./ din Perioduntoi iWft. 23. filf 620 Printed in Denmark if! rights reserved Human histologic evaluation of HTR polymer and freeze-dried bone allograft A case report Copyright S Munksgaard IWf> JOUHNA1Dl ISS.\ 0303-6979 Stuart J.Froum Department of Surgical Sciences (Periodontics) and Implants. New York University Dental Center. 345 E. 24th Street. New York. NY 10010-4099. USA Frown SJ: Human histologic evaluation of HTR polymer and freeze-dried bone allograft. A cast report. J Clin Periodontoi IW6; 23. 615-620. © Munksgaard. 1996. Abstract. This ca^e report compares flic results emplacement of HTR™ (Hard Tissue Replacement Synthetic Bone™) and freeze-dried bone allografl (FDBA) in the same patient. 2 notches were made in each of 6 teeth, I at the gingival margin and the other at the most apical level of calculus. Soft tissue reponses with both materials included probing depth reduction, gingival shrinkage and gain in clinical attachment. Histologies! sections of block extraction sites 30 months after placement of either material showed no signs of new attachment. Little or QO inflammation was present with both materials. There was also an absence of all FDBA particles which assumes complete resorption of this material prior to 30 months. Gingival shrinkage resulted in the exposure of the gingival and calculus notch in all FDBA treated siles. Gingival epithelium was found adjacent to ihe gingival notch in all HTR™ treated sites. The calculus notch in HTR™ treated sites was lined by junctions! epithelium with connective tissue and bone opposing the adhesion. HTRrM particles were present and surrounded by connective tissue or bone. HTRIM appears to serve as a scaffold for new bone formation when in close contact with alveolar bone. Key words: human histology: HTR: freeze-dried bone allograft Accepted tor publication 3 August 1995 Various bone replacement grafts have been used to facilitate restoration of bone and, in some instances, periodonlal support tost to periodontal disease. Reviews by Froum & Gome/ (1993) and Garrett & Bogle (1994) support the use of various bone graft and bone substitute materials to treat such defects, the latter noting that "conclusive evidence now exists that some (issue regeneration occurs after regenerative procedures, in which bone grafts arcused". Hancock il989) documented this with autogeneous and allogenic bone grafts in the management of vertical type osseous defects. Freeze dried bone allografl I-DBA and HTR™ (Hard Tissue Replacement) Synthetic BoneIN1 alloplast (Bioplant Inc.. New York. NY) are 2 materials for which positive clinical results have been reported. Sepe et al. (1978), Sanders ct al. (1983) have described clinical results with KDBA alone and in combination with autogenous hone. In reviewing the results. Mellonig el al. (1991) reported 50% to 100% bone fill in 220 of 327 test sites treated with FDBA and surgically re-entry 6 months or more later. Rumelhart el al. (1989) found no statistically significant difference in clinical bone fill, levels of probing deplh reduction or gain in attachment in a comparison of FDBA with decalcified fiee/e dried bone allografl (DFDBA). Although conclusive histologic evidence of regeneration following use of DFDBA has been presented (Bowers et al 1989a. 1989b). no such studies with FDBA have been reported. Treatment of human infrabony and furcation defects with HTR1 Nl have shown favorable clinical results. Yukna (1991) compared HTRIVI with open flap debridement in 21 patients using 6-month re-entry data. HTR™ resulted in greater mean defect fill, (2.2 mm versus 1.00 mm), decreased probing depth I.V2 mm versus 2.3 mm) and increased gain in clinical attachment (1.9 mm versus 1.0 mm) versus the flap debridemem control. More recently. Yukna (1994) compared HTR™ to autogenous osseous eoagulum (AOC) at 15 paired mandibular Class II furcations in 9 subjects and reported improved results with HTR1 M in enhancing horizontal furcation fill (1.9 mm versus 0.8 mm) and in % of defect fill (44.4"/,, versus 17.1%). Improved clinical results with HTR™ are however, not universal. In a 12-month. 15-palient clinical evaluation of bilateral infrabony defects treated with either open flap debndement or HTR™ polymer grafts. Shahmiri et al. 616 I'nmnt (tW2l round no statistical significant differences in pocket responses. While human histologic studies have demonstrated bone formation around HTR™ panicles (Yukna & Greer I1M2. Suzki ei al. I989), Stalil et al. (1990) have reported a "gamut of healing respon-^ in human block sections taken after 4 to 26 weeks. HTR™ panicles were surrounded lor the mosl parl b> dense collagen with peripheral bone formal ion seen only occasionally. C linical closure consisted a ol' long epithelial adhesion at 7 sites and limited new attachment in 4 instances. The present case report describes the 30-month post-healing clinical and histological outcomes of the use ol" FDBA or HTRIM al 6 intraosseous defects in tin- same indi- •. idual. Material and Methods A 56-year-old man presented with a medical history which included angina (last attack 3 years earlier) and an allergy to penicillin. He had slopped a cardizam regimen I year earlier. His physicians reported no contra-indications to treatment or medication and no restriction on the use of epinephrine in obtaining local anesthesia. The patient's chief complaints were occasional pain in the region of the right maxillary hihlc I I'iv and HTR™ surgical clinical data grafted Mies (mm) I I > H A FDBA mesial fld mesial #7 lingual (Hi HTR mesial #10 mesial e l l lingual ~l> I ' IV op nrob. dep. 6.4 - 5 8.2 1.4 7.(1 K.I I'r. op n\^ dep 2.9 v: 4 . 0 4.5 4.? 4.6 Table 2. Clinical measurements (mini al sites 30 months alter placemen! of r i H i \ and HTR™, before block sections IDCM.ll Sh mesial »7 lingua) ttX HTR mesial " l i t mesial tfl 1 lingual #<•> l'i»t. op proh. dep. 3.3 },8 4.5 3.5 3.2 ».6 Rec. I.I 1 5 I S 0.3 n 1.0 An level change 2.5 2.2 ] 9 3.6 3.7 3.5 premolar. gingival bleeding and difiiciiliy in chewing because of his missing maxillary molars. He reported "swelling and pus" palatal to the maxillary incisors. Although whole-mouth periodontal treatment was provided, only that a! teeth 13. 12. II. 21, 22. 23 is considered here. A fixed splint was in place from tooth 16 to tooth 22 and a single temporary crown on #23. Teeth 14 and 15 had pockets extending to the apices. Although obviousl) hopeless, as they were s\mpiom-tree and in good function, they were retained, symptoms permitting. They were scheduled for removal when all other remaining maxillary teeth were to be extracted. Teeth 12 22 had probing depths of 5 8 mm and responded to light probing with bleeding and exudate. Independent evaluation by 2 members of the Department of Diagnosis and the Department of Prosthodontics concurred with the patient's wish that all of the remaining maxillaiA teeth should be extracted. The patient insisted that "he wants no further problems" and desired a complete maxillary demure fitted. Based on that information, he was referred for possible participation in our evaluation sltidy. The nature of the slud_\ was explained Fig. I. Pre-surgical clinical apparcancc after removal of tooth no. Id (pontic). A fixed splini extends from loolh IK>. 15 to lonth no. 22. A temporary crown is present on tooth no. 23. Hi I B . ^^^^^k ^^^^^B Fig. 2. Flap reflection reveals delects on teeth nos. 21 to 23. Note supraeivstal calculus. Evaluation of HTR polymer 617 Fig 3. Histologic overview of mesial surface of tooth no. 13. ?U months after FDHA placement. Note the sulcul.ii epithelium apical to the remnants of the calculus notch (arrow). Hematoxylin-eosin stain. 35Xmagnification. l-'iy •/. Histological appearance of mesial o( tooth no. 13 apical to [is:. 3 wiih no evidence of new attachment, new oemenlum or graft particles. Hematoxylin-eosin stain. 35Xmagnr6- cation. to the patient, whereb} 2 types of bone replacement materials would he used and the teeth subsequently extracted in block section. 6 to 12 months following surgery. This protocol would change only if symptoms demanded early extraction. The patient \\a> informed that he could withdraw from the study at any lime and signed a consent form. All clinical measurements were made by the author. Scaling and oral hygiene instructions were provided 1 month prior to surgery. The maxillary right Isi molar pontic was severed from the splint at the time of scaling to reduce the ocelusal stress on the right premolar teeth (Fig. I). No attempt was made to remove subgingival calculus. Oral hygiene instructions were repeated weekly until surgery. Measurements I month after scaling and oral hygiene instruction, pre-surgical measurements were recorded for teeth #13 23 (Table I). Dentinal notches were made with a no. V2 round bur at the gingival margin immediately apical to the crown margin on each tooth at the greatest probing depth site of each test tooth. Measurements were made with a Bole\ gauge, siker point and locking pliers to the nearest 0.1 mm (Froum et al. 1982). The distance from the gingival notch to the base of the pocket was recorded, as was the degree of mobility of the splint and individual tooth. Following flap reflection, a 2nd notch was made through the most apical point of calculus at the invoked root site (Froum el al. 1983). The distances from that notch to the bone cresi and from the notch to the deepest point of the osseous defect were recorded. Soft tissue measurements were repeated from the most coronal notch I month before block section. Surgical procedure Under local anesthesia, buccal and palatal full thickness mucoperiosteal flaps were reflected (Fie. 2). Following root/calculus notching, the root and osseous defect were thoroughly debnded. The root was debrided of all visible hard and soft deposits to the base of the osseous defects. Debridemem was completed with ultrasonic and hand instruments. Inlra-marrow penetration was performed at 2 sites (mesial surfaces of teeth nos. 12, 22). Spontaneous bleeding from the walls of the bony crest was evident al all other sites. 3 adjacent sites (I 3, 12. I I > were tilled with FDtJA (crushed cortical bone. 200 350 ftm. University of Miami Tissue Bank. Miami, FL) and the 3 other sites (21. 22. 231 were filled with HTR™ (#40). To minimize the risk of one material migrating into the site of another, it was decided against use of different materials at alternate sides. The 2 bordering sites receiving different materials, tooth I I (lingual) and tooth 21 (lingual) had "contained" circumferential defects from which material migration seemed unlikely. Both materials were prepared with saline solution at least 30 min before implantation. Sues were o\cifilled with graft materials and covered completely by the eoronally positioned and sutured flap. A periodontal dressing (Coe Park. GC America Inc., Chicago. ID was placed. Tetracycline 250 mg q id was prescribed for 2 weeks, after which the sutures were removed. Professional plaque removal was provided weekly for 6 weeks and then at 2-week intervals for 6 618 Frown months. The patient was then unable to retiiin for about 2 years, but reported professional "cleanings" at 3- to 6- month intervals and some compliance in home eare regimen. Sealing was then provided 3x over a 4-week period and appropriate measurements were recorded and radiographs were taken. The maxillary teeth, nos. i 13. I 2. M.2I. 22. 23), were then extracted in block section. All individual specimens were deealeilied and prepared tor histologic examination. Step-serial mesial ly cut sections IS /im) of teeth 13. 12. 22 and 23 and step-serial buccal-lingual sections of teeth 11 and 21 were prepared and stained selectively with hematoxylin- eosin and Mallorv-inehrome stains Results Clinical observations Preoperative probing and bone osseous depths ranged between 6.9 mm and 8.2 mm, and 2.9 mm and 4.6 mm. respectively No mobility of either (he splinted teeth or the single canine was observed at any stage. There was no bleeding on probing recorded before block section removal, and post-surgical healing was uneventful. By 30 months, all sites showed significant pocket reductions ami gain in clinical attachment level (Table 2). Post-surgical probing depths at sites treated with FDBA were between 3.5 to 4.5 mm. and those treated with HTRIM. 3.2 to 3.(> mm. Gain in attachment was between 1.9 and 2.5 mm. and between 3.5 and 3.7 mm. with (•DBA and HTR™ sites, respectively. (imyiv.il shrinkage was between 1.1 and I.S mm at FDBA sites, and between IM and 1.0 at HTR™ sites. Histological Observations Freeze dried bone sllograft Post surgical-shrinkage led to exposure of the gingival and calculus notches in all specimens. Sulcular epithelium was located apical to the calculus notch (Fig. 3). Further apical to that notch, attachment consisted of junctional epithelium and inserted collagen fibers with no evidence of cementogenesis either supra- or sub-crestallj (Fig. 4). Minimal inflammation was presenl in the underlying Fig. 6. Histologic overview of silc of mesial of tooth no. 22. Note bone surrounding the HTRIM panicle ;ii die osseous crc^t and the limited inflammation seen in chc soli tissue opposite the HTR™ particle. Mallory trichrome stain, 35xmagnification, Fig 5 Histological appearance of mesial of tooth no. 23. 30 months after HTR™ graft placement. Note presence of long junctional epithelium at the apical aspect of the calculus notch. Hematoxylin-eosin stain, 35xmagnification. Fig. 7. HILIII magnification ol crestal bone as shown in Fig. 5. Note that the HTRIM particle is surrounded ;U the crcasi with bone that apparently abuts the panicle ul its periphery. Hematoxylin-eosin stain. I2> - magnification. Evaluation nf HTR polymer 619 connective tissue with no evidence of I-DBA particles in any section. Alihough no evidence of active oseogenesis was present, reversal lines were noted along ihc pcriodonlal aspect of the alveolar hone and along the cresl. Hard tissue replacement Synthetic Bone™ alloplast The gingival epithelium faced the coronal notch at all siles and a longjunctiona! epithelium was seen extending to the apical aspect of the calculus notch (Fig. 5). Particles of HTR1M were present, some of which were in close proximity to the crest, and surrounded by a connective tissue capsule with limited crestal remodeling and new bone formation. However, other sections showed particles surrounded by lamellar bone, suggestive of crestal remodeling and new bone formation (Figs. 6. 7|. Minimal gingiuil inflammation vwis seen in all sections, with no evidence of cemenlogenesis opposite [he alveolar crests. Discussion The results described here are based on only 6 sites in a single patient and cannot be considered universal. Both bone replacement materials, t DBA and HTR1A1 Synthetic Bone ,illoplast. resulted in improved clinical parameters: decreased probing depih and new clinical aitachment. Gingival shrinkage occurred at all sites and little gingival inflammation was evident. The patient's presentation with minimal deposits, inflammation and bleeding on probing after a 2-year absence was an indication that he did have professional supportive therapy (as he reported) and did maintain a high level of homecare. However, in view of the pre-extraction scaling and root planing, lhe degree to which bone replacement materials affected the healing responses remains unclear. The observation that neither material led to periodontal regeneration or new attachment, is ai variance with a previous report which documented limited new attachment with HTR™ material in 4 of 11 block sections (Stahl et ul. I WO). The negative response reported here may relate \o the position of the calculus notches which were coronal to lhe existing alveolar crest at the time of implant surgery. Had it been formed, new attachment would have been supraeresial. Such new attachment has been shown to occur to a limited degree only in select cases wilh iliac marrow (Dragoo & Sullivan 1983) and with barrier membranes and coronal anchorage (Stahl & Froum 1991). There was no evidence of any FDBA particles in any section 30 months after implantation. This is of interest in light of a recent publication (Becker et al. 1994) which reported little or no resorption of decalcified Freeze-dried bone 3 to 13 months after insertion in fresh human extraction .sockets. The difference between (heir histological results and those described here may arise from differences in grafted sites, material used (DFDBA, FDBA), the time of hislological examination (13 months versus 30 months), or a combination of ihese factors. HTR™ particles were present in all sections 30 months after surgery and appeared to be well tolerated, wilh little or no surrounding inflammation. Such particles, when surrounded by connective tissue, were in close proximity to the crestal bone position. Other particles were partially or completely surrounded by bone. This suggests that bone grew around them as part of osseous remodeling. The lack of cemenlogenesis indicates that new attachment did not form at any of the experimental sites with either of lhe materials examined. Osleoclasts were nol observed, even 30 months after surgery, and there was no evidence of resorption of HTR™ particles. It is impossible, in the present case, to determine which variables were responsible for the clinical improvements seen following surgery. Several authors (Nyman et al. 1975. Rosling et al. 1975. [\>i-on ci al. 1978, Froum et al. 1982) have reported significant gains in clinical attachment, probing depth reduction and osseous till, following open debridemeni alone. Perhaps (he thorough debridement performed here is responsible for the clinical results. Human histological studies, however, have found liule or no new bone growth with open debridement alone (Stahl et al. 1982). In the present study, new bone growth was evident at sites where HTR particles had been deposited, and remained close to the alveolar bone. Zusammenfassung Histohgische Auswertung von SITR Polymer und gefriergetrockneterH allogenen Knochenimplantat am Menschen. Ein Fallberichl Mil diesern Fallbericht wird beahsicliiiui. Jie Resultate des Einbrmgens von HTR'^' (Hartgewebeersat2 mit synthetischem Knochen ™), sum gieichen Paticnten mil gefriergclrockuelem. allogencn Knochenimplanta! (FDBA) /u vergteichen. An jedem einzdnen eincr Gesamtheit von 6 Zahnen wurdeH 2 Kcrben angebracht. die cine am Gingivalsitum und die andere an dem am weitesteo apikul gelcgcnen Zahnsteinniveau. Retluktion der Sondierungstiefe, Schrumpfung dei Gingiva und klinischer Atlachrnentgewinn wurden uls Reaktion des Wekhgewebes auf die heiden Knocherersalzmaierialien aufgel; il.il und evaluiert. Histologische Schniite dcr 30 Monale nach dem Einbnngen beider Muterialien voigenommenen Btockextraktion, eihrachten keisie Hinweise aufneucs Attachment. Beide Materialien verurMtchten lecine odor nur geringe Entzflndung. Weiterhin vvurden koine I-DBA Partikcl gesehen, \va^ die Annahme tiahelegt. datt die vollstandige Resorption dieses Maleriats vor dem Abtauf dcr 30 Monatc stalll'and. An alien mil FDBA behandclien Stellen. haile das Schrumpl'en der Gingiva die freileaunt: der am Gingivissaum und am apikaleo Zahnsteinniveau gesetzten Kerben /ur Folge. An alien mil HTR™ behandekeu Stellen wurdc direkt neben der Gingivakerbc gelegcnes Gingivaepithel vorgefunden. An den mil HTRrvi behandelien Stellen wurdc weiterhin die Zahnsteinkerbe mit, von Bindegewebe durchsetztem, Saumepithel und adhSsionshindcrndem Knocben abgedeckt. Von Bindegewebe «der Knocben umgebene HTRTVI Panikel waren vorhanden. In engem Kontakt mil dem Alveolarknochen befiudlicfaes HTR™ scheint ills (ieriisl I'iir die Neubildung von Knochen zu dienen. Resume Evaluation histologujue humaine d'albgreffe d'o,\ sichi el congele el d'vn polymere HTR. Un cas clinique Cc cas compare les resultats du placement de HTRrM (os synllietique de remplaccmenl de tissus durs) et dc KDBA (atlogreft'e d'os ie- CIIL' el congele) die/ le memc patient. Deux entailles ont ete effectuees au niveau de six dents, t'unc au niveau de la gencive marglnale et 1'iiutrc au niveau le plus apical du larlre. Les reponses des lissus mous avoc ies deu\ materially eoniprenaienl la reduction dc prolondeur au sondage, ki diminution dc 1'inflammaiion gingivale et le gain d'atlache dinique. De.s coupes histologtques dci sues obtenus trenle inois apres le placement des deux materiaun ne montraient aucun stgne de nouvelie attache, Pcu ou pas d'inSammation etail prescnle avec les deux l>pes dc maleriaux. II y avail egalement une absence de particules I-'DBA. ce qui pourrait indiquer une rcsorpiion complete dece materiel avanl treiUe mois. La diminution de Tocdeme gingival .1 results a I'exposilion des entailles umgivates cl dc sous-lartrc dc tou.s les sites iraites par FDBA. L'epilhelium gingival .i etc trouve adjacent a I'entaille gmgiv.ile de IOUS les sites trailes HTR™. L'enlaillc sous-tarte 620 Frown dcs silcs Lrailes HTR1M £tait recouverle par do I'epithilium dc jonction avec ilu tissn conjonclif" el de I'os centre LL-UL- adhesion. Lcs particules HTR™ etaient presentes et entourfe par du lissii conjonctif ou ile l"os. HTR™ sanble done servir d'echaffaudage a la neoformation osseuse lorsqu'il esi place en contact iimnic avec I'os alveolaire. References Becker. W, Becker, B. E. & Caffesse, R. (1984) A comparison of demineratized freezedricd and aulologous bone to induce bone formation in human exlraction sockets. Journal oj Periodontology 65, [128-1133. Bowers, G. M., ChadrofT, B.. Carnevale, R.. Melonig, J.. Corio. R . Emerson, J . Ste\- ens, J. & Romherg, E. (1989a) Htstoiogic evaluation oi now attachmeni apparatus formation in humans (II). Journal oj Periodontology SQ, 675 6K2. Bowers, (i. M., Chadroff. B., Carnevale. R.. Melonig, .[.. Corio. R.. Emerson. J . Stevens. J. & Romherg. E. (1989a) rtistologk evaluation of new attachment apparatus formation in humans (III). Journal oj I'niodontology60,683- 693. Dragoo, M. R. & Sullivan. M- C. (1478) A clinical and histologies] evaluation ol autogenous iliac bone grafts in humans (11. Wound healing 2 to 8 m o n t h s . Journal <>l Periodontology 44. 599 fi 12 Froum. S. .1. & Gomez, C. (1993) Periodontal regeneration. Current Opinion in Periodontotogyl\ h\,\2&. Froum, S. J., Coran, M., Thaller. B., Scopp, I. W & Stahl. S. S. (1982) Periodontal healing following open debridement Bap procedures. Journal ol Periothmtolo<>\ 53. X 14. Kroum. S. .1.. Kushner. I. & Stahl, S. S. (19831 Mealing responses of human imra osseous lesions following the use of debridement. grafting and citric acid root treatment. 1. Clinical and histologic observations six mouths poslsurgery. Journal of Pcriodontology 54. 67 76. (ianett. S. & Bogle. S. (1994) Peridontal regeneration with bone grafts. Current Opinion in Periodontology IV 168 177, Hancock, b. B. (19X9) Regeneration procedures. Proceedings of the World Work' shop in Cliiucti! Perhiihiuics VI, 1-20. Mellonig, J. T., Bowers. G. M.. Bright. R. W. & Lawrence. J. J. (19761 Clinical evaluation of freeze-dried hone allografts in periodontal osseous detects. Journal oj Periodontology 47. 125 131. Nyman, S., Rosling, B. ct Lindlie. J. (1975) Effect ill professional tooth cleaning on healing alter periodontal surgery. Journal of Clinical Periodontology 2, K((. Poison. A. M. & Hcjl. L. C. (1978) Osseous repair in inlrabony periodontal defects. Journal of Clinical Periodontology 5. 13 23. Sanders. J. J., Sepe, W. W., Bowers, ti- M. et ;il (19X3) Clinical evaluation of Ireezedried bone allografts with and without autogenous bone grafts. Journal of Periodontology 54, 1-8. Rummelhart, .1. M., Mellonig. J. T., Gray, J. L. & Towle, H. J. (1989) A comparison of freeze-dried hone allogral'i and deminerali «d t'ree/e-dned bone allograft in human periodontal osseous delects. Journal of Periodontology 6ft. f>55 663. Sepe. W. W. Bowers. G. M.. Lawrence, J. J., Indelaender, (i. E. & Koch, R. W. (1978) Clinical evaluation of freeze-dried bone allografts in periodontal osseous defects (II). Journal of Periodontology 49. 9 14. Shahmiri, S.. Singh. J. J. & Stahl. S. S. (1992) Clinical response to use of HTRIM polymer implant in human intrabony lesions. International Journal ol Periodontks and Restorative Dentistry 12. 295 300. Stahl, S. S. & Froum, S. J. (1991) Healing of suprabony lesions treated with guided tissue regeneration and coronally anchored Haps. Case report. Journal of Periodontology 18. 69 84. Stahl. S. S., Froum. S. J. & Kuahner. 1. (19X2) Periodontal healing following open debridement Bap procedures. Journal of Periodontology 53. 15 21. Stahl. S. S.. Froum. S. J. & Tarnow. D. (1990) Human clinical ;md histological responses to the placement of HTR polymer particles in 11 intrabony lesions. Journal of Periodontology 53. 6 14. Suziki. J. B.. Babcock-Goodman, S. &. Phillips, B. (19X9) Comparison of clinical healing of human periodontal defects with HTR synthetic grails. Journal of Dental Research 409 (special issue) (abstr. 1 822). Yukna. R. A. (1991) HTR polymer grafts in human periodontal osseous defect: 6 month clinical results. Journal of Perindontology 61. 633 (42. Yukna, R. A. (1994) Clinical evaluation of HTR polymer bone replacement gratis in human mandihular class II molai furcations. Journal of Periodontology 65, 342 3-19. Yukna. R, A. & Greer, Jr. R. (). 11992) Human gingival tissue response to HTR polymer. Journal oj Biomedical Materials Research 26. 517 527. Address: Stuart J. Frown Department oj Surgical Sciences • Periodontks) and Implants New York University Dental Center 345. E 24th Street New York ,V> 10010-4099 ! SA