandible and paracrestalpalatal incisions in the maxilla were performed priorto the elevation of a full-thickness flap. Small perfo-

Table 1.

Comprehensive Data of Patients Treated With a New CollagenBarrier

Age 2 Weeks: 4 Weeks:Patient Gender (years) Indication Area Exposure Exposure

1 M 63 Edentulous space maxilla Anterior Yes No

2 F 62 Distal extension mandible Posterior Yes No

3 M 34 Single gap mandible Posterior Yes No

4 F 29 Single gap mandible Posterior Yes No

5 M 46 Distal extension maxilla Anterior Yes No

6 F 51 Distal extension mandible Posterior Yes Yes

7 M 36 Distal extension mandible Posterior Yes No

8 F 22 Edentulous space maxilla Anterior Yes No

9 F 36 Single gap maxilla Anterior Yes No

10 F 38 Edentulous space mandible Posterior Yes Yes

11 F 44 Distal extension mandible Posterior No No

12 F 47 Edentulous space mandible Posterior No No

13 M 65 Single gap maxilla Anterior No No

14 F 42 Single gap maxilla Anterior No No

15 F 63 Distal extension mandible Posterior No No

16 F 49 Single gap mandible Posterior No No

0327_IPC_AAP_553351 11/5/01 4:45 PM Page 1617

Volume 72 • Number 11

Case Series

1618 Observations on a New Collagen Barrier for GBR Procedure

rations in the cortical plate were prepared, and thedefects were laterally augmented with deproteinizedbovine bone mineral (DBBM)¶ enriched with thepatient’s own venous blood. A new bovine collagentype I barrier for bone regeneration� was utilized (Fig.1). By horizontally releasing the periosteum, com-plete coverage of mobilized flaps was achieved withtension free sutures (silk 3.0, Dexon 5.0) (Fig. 2).

Postoperative treatment included amoxicillin 750mg 3 times per day for 7 days (cephoral 1,000 mg2 times per day in cases of allergy), ibuprofen 400mg 4 times per day for 5 days, and chlorhexidinemouthrinses twice a day for 14 days.

Sutures were removed 2 weeks after surgery.Patients were reexamined at weeks 1, 2, and 4 andat months 3 and 6 after the first-stage treatment.Patients exhibiting a soft tissue dehiscence wereinstructed to use chlorhexidine gel 3 times a day onexposed areas and were reexamined weekly. Pho-tographs were taken in a 1�1 format to documentthe process of reepithelialization and calculate thedehisced areas. After 7 months of healing, second-stage surgery was performed and the membranes ortheir remnants removed for histological purposes. Atthat point, successful implant installation concludedthe study. Biopsies for histologic analysis were takenfrom macroscopically identified remnants that wereidentified during the reentry procedure.

Morphometry and Statistical AnalysisSpecial software# was utilized for quantitative analy-sis. Clinical photographs were digitized,** and eachimage was calibrated# by reproducible landmarks.Although the photographs were not standardized, itwas possible to calibrate each single image for eachexaminer by measuring the tooth adjacent to thedefect in width and height. To standardize the results,3 examiners performed the measurements 3 timeseach. The examiners were calibrated, and the land-marks for calibration of the images remained iden-tical.

Data per site and count were pooled to calculatethe mean value. The distribution of mean values wastested by graphic analysis (P-P-Plot) and Kol-mogorov-Smirnov-test. Statistical analysis was per-formed after the normal distribution of the values hadbeen tested. Due to the low number of samples, theWilcoxon test for paired comparison was used to esti-mate the differences between the mean value of thedehisced area after 2 and 4 weeks of healing.

Correlations were tested between the occurrenceof the dehiscence and anterior or posterior implantregion; dehiscence and gender; the dehisced areaand implantation site; and the dehisced area andpatient’s age. Correlations were estimated by cross-table analysis, calculating the correlation coefficientsfor nominal scaled parameters (Ψ, Cramer-V, con-tingency coefficient, uncertainty coefficient). Statis-tical calculations were carried out using a statisticalsoftware package.††

Figure 1.The new collagen barrier membrane (M) covers the DBBM-augmented area without additional fixation.The barrier’s textureprevents micromovements under the flap (Table 2, patient 5).

Figure 2.Full, tension-free, soft tissue closure is achieved by releasing theperiosteum.

¶ BioOss, cancellous granules 0.25 to 1.0 mm in size, GeistlichBiomaterials, Baden-Baden, Germany.

# Scion Image for Windows, Scion Corporation, Frederick, MD.** Windows 95, Microsoft Corporation, Redmond, WA.†† SPSS version 9.0, SPSS Inc., Chicago, IL.

0327_IPC_AAP_553351 11/5/01 4:45 PM Page 1618

J Periodontol • November 2001

Case Series

Friedmann, Strietzel, Maretzki, Pitaru, Bernimoulin 1619

HistologySamples of the barrier and adjacent tissue obtainedduring the reentry surgery were fixed in formalin (4%)and embedded either in paraffin or glutarmetacry-late.‡‡ Sections were prepared by a sledge micro-tome§§ in paraffin blocks and rotation microtome inglutarmethacrylate blocks§§ and routinely stainedwith hematoxylin and eosin. Qualitative analysisfocused on the stability and integrity of the removedmaterial and the collagen structure.

RESULTSAll 16 patients completed the study. Different pat-terns of soft tissue healing following membrane place-ment (stage 1 surgery) were observed. Six out of 16

sites healed uneventfully. Ten sites experienced a flapdehiscence accompanied by barrier exposure duringthe initial healing period. In all 10 cases, the dehis-cences occurred within the first 14 postoperative days,prior to or immediately after suture removal (Fig. 3).Healing by secondary epithelialization was achievedin all exposed sites. Seven out of 10 dehiscenceswere completely closed within the following 14 days(Figs. 4 through 6). In one of the remaining 3 sites,the entire exposed area was completely covered bythe new gingiva, but some deproteinized bovine bonematerial (DBBM) granules were partially exfoliated.

Figure 3.Patient 5. Dehiscence occurred prior to suture removal at day 14postoperatively.

Figure 4.Patient 5 demonstrates complete soft tissue healing 2 weeks later.

‡‡ Technovit 8100, Kulzer, Bensheim, Germany.§§ Leitz, Wetzlar, Germany.

Figure 5.Three-month visit.The keratinization of the new tissue increased inthe previously exposed area.

Figure 6.Improved shape of maxillary ridge after successful augmentationutilizing the collagen barrier and deproteinized bovine bone materialat reentry. Seven months of healing, patient 5.

0327_IPC_AAP_553351 11/5/01 4:45 PM Page 1619

Volume 72 • Number 11

Case Series

1620 Observations on a New Collagen Barrier for GBR Procedure

However, this case was considered to be completelyclosed, because there was no exposed membranesurface after 4 weeks (Fig. 7). The other 2 patientsexhibited exposed barriers at the 4-week examination(Table 2). One of these exposed areas increased fromweek 2 to week 4. However, 14 days later at the 6-week appointment, these 2 sites were completelycovered by new gingiva. When dehiscences werepresent, neither the soft tissues nor the barriersshowed any clinical signs of inflammation or degra-dation, or any signs of swelling, redness, or exuda-tion. This finding remained constant throughout theperiod of exposure.

The overall percentage of exposure was 62%.Although the percentage was higher in the maxillathan in the mandible (71% versus 55%, respectively),no specific pattern of distribution could be detected.The exposure frequencies were equally distributedbetween the 2 surgeons involved in the treatment.The complete gingival healing revealed statisticallysignificant differences between week 2 and week 6 inall 10 previously exposed sites (P = 0.008; Wilcoxonnon-parametric test). No significant correlations weredetected between dehiscence and implant region (i.e.,anterior or posterior, P = 0.726) or dehiscence andgender (P = 0.330). The size of the dehiscence didnot correlate with either the site of implantation (P =0.179) or the patient’s age (P = 0.372).

All 16 patients presented improved dimensions inthe augmented region at the reentry for second-stagesurgery. Fifteen patients revealed sufficient mature,new bone formation for adequate implant insertion.

In one patient, the buccal half of the augmented sitewas not mineralized and required an additional aug-mentation procedure. Solid titanium plasma-sprayedimplants� � were inserted in a transmucosal manneraccording to the manufacturer’s protocol. A total of30 implants were installed; 6 of them were single-tooth implants.

Morphometric ResultsThe mean exposed surface area of the exposed bar-riers at 2 weeks was 35.55 mm2 (Table 2). In 2 out of10 cases, the exposed surface was not covered bysoft tissue at 4 weeks. In one of these cases, theexposed area decreased by 40% (patient 10), whereasit increased by 4 times in the other case (patient 6).Nevertheless, both cases exhibited complete soft tis-sue coverage at 6 weeks.

HistologyThe histological aspects of the barrier remnantsrevealed a dense, homogeneous fibrous structure sim-ilar to fibrocartillage. The fibrils were oriented in a

Figure 7.Exfoliation of some DBBM material (arrows). After a new gingivalayer completely closed the dehiscence, no exposed fragments of thebarrier were visible (Table 2, patient 8).

�� ITI Dental Implant System, ITI Straumann, Freiburg in Breisgau, Germany.

Table 2.

Measurements of Dehisced Areas andSpecification of Biopsies

Dehiscences (mm2 ± SD)*Biopsy After

Healing After Healing After 7 MonthsPatient 2 Weeks 4 Weeks of Healing

1 50.37 ± 2.87 Complete +

2 1.17 ± 0.26 Complete –

3 146.97 ± 19.95 Complete –

4 20.52 ± 1.17 Complete –

5 40.59 ± 25.48 Complete –

6 9.30 ± 2.25 39.12 ± 5.08 +

7 9.98 ± 1.43 Complete –

8 32.77 ± 6.87 Complete +

9 4.87 ± 0.66 Complete +

10 38.96 ± 2.18 22.72 ± 0.57 +

Mean 35.55 ± 42.65

* All sites showed statistically significant improvement at the 4- or 6-weekappointment, respectively (P = 0.008; Wilcoxon test for non-parametricpaired analysis).+ biopsy obtained; – no biopsy.

0327_IPC_AAP_553351 11/5/01 4:45 PM Page 1620

J Periodontol • November 2001

Case Series

Friedmann, Strietzel, Maretzki, Pitaru, Bernimoulin 1621

parallel manner, and no cells inside the barriers couldbe detected at any magnification. The presence oferythrocytes aligning collagen material was likelyrelated to the surgical dissection of the biopsies. Nosigns of cell infiltration inside the barriers or in theperiphery were visible. No differences in appearancewere observed between membranes with and withoutexposure.

Barrier degradation could not be attributed tomacrophages. It is possible that bacteria adheredto the exposed membrane surfaces; however, nogranulocytic infiltration was observed at 7 months(Fig. 8).

Tissue apposition and ingrowth into the gapsbetween the collagen layers occurred independentlyof membrane exposure. Connective tissue with col-lagen fibers and elongated fibrocytes, as well as bonedeposits, was revealed by undecalcified sections(Figs. 8 and 9). In some barrier remnants, DBBMparticles were detected between 2 collagen layers(Fig. 10). Active osteoclasts were located on the tra-beculae of the woven bone, possibly digesting rem-nants of DBBM (Figs. 11 and 12).

DISCUSSIONThe new collagen membrane was tested as a suit-able barrier for guided bone regeneration. At reen-try, a lateral tissue augmentation of the alveolar ridgewas observed in all operated patients. When mem-brane exposure occurred, new gingiva covered thedehisced areas within 14 days after suture removal

in the majority of the cases. Two patients experi-enced a longer period of barrier exposure. While onepatient showed a decrease in the size of the dehiscedarea 4 weeks postoperatively, the second patientshowed an increase (Table 2). However, these 2 siteswere completely healed 2 weeks later; i.e., 6 weeksafter surgery. In both cases, the barriers maintainedtheir macroscopic integrity during the healing period.In general, collagen fibrils can be degraded intra-and extracellularly; while fibroblasts can do both,bacterial collagenases act extracellularly.18 It is con-ceivable that the dehisced areas were exposed tooral bacteria and their collagenolytic enzymes. In

Figure 8.Apposition and ingrowth of fibrous connective tissue into the gapbetween 2 layers of the collagen membrane (arrows). Erythrocytesdocument surgical dissection of the biopsies.Any cells that wouldindicate an inflammatory reaction are totally absent. (Biopsy frompatient 8; original magnification ×30).

Figure 9.Apposition of new cortical bone (B) in direct contact with the barrier(M) surface, despite previous exposure to the oral cavity. (Patient 6;original magnification ×130).

Figure 10.DBBM particles among collagen layers of the barrier after 7 monthsof healing. Osteoclast (arrows) (original magnification ×250).

0327_IPC_AAP_553351 11/5/01 4:45 PM Page 1621

Volume 72 • Number 11

Case Series

1622 Observations on a New Collagen Barrier for GBR Procedure

Figure 12.Bone formation (arrows) on the barrier’s surface document a highbiocompatibility of the membrane (M) material, demonstrating itsstability during 7 months of healing (original magnification ×130).

this respect, it is noteworthy that barrier remnantscould be retrieved at 7 months’ reentry in 50% of theexposed barriers. Moreover, the possible enzymaticdegradation in vivo seems not to have affected theprotective role of the barrier. In fact, all dehiscedareas healed completely, and the temporary expo-sure did not interfere with the ongoing processes ofmineralization and maturation. This was demon-strated by the bone apposition in close contact to thebarrier.

There are only a few published studies focusingon the stability of bioabsorbable or degradable bar-riers in humans. Simion et al. found no statisticallysignificant difference in bone fill in fenestration ordehiscence-type defects when PLA/PGA barriers

(tests) or ePTFE barriers (controls) were used.11

However, the authors stated that the barrier effectlasting longer than 4 to 6 weeks could be more ben-eficial for the outcome of GBR. Using PLA/PGA bar-riers for the same indication, Lundgren et al. pre-sented a series of cases, where they observed noremnants of the membranes after 6 to 7 months ofhealing.10 Another controlled study compared theaugmentation results with an ePTFE barrier versusa collagen membrane.15 DBBM was the filler mate-rial for both. Although no statistically significant dif-ferences were found in cases with uneventful heal-ing, the measurements of volumetric gain in caseswith barrier exposure revealed a significant advan-tage for collagen membranes. In that study, implantinstallation was combined with a defect augmenta-tion. Bone defects (dehiscence or fenestration type)occurring buccally or lingually from the crest of thealveolar ridge due to implant insertion are, accord-ing to their shape, mainly self-retaining. The appli-cation of a membrane barrier in self-retaining defectshas been controversial.14 Zitzmann et al. claimedthat in addition to the barrier effect, the defect mor-phology was a decisive factor for hard tissue for-mation. Furthermore, they reported an advanceddegradation of the collagen barrier in cases of incom-plete wound closure.15 These results differ from thoseof the present study. All defects in our study werenon–self-supporting; thus, the new collagen barrierhad to be not only cell occlusive, but also providea permanent mechanical stability over a 6-monthperiod. Our data support the permanent stability of

Figure 11.Osteoclasts (OC), possibly digesting remnants of DBBM, do notinterfere with the barrier surface which is located underneath thelayer of new bone formation (original magnification ×250).

0327_IPC_AAP_553351 11/5/01 4:45 PM Page 1622

J Periodontol • November 2001

Case Series

Friedmann, Strietzel, Maretzki, Pitaru, Bernimoulin 1623

the new barrier up to 6 months. Furthermore, thecomplete absence of any signs of inflammation, evenin patients who experienced barrier exposure, histo-logically favors the beneficial effect of the new bar-rier on soft tissue healing. Finally, the collagen andbone apposition visible on the membrane remnantsdocuments this barrier’s high grade of biocompati-bility.

ACKNOWLEDGMENTSThis study was supported by ColBar R&D Ltd., RamatHasharon, Israel.

REFERENCES1. Buser D, Dula K, Hirt HP, Schenk R. Lateral ridge aug-

mentation using autografts and barrier membranes: Aclinical study with 40 partially edentulous patients. JOral Maxillofac Surg 1996;54:420-432.

2. Berglundh T, Lindhe J. Healing around implants placedin bone defects treated with Bio-Oss. An experimentalstudy in the dog. Clin Oral Implants Res 1997;8:117-124.

3. Fiorellini J, Engebretson S, Donath K, Weber H. Guidedbone regeneration utilizing expanded polytetrafluo-roethylene membranes in combination with submergedand nonsubmerged dental implants in beagle dogs. JPeriodontol 1998;69:528-535.

4. Hämmerle C, Karring T. Guided bone regeneration atoral implant sites. Periodontol 2000 1998;17:151-175.

5. Buser D, Hoffmann B, Bernard JP, Lussi A, Mettler D,Schenk RK. Evaluation of filling materials in mem-brane-protected bone defects. A comparative histo-morphometric study in the mandible of miniature pigs.Clin Oral Implants Res 1998;9:137-150.

6. Hall EE, Meffert RM, Hermann JS, Mellonig JT, CochranDL. Comparison of bioactive glass to demineralizedfreeze-dried bone allograft in the treatment of intra-bony defects around implants in the canine mandible.J Periodontol 1999;70:526-535.

7. Hämmerle CHF, Chiantella GC, Karring T, Lang NP.The effect of a deproteinized bovine bone mineral onbone regeneration around titanium dental implants.Clin Oral Implants Res 1998;9:151-162.

8. Lang N, Hämmerle C, Brägger U, Lehmann B, NymanS. Guided tissue regeneration in jawbone defects priorto implant placement. Clin Oral Implants Res 1994;5:92-97.

9. Nowzari H, Slots J. Microscopic and clinical study ofpolytetrafluoroethylene membranes for guided boneregeneration around implants. Int J Oral MaxillofacImplants 1995;10:67-73.

10. Lundgren D, Sennerby L, Falk H, Friberg B, Nyman S.The use of a new bioresorbable barrier for guided boneregeneration in connection with implant installation.Case reports. Clin Oral Implants Res 1994;5:177-184.

11. Simion M, Misitano U, Gionso L, Salvato A. Treatmentof dehiscences and fenestrations around dentalimplants using resorbable and nonresorbable mem-branes associated with bone autografts: A comparativeclinical study. Int J Oral Maxillofac Implants 1997;12:159-167.

12. Wiltfang J, Merten H-A, Peters J-H. Comparative studyof guided bone regeneration using absorbable and per-manent barrier membranes: A histologic report. Int JOral Maxillofac Implants 1998;13:416-421.

13. Aaboe M, Pinholt EM, Schou S, Hjorting-Hansen E.Incomplete bone regeneration of rabbit calvarial defectsusing different membranes. Clin Oral Implants Res1998;9:313-320.

14. Hockers T, Abendsur D, Valentini P, Legrand R, Häm-merle CHF. The combined use of bioresorbable mem-branes and xenografts or autografts in the treatment ofbone defects around implants. A study in beagle dogs.Clin Oral Implants Res 1999;10:487-498.

15. Zitzmann NU, Naef R, Schärer P. Resorbable versusnonresorbable membranes in combination with Bio-Oss for guided bone regeneration. Int J Oral Maxillo-fac Implants 1997;12:844-852.

16. Buser D, Dula K, Belser UC, Hirt H-P, Berthold H.Localized ridge augmentation using guided bone regen-eration. II. Surgical procedure in the mandible. Int JPeriodontics Restorative Dent 1995;15:10-29.

17. Langer B, Langer L. The overlapped flap: A surgicalmodification for implant fixture installation. Int J Peri-odontics Restorative Dent 1990;10:208-215.

18. Ten Cate AR. Cytoskeleton, junctions and fibroblasts.In: Ten Cate AR, ed. Oral Histology: Development,Structure and Function, 5th ed. St Louis: The MosbyCompany; 1998:50-69.

Send reprint requests to: Dr. Anton Friedmann, Depart-ment of Periodontology and Synoptic Dentistry, School ofDentistry, Charité, Humboldt University Berlin, Augusten-burger Platz 1, 13353 Berlin, Germany. Fax: 49 30 450562931; e-mail: anton.friedmann@charite.de.

Accepted for publication May 4, 2001.

0327_IPC_AAP_553351 11/5/01 4:45 PM Page 1623