Growth Factor–Mediated Treatment of Recession Defects: A ... · recession-related treatment with CTGs does not call for barrier membranes; therefore, none were used. In the test

Growth Factor–Mediated Treatmentof Recession Defects: A RandomizedControlled Trial and Histologicand Microcomputed TomographyExaminationMichael K. McGuire,* E. Todd Scheyer,* and Peter Schupbach†

Background: The primary aims of this two-part prospective study were: 1) to compare the safety and ef-ficacy of beta-tricalcium phosphate (b-TCP) + 0.3 mg/ml recombinant human platelet-derived growth fac-tor-BB (rhPDGF-BB) with a bioabsorbable collagen wound-healing dressing and a coronally advanced flap(CAF) to a subepithelial connective tissue graft (CTG) in combination with a CAF in subjects with gingivalrecession defects using a randomized, controlled, split-mouth design; and 2) to compare, through histologicand microcomputed tomography (micro-CT) examination, the periodontal regenerative potential of thesetwo therapies in surgically created gingival recession defects in restoring missing cementum, periodontalligament (PDL), and supporting alveolar bone.

Methods: In the randomized controlled trial (RCT), 30 patients with Miller Class II buccal gingival recession,‡3 mm deep and ‡3 mm wide in contralateral quadrants of the same jaw were treated and followed for 6 months.Using a split-mouth design with similar bilateral recession defects, test sites were treated with 0.3 mg/ml rhPDGF-BB + b-TCP + bioabsorbable collagen wound-healing dressing; contralateral control sites were treated with a CTG,each in combination with a CAF. In the histologic/micro-CT study segment, recession defects were created in sixteeth,each requiringextraction fororthodontic therapy.Thesedefectswerecreatedwitha recessiondepth ‡3mm,theosseouscrest2 to3mmapical to thegingivalmargin,andwith2 to3mmofkeratinized tissue.Thedefectsweretreated with a CTG (control) or rhPDGF-BB + b-TCP + wound-healing dressing (test), plus CAF. Nine months aftersurgical correction, en bloc resections were obtained and examined histologically and with micro-CT.

Results: In the RCT, test and control treatments demonstrated clinically significant improvements frombaseline through month 6. Statistically significant results favoring the CTG were found in recession depthreduction (-2.9 + 0.5 mm, test; -3.3 + 0.6 mm, control; P = 0.009), root coverage (90.8%, test; 98.6%, con-trol; P = 0.013), and -3.9 – 0.7 mm, control, -3.3 – 1.3 mm, test, recession width reduction (P = 0.035),whereas mid-buccal probing depth (PD) and PD reduction (PDR) reduction favored the test group (1.4 –0.4 mm, test; 1.8 – 0.1 mm, control; P < 0.001 PD and -0.0 mm test; +0.4 mm control PDR). For all otherparameters, the two treatments were statistically equivalent, including increases in keratinized tissue, es-thetic results, and subject satisfaction. In the histologic/micro-CT portion, all four sites treated with rhPDGF-BB + b-TCP showed evidence of regeneration of cementum, PDL with inserting connective tissue fibers, andsupporting alveolar bone, whereas neither CTG-treated site exhibited any signs of periodontal regeneration.

Conclusions: CTG and rhPDGF-BB + b-TCP + wound-healing dressing are effective treatment modalitiesfor clinically correcting gingival recession defects. In addition, the current study demonstrated that regen-eration of the periodontium in gingival recession defects was possible through a growth factor–mediated ap-proach. J Periodontol 2009;80:550-564.

KEY WORDS

Connective tissue; gingival recession/surgery; grafts; histology; recombinant platelet-derived growthfactor-BB; rhPDGF-BB.

* Private practice, Houston, TX.† Department of Periodontics, School of Dental Medicine, University of Pennsylvania, Philadelphia, PA.

doi: 10.1902/jop.2009.080502

Volume 80 • Number 4

550

The successful resolution of patient-centeredconcerns related to gingival recession defects,i.e., improved esthetics, reduction or elimina-

tion of root dentinal sensitivity, decreased incidenceof cervical caries, and improved ability to controlplaque is critical to effective treatment outcomes whenaddressing recession-related mucogingival deficits. Anadditional, although perhaps more elusive, therapeuticgoal is the regeneration of the lost attachment appa-ratus common to all recession defects, including theformation of new cementum with inserting connectivetissue fibers and supporting alveolar bone.1 Two recentsystematic reviews2,3 of randomized controlled clinicaltrials verified the effectiveness of the subepithelialconnective tissue graft (CTG), guided tissue regener-ation (GTR), and the coronally advanced flap (CAF) inproducing statistically significant gains in root coverageand clinical attachment. In both reviews, the subepithe-lial CTG seemed to be more effective than GTR inresolving or reducing gingival recession. Additionalstudies4,5 examining the long-term results of the CTGseem to support its long-term efficacy in maintainingroot coverage. Supported by such evidence-based data,the CTG is viewed by many clinicians as the goldstandard treatment in reducing or eliminating gingivalrecession. Less certain, however, is its ability to satisfythe other major goal of recession therapy: regenerationof the lost attachment apparatus, including the forma-tion of new cementum with inserting connective tissuefibers and supporting alveolar bone.

A number of studies suggested that the CTG has atleast limited potential to regenerate tissues of the peri-odontium. In a case study examining a tooth removeden bloc following root fracture, Harris6 demonstratedregeneration coronal to the original gingival margin.Likewise, Goldstein et al.7 demonstrated regenerationof new bone, cementum, and periodontal ligament(PDL) 15 months after CTG with a CAF in a maxillaryfirst bicuspid with 5 mm of recession. Bruno andBowers8 also found 0.5 to 1 mm of regenerated bone,cementum, and PDL just coronal to the alveolar crest,with connective tissue covering the majority of the re-maining exposed root surface.

Although some evidence for regeneration exists,other studies suggested that attachment through a longjunctional epithelium (LJE) or connective tissue adhe-sion is more likely to result following connective tissuegrafting. In a study examining shallow recessions usinga double-pedicled graft with connective tissue, Harris9

demonstrated an LJE extending close to the originalbony crest or connective tissue fibers running parallelto the root surface. In a histologic study examining enblocresectionsofmaxillarypremolarsfollowingconnec-tive tissue grafting, Majzoub et al.10 also found predom-inantly an LJE adjacent to exposed dentin along withevidence of parallel-oriented connective tissue fibers.

In a third histologic study, Cummings et al.11 comparedtheCTGtoacellulardermalmatrixand reportedfindingsfor each to be a combination of an LJE and connectivetissue, without evidence of periodontal regeneration.

Although considered by many as the most effectivetreatment for root coverage, the CTG has a numberof disadvantages: a distant donor site is required,increased morbidity may be associated with graftharvest, and limited amounts of donor tissue may in-crease the total treatment time for patients with mul-tiple recession defects. Considering these limitations,as well as the variable results seen in regeneratingtissues of the attachment apparatus, recombinantgrowth factor technology may offer viable alternativesto the CTG, including regeneration of cementum,PDL, and supporting alveolar bone.

Since the late 1980s, when research first revealedthat platelet-derived growth factor (PDGF) promotedregeneration of cementum, inserting connective tis-sue fibers and bone, >100 studies have underscoredPDGF’s positive effects on chemotaxis and the prolif-eration of animal and human PDL and alveolar bonecells.12-14 Several recent human histologic stud-ies15-17 dramatically revealed PDGF’s ability to sup-port periodontal regeneration in severe periodontalinterproximal and Class II furcation defects.

Basedonthepositiveclinicalandhistologicfindingsoftheprecedingstudies,McGuireandScheyer,18 inasplit-mouth case series, recently evaluated multiple clinicalparameters following the treatmentofgingival recessiondefects with beta-tricalcium phosphate (b-TCP) and acollagenmembrane, both saturated with recombinanthuman PDGF-BB (rhPDGF-BB). Comparing the re-sults achieved after 6 months with PDGF to thoseachieved with the subepithelial CTG, both proce-duresachievedrootcoveragewith£1mmofresidualre-cession depth. Similar improvements were seen inprobing attachment levels and height of keratinized tis-sue for the experimental treatment group and the CTGgroup.Giventhe favorabletissueresponsestorhPDGF-BB with b-TCP and a collagen membrane, and resultscomparable to theCTG,aproperlypowered,controlledclinical trial was deemed warranted.

Based upon the above initial feasibility study find-ings as well as earlier histologic regenerative findingsin restoring the complete attachment apparatus withan rhPDGF-BB–mediated approach for intrabonyand Class II furcationdefects, a single-center, two-partprospective study was undertaken with the followingobjectives: 1) to compare, in subjects with recession-type defects, the safety and efficacy of b-TCP +0.3 mg/ml rhPDGF-BB with a bioabsorbable collagenwound-healing dressing‡ and a CAF to a subepithelialCTG in combination with a CAF using a randomized,

‡ CollaTape, Integra LifeSciences, Carlsbad, CA.

J Periodontol • April 2009 McGuire, Scheyer, Schupbach

551

controlled, split-mouth design; and 2) to compare theperiodontal regenerative potential of the abovetwo therapies in surgically created gingival reces-sion defects in restoring missing cementum, PDL, andsupporting alveolar bone through histologic and micro-computed tomography (micro-CT) examinations.

MATERIALS AND METHODS

Part I: Randomized Controlled Trial (RCT)Study population. Thirty patients (four males and26 females) with Miller Class II19 buccal gingival re-cession ‡3 mm deep and ‡3 mm wide in contralateralquadrants of the same jaw who met the inclusion/exclusion criteria were eligible to enter the study. In-clusion criteria included the following: patients musthave read, understood, and signed an informed con-sent form; been able and willing to follow study proce-dures and instructions; had at least one pair of teethwith buccal recession type defects ‡3 mm deep and‡3 mm wide (selected defects were located in contra-lateral quadrants of the same jaw, with similar size andmorphology with respect to depth and width, and se-lected teeth subjected to root canal treatment wereasymptomatic and without technical remarks; theroot canal treatment should have been completed atleast 6 months before inclusion); had presence of ker-atinized tissue; and in case of adjacent teeth with re-cession defects, only one tooth at each side actedas a test or control tooth; no adjacent teeth weregrafted during surgeries. Exclusion criteria includedthe following: subjects who were participating in otherclinical trials; had a history of cancer or human immu-nodeficiency virus; had an allergy to yeast- derivedproducts; failed to maintain good plaque control;had Class V restorations; had teeth with extremelyprominent root surfaces; had any systemic acute in-fections in the areas intended for surgery; had a his-tory within the previous 6 months of weekly or morefrequent use of smokeless chewing tobacco, pipe orcigar smoking and cigarette smoking; had takenchronic therapeutic doses of medication known to af-fect bone metabolism; and female subjects who werepregnant or lactating or sexually active females whowere of childbearing potential and were not using hor-monal or barrier methods of birth control. The follow-ing teeth did not qualify: molars; teeth with axialmobility; teeth with interproximal loss of attachment;and teeth with crowns or veneers. All patients wereenrolled from the authors’ (MKM and ETS) privatepractice from June to October 2006. The patient pop-ulation included subjects 18 to 70 years of age (meanage: 43.8 – 10.7 years). A signed institutional reviewboard–approved consent form was obtained fromeach patient. Of the 30 patients, one was of Hispanicorigin, and the rest were white. All of the patients werecurrent non-smokers; 14 had smoked 11.0 – 11.4 cig-

arettes/day for a mean of 6.6 – 6.6 years. No pa-tient smoked in the 6 months prior to study entry.All treated defects required marginal gingivalkeratinized tissue. At baseline, keratinized tissuemea-sured 1.9 – 0.6 mm for the rhPDGF-BB group and 1.9 –0.8 mm for the subepithelial CTG group. All occlusalinterferences were removed through occlusal adjust-ment, and biteguards were constructed as needed.The split-mouth design of this study allowed eachpatient to serve as his or her own control.

Clinical evaluation. Atbaselineandateachpost-sur-gical follow-upvisit, the treatedsiteswereexaminedclin-ically, photographic documentation was performed, anddefect measurements were recorded. Following initialbaseline screening and surgery, six follow-up visitsthrough week 24 were mandated for this study. Radio-graphs were taken at baseline and at week 24. The pri-mary efficacy endpoint was the change in recessiondepth at 6 months. Secondary endpoints included clin-ical attachment level (CAL), probing depth (PD), PD re-duction (PDR), height of keratinized tissue, percentageof root coverage, recession width, clinician rating ofcolor/texture of treatment sites, the subject’s estheticsatisfaction, and subject perception of pain/discomfort.

Baseline parameters included the following: gingi-val recession depth, recession width measured at thelevel of the cemento-enamel junction (CEJ), height ofkeratinized tissue from the free gingival margin to themucogingival margin, buccal and proximal PDs, CALmeasured from the CEJ, alveolar bone level, root den-tin hypersensitivity, inflammation score, and plaquescore for both test teeth. At baseline, there were nostatistical differences between the teeth allocated tothe two study treatments with regard to recessiondepth (P = 0.349), CAL (P = 0.672), mid-buccal PD(P = 0.681), proximal PD of the mesial surface (P =0.464), proximal PD of the distal surface (P = 0.804),height of keratinized tissue (P = 1.000), or recessionwidth (P = 0.922).

Evaluation of healing, based on the presence orabsence of inflammation using a visual analog scale,with ‘‘much worse than expected’’ on one end and‘‘much better than expected’’ on the other end, wasmade at weeks 1, 2, and 4.

Root dentin hypersensitivity was assessed using aconventional blast of air for 3 seconds at the exposedroot surface. The hypersensitivity was recorded asnone, mild, moderate, or severe.

Visible signs of inflammation were examined andrecorded based on the modified gingival index ofLamster et al.20

Plaque scores of test and control teeth were calcu-latedbasedonthepresence(1)orabsence(2)ofplaqueon the buccal and lingual surfaces of the study teeth.

Gingival color and texture were assessed by compar-ing test and control teeth to surrounding tissues and

Growth Factor–Mediated Treatment of Recession Defects Volume 80 • Number 4

552

scoring throughquestionnaires, i.e.,morered, lessred,or equally red and more firm, less firm, or equally firm.

The patient’s perception of pain was assessed us-ing a visual analog scale, with ‘‘no pain’’ and ‘‘extremepain’’ on each end, for graft and donor sites.

A visual analog scale measuring the patient’s es-thetic satisfaction, with ‘‘disappointed’’ at one endand ‘‘fully satisfied’’ at the other end, was completedat baseline and week 24.

All pre- and postoperative clinical assessmentswere made by a single, masked examiner who wasnot the operating surgeon. Training and calibrationwere conducted prior to study initiation to ensure in-traexaminer reproducibility of measured outcomes.

Surgical procedure. All subjects received oral hy-giene instructions and were not scheduled for surgeryuntil they were capable of demonstrating adequatesupragingival plaque control. Immediately prior tosurgery, the treating surgeon (or delegated dentalassistant) opened an envelope with the subject’s iden-tifying number and treatment assignment. The subep-ithelial CTG and rhPDGF-BB + b-TCP treatments wereassigned to the right or left side of the subject’s mouthaccording to a predetermined randomization sched-ule. Subjects were not informed about which treat-ment each study tooth had received. The examiningclinician remained masked to the randomization codefor the duration of the study.

Test and control sites were treated as described re-cently by McGuire and Scheyer,18 with the followingexception: a bioabsorbable collagen wound-healingdressing§ saturated with rhPDGF-BB was placed overthe grafted test root surfaces in place of a collagenbarrier membrane. The first surgery was performedon the left side in all subjects. The second surgerywas performed immediately after the first surgery.

Statisticalmethods.Descriptive statistics were per-formed by treatment and provided the summaries forcontinuous variables displayed as sample size (n) andmean and its SE, SD, median, and range and for cate-gorical variables displayed as count and percentage.

One-way analysis of variance (ANOVA) was usedto compare the difference between treatments forthe continuous measurements. The Fisher exact testwas used to compare the difference between treat-ments for the categoric parameters.

For exploratory purposes, analysis of covariance(ANCOVA) was also used for the primary and second-ary variables (CAL, PDR, height of the keratinized tis-sue, and recession width); the treatment comparisonwas adjusted for the baseline value, site (left/right),and sequence of treatments.

All statistical tests were two-sided.Sample size determination. Power calculations at

the 5% significance level determined that 20 evaluablesubjects were needed to detect a 1.0-mm change in

recession depth, with 95% power and assuming awithin-subject variation (SD, estimated from previousstudies with similar inclusion/exclusion criteria) of 1.0mm. This study elected to recruit 30 subjects withbuccal recession–type defects. All 30 subjects com-pleted the study.

Part II: Histologic/Micro-CT ExaminationStudy population. Two patients were enrolled in thisportion of the study. Each required first premolarextraction for orthodontic treatment. One patient re-quired the removal of all four first premolars, whereasonly the maxillary first premolars required extractionin the second patient. Therefore, six teeth were sched-uled for removal. Gingival recession defects weresurgically created buccal to all six of these teeth.Following a thorough explanation of the nature ofthis portionof the study, which included information re-garding the potential risks as well as an understandingof the lack of any clinical benefit derived fromrecession-related procedures, informed consents weresigned, and institutional review board approval was ob-tained. For participating in this study, orthodontictreatment was provided without cost to the patientswho otherwise could not afford orthodontic care.

Basic study design. Surgically created recessiondefects were designed with the following characteris-tics: recession depth ‡3 mm, osseous crest 2 to 3 mmapical to the apically positioned gingival margin, anda maximum of 2 to 3 mm of keratinized marginal gin-giva ( Fig. 1). In an effort to replicate a naturally con-taminated root surface, the surgically created defectswere left exposed to the oral environment for 2months. Four of the six defects were grafted withrhPDGF-BB + b-TCP + a collagen wound-healingdressing, and two were grafted with CTGs. As in theRCT portion of the study, CAFs were used to coverthe grafted sites (Figs. 2 and 3). Discrete referencenotches were placed at the pregrafted free gingivalmargin and at the osseous crest of all sites (Fig. 4).In each patient, one tooth, as determined by a ran-domization schedule, served as the control and re-ceived a CTG. All other premolars received therhPDGF-BB–mediated therapy. Nine months aftercorrection of the surgically created recession defects,conservative en bloc resections were obtained for his-tologic and micro-CT examination. The techniqueused preserved as much ridge as possible to minimizethe chance for a ridge defect following the en bloc re-section. The following technique required precise in-cisions into soft tissue, followed by cuts into hardtissue. A full-thickness incision was made from eachof the line angles of the buccal free gingival marginto the center of the adjacent papillae. Then a verticalincision was made interproximally through the

§ CollaTape, Integra LifeSciences.


553

gingiva, extending into the mucosa. Next, the mesialand distal incisions were connected with an incisionparallel to and below the mucogingival junction.The surgeon, using a piezosurgical instrument, cutthe tooth from the facial to the lingual aspects, break-ing the mesial and distal contacts. As soon as the con-tact was broken, the cut was angled through the tooth

toward the central fossa, with the mesial and distalcuts joining at that point. These cuts were made ap-proximately to the depth of the CEJ. A separate cutwas made through the cortical plate and the buccalportion of the root 2 to 3 mm below the pregraftedosseous crest following the full-thickness soft tissueincisions. This cut was connected to the initial inci-sions with an oblique cut, and the buccal portion ofthe tooth, including the graft,was removed for analysis.

Figure 4.Reference notches are seen at the preoperative gingival margin andalveolar crest in the surgically created defects prior to correction of therecession defect with rhPDGF-BB + b-TCP or a CTG with CAF.GN = gingival notch; ON = osseous notch.

Figure 1.Surgically created gingival recession defect with osseous crest 2to 3 mm apical to the newly created gingival margin.

Figure 2.Gingival recession corrected 2 months after defect creation withrhPDGF-BB + b-TCP or CTG, each with a CAF.

Figure 3.Nine months after correction of a surgically created recession defect.


554

The remaining part of the tooth was extracted in a rou-tine fashion. The resulting alveolar defects were im-mediately grafted with freeze-dried bone allograft +rhPDGF-BB. Three months later, active orthodontictherapy recommenced uneventfully. No postopera-tive complications were observed in any of the testor control sites.

Micro-CT analysis. The specimens were scannedusing a high-resolution micro-CT systemi in multislicemode. Each image data set consisted of ;800 micro-CT slice images.

The specimens were scanned in high-resolutionmode with an x-, y-, and z-resolution of 16 mm. The im-age data sets were used to produce three-dimensional(3D) views of the specimens using a special software.¶

The following parameters were examined usingmicro-CT analysis: two-dimensional and 3D anatomyof preexisting bone and newly formed bone adjacentto the root surfaces, the 3D configuration of the PDLspace, and micro-CT corroboration of histologic find-ings of periodontal regeneration by identifying newbone connected by a PDL-like space in relation tothe gingival margin and osseous reference notches.

Operative procedure for surgically created re-cession defects. Following the administration of localanesthetic, gingivectomies were performed where re-quired, resulting in 2 to 3 mm of keratinized marginaltissue adjacent to each test and control tooth. To allowelevation of a full-thickness mucoperiosteal flap,mesial and distal vertical releasing incisions were con-nected to a carefully placed intrasulcular incision.Next, where necessary, ostectomies to remove thebuccal cortical plate were performed, enabling the os-seous crest to be repositioned 2 to 3 mm apical to thenewly positioned gingival margin (Fig. 1). Then themucoperiosteal flap was apically repositioned, result-ing in a recession defect ‡3 mm as required by thestudy design. 5.0 gut sutures were used to close allincisions (Fig. 5).

RESULTS

Part I: RCTTest and control treatments proved to be equallyeffective and were comparable in terms of patient sat-isfaction with postoperative discomfort, healing, andthe final esthetic result (Fig. 6). On a 10-cm visual an-alog scale, no statistically significant differences wereobserved between the treatments in response to an es-thetic satisfaction questionnaire. The subjects ratedpostoperative discomfort, which included bleeding,swelling, and sensitivity, as similar for the two treat-ment sites. All subjects had mild or no discomfortdue to bleeding, swelling, and sensitivity and con-tinued to improve from weeks 1 through 4. No sta-tistically significant difference in pain scores wasobserved between the study treatments using a 10-

cm visual analog scale. At the 24-week postoperativevisit, 97% of the subjects commented that they expe-rienced no difference in discomfort between the twotreatment sites. Likewise, no statistically significantdifferences in the clinical rating of color/texture ofthe tissues were observed between the treatmentsthroughout the study. At week 24, 100% of treatmentsites were rated by the masked examiner as equallyfirm compared to surrounding tissues. One hundredpercent of sites treated with rhPDGF-BB were scoredas equally red, whereas 90% of CTG sites were ratedas equally red compared to surrounding tissues.

Tables 1 and 2 summarize the primary and second-ary efficacy endpoints. Test and control treatmentsresulted in clinically significant improvements frombaseline at the 24-week postoperative visit for allmeasured parameters. For the primary endpoint,change in recession depth at 6 months, treatment withrhPDGF-BB + b-TCP resulted in a mean recessiondepth reduction of -2.9 – 0.5 mm, whereas CTG treat-ment caused a mean reduction of -3.3 – 0.6 mm(ANOVA, P = 0.009; ANCOVA, P = 0.007); this statis-tically significant difference favored the CTG-treatedsites. Likewise, although both treatments resultedin significant clinical improvements from baseline,the reduction in recession width and the percentageof root coverage statistically favored the CTG. At24 weeks, a mean reduction in recession width of

Figure 5.Surgically created defect with flap apically repositioned to create arecession defect ‡3 mm .

i mCT 40, Scanco Medical, Bassersdorf, Switzerland.¶ Scanco Medical.


555

-3.3 – 1.3 mm and -3.9 – 0.7 mm was seen for the testand control groups, respectively (ANOVA, P = 0.035;ANCOVA, P = 0.021). The mean percentage of rootcoverage at 24 weeks was 90.8% and 98.6% for thetest and control groups, respectively (ANOVA, P =0.013).

As noted in Table 1, there were no statistically sig-nificant differences in the height of keratinized tissueat the 24-week postoperative visit between the studygroups’ treatments (rhPDGF-BB + b-TCP, 2.9 – 0.6mm; CTG, 3.2 – 0.8 mm; P = 0.113). In addition,both treatments resulted in statistically equivalent ab-solute changes in keratinized tissue from baseline toweek 24 (ANOVA, P = 0.115; ANCOVA, P = 0.072).

Table 2 summarizes the changes in CAL and PD re-ductions from baseline to week 24 in the test and con-trol groups. CAL was measured, to the nearest 0.5mm, from the CEJ, the fixed reference point. Therewere no statistically significant differences in CAL ob-served between the study treatments at the 24-weekpostoperative visit (rhPDGF-BB + b-TCP, 1.7 – 0.5mm; CTG, 1.8 – 0.4 mm; P = 0.335) or for the absolutechange from baseline to the 24-week postoperativevisit (rhPDGF-BB + b-TCP, -2.9 – 0.6 mm; CTG,-2.9 – 1.0 mm; ANOVA, P = 0.871; ANCOVA, P =0.385). PD was measured to the nearest 0.5 mm fromthe gingival margin to the bottom of the probablepocket of the buccal, mesial, or distal surface. There

were no statistically significant differences observedbetween the study treatments for proximal PD of themesial or distal surface at the 24-week postoperativevisit or for PDR from baseline to the 24-week post-operative visit. However, there was a statistically sig-nificant difference between the study treatments forPD of the mid-buccal surface at the 24-week postop-erative visit, favoring the test treatment (rhPDGF-BB +b-TCP, 1.4 – 0.4 mm; CTG, 1.8 – 0.1 mm; P <0.001)and for the PDR from baseline to the 24-week postop-erative visit (rhPDGF-BB + b-TCP, 0.0 – 0.6 mm;CTG, 0.4 – 0.6 mm; ANOVA, P = 0.004; ANCOVA,P <0.001).

With regard to the safety results, 25 (78.1%) sub-jects experienced 75 adverse events (AEs) duringthe study. All AEs were mild or moderate in severity.The most common AE was mild contusion, occurringin 16 (50.0%) subjects, followed by face swelling in 13(40.6%) subjects. There were no serious AEs duringthe study.

Part II: Histologic/Micro-CT ExaminationNine months after treatment with a CTG or rhPDGF-BB + b-TCP + a bioabsorbable collagen wound-heal-ing dressing, all six sites retained 100% root coverage.Both therapeutic approaches restored the protectivesoft tissue morphology and esthetic balance of themucogingival complex. However, clear and distinct

Figure 6.Representative control (A and C) and test (B and D) pre- (left panel) and postoperative (right panel) sites at baseline and 6 months later.


556

differences were found in each treatment’s ability torestore the lost attachment apparatus common toall recession defects. The following sections examine,in detail, two representative sites from the study (theother sites will be the focus of a companion article):one grafted with a CTG and the other with rhPDGF-BB + b-TCP + a collagen wound-healing dressing,each with a CAF. Both grafted areas are from awell-developed, well-nourished 41-year-old Hispanicmale with bimaxillary protrusion associated with se-vere maxillary and moderate mandibular incisor

crowding. The patient required removal of all four firstpremolars. One site, tooth #21, was grafted with a CTG.The other three sites were grafted with rhPDGF-BB +b-TCP.CTG results: tooth #21. Figure 7 illustrates the surgi-cal timeline of this study. Two months after creation ofthe recession defect (Figs. 7A and 7B), osseous andgingival margin reference notches were placed atthe buccal osseous crest and precorrected gingivalmargin and the area grafted with a CTG from the pal-ate. Then a buccal flap was coronally advanced supe-rior to the CTG; the outcome at 9 months is seen inFigure 7C. A conservative en bloc resection was

Table 1.

Recession Depth, Recession Width,Percentage of Root Coverage, andHeight of Keratinized Tissue at Baselineand Week 24

Parameter Baseline Week 24

Absolute Change From

Baseline to Week 24

Recession depth (mm)Test

Mean (SE) 3.2 (0.1) 0.3 (0.1) -2.9 (0.1)Range 3 to 5 0 to 2 -4 to -2

ControlMean (SE) 3.4 (0.1) 0.1 (0.0) -3.3 (0.1)Range 3 to 5 0 to 1 -5 to -3

P value 0.349* 0.015* 0.009*; 0.007†

Recession width (mm)Test

Mean (SE) 4.0 (0.1) 0.7 (0.2) -3.3 (0.2)Range 3 to 6 0 to 3 -6 to 0

ControlMean (SE) 4.0 (0.1) 0.1 (0.1) -3.9 (0.1)Range 3 to 6 0 to 2 -6 to -2

P value 0.922* 0.021* 0.035*; 0.021†

Root coverage (%)Test

Mean (SE) 90.8 (2.8)Range 57 to 100

ControlMean (SE) 98.6 (1.0)Range 75 to 100

P value 0.013*

Height of keratinized tissue (mm)Test

Mean (SE) 1.9 (0.1) 2.9 (0.1) 1.0 (0.1)Range 1 to 3 2 to 5 0 to 3

ControlMean (SE) 1.9 (0.1) 3.2 (0.1) 1.3 (0.1)Range 1 to 4 2 to 5 0 to 3

P value 1.000* 0.113* 0.115*; 0.072†

* One-way ANOVA.† ANCOVA model adjusted for recession depth, recession width, and height

of keratinized tissue at baseline, side (left/right), and sequence oftreatments.

Table 2.

CAL, Mid-Buccal Depth Reduction, andProximal PDs at Baseline and Week 24

Parameter Baseline Week 24

Absolute Change From

Baseline to Week 24

CAL (mm)Test

Mean 4.6 (0.1) 1.7 (0.1) -2.9 (0.1)Range 4 to 7 1 to 3 -4 to -2

ControlMean 4.7 (0.2) 1.8 (0.1) -2.9 (0.2)Range 4 to 7 1 to 3 -6 to -2

P value 0.672* 0.335* 0.871*; 0.385†

Mid-buccal depth reduction (mm)Test

Mean 1.4 (0.1) 1.4 (0.1) 0.0 (0.1)Range 1 to 2 1 to 2 -1 to 1

ControlMean 1.4 (0.1) 1.8 (0.1) 0.4 (0.1)Range 1 to 2 1 to 3 -1 to 1

P value 0.681* <0.001* 0.004*; <0.001†

Proximal PD (mesial; mm)Test

Mean 2.3 (0.1) 1.9 (0.1) -0.4 (0.1)Range 1 to 3 1 to 3 -1 to 1

ControlMean 2.2 (0.1) 1.9(0.1) -0.3 (0.1)Range 1 to 3 1 to 3 -1 to 1

P value 0.464* 0.871* 0.568*; 0.962†

Proximal PD (distal; mm)Test

Mean 2.3 (0.1) 1.9 (0.1) -0.4 (0.1)Range 1 to 3 1 to 3 -2 to 1

ControlMean 2.4 (0.1) 2.0 (0.0) -0.4 (0.1)Range 2 to 3 2 to 3 -2 to 0

P value 0.804* 0.190* 0.720*; 0.211†

* One-way ANOVA.† ANCOVA model adjusted for CAL, mid-buccal depth reduction, and

mesial/distal proximal PD at baseline, side (left/right), and sequence oftreatments.


557

performed at 9 months (Fig. 7D) for microscopicground section and micro-CT examination.

Nine months after CTG, no evidence of periodontalregeneration was evident. Figure 8A is a micro-CT im-age of the en bloc resection obtained at 9 months fol-lowing CTG correction of the recession defect. Thebottom arrow points to the osseous reference notch,and the top arrow denotes the gingival margin refer-ence notch. Neither notch is obscured with overlyingregenerated bone. As noted in Figures 8A and 8B, nochange in the level of the osseous crest occurred dur-ing the 9-month period after CTG grafting. Ground-section histologic examination at 9 months (Fig.8C) confirmed the micro-CT findings. An LJE extends

apically, ending just superior to the original osseouscrest. Abundant connective tissue without evidenceof bone or cementum regeneration is seen coronalto the osseous crest. A bucco-lingual micro-CT viewemphasizes the complete lack of bone regenerationseen in the en bloc resected specimen 9 months afterCTG + CAF treatment (Fig. 9A). This lack of bone re-generation was also demonstrated in an additionalbucco-lingual micro-CT image through the centralportion of the tooth; no bone regeneration is seen cor-onal to the osseous notch (Fig. 9B).

rhPDGF-BB + b-TCP graft results: tooth #12. Asin the above CTG specimen, Figure 10 represents theidentical surgical timeline, except that this site was

Figure 7.A) Clinical photograph of gingival recession defect 2 months after its surgical creation. Osseous and gingival reference notches were placed at thetime of surgical correction of the recession defect. B) In each case, the osseous crest was placed ;3 mm apical to the precorrected gingival margin.C) 100% root coverage is maintained 9 months after correction of the recession defect with the CTG. D) Conservative en bloc resection of tooth#21 along with adjacent hard and soft tissues.

Figure 8.A) A micro-CT image of the en bloc resection obtained 9 months after CTG correction of the recession defect. The bottom arrow represents theosseous reference notch (ON), and the top arrow denotes the gingival margin reference notch (GN). No evidence of bone regeneration is evident9 months after connective tissue grafting. B) Osseous and gingival margin reference notches were placed at the time of surgical correction of therecession defect. C) Histologic ground section confirms no evidence of periodontal regeneration 9 months after connective tissue grafting.Box: An LJE is seen extending just coronal to the original osseous crest along with abundant non-inserting connective tissue (CT) fibers.OC = osseous crest; RC = root canal. (Toluidine blue/Azur II; original magnification: ·12).


558

grafted with rhPDGF-BB + b-TCP + a collagen wound-healing dressing with an overlying CAF. As in the for-mer grafted site, the osseous crest was placed ;3 mmapical to the precorrected gingival margin.

Nine months after grafting with rhPDGF-BB +b-TCP, micro-CT revealed substantial bone regener-ation beginning at the osseous notch and extendingcoronally toward the gingival reference notch (Fig.11). The regenerated tissue appears as dense corticalbone analogous to the lamina dura observed in dentalperiapical radiographs.

Further identification of the osseous and gingivalreferencenotches,critical todetermining thepresence

and extent of periodontal regeneration, is seen in addi-tional sagittal micro-CT imagery (Fig. 12). Boneregeneration coronal to the osseous notch is readilyappreciated in this micro-CT image.

Ground mineralized histologic sections, used in thisstudy for increased clarity over demineralized serialsections, confirmed and amplified the micro-CT re-sults, demonstrating robust coronal bone and cemen-tum regeneration with a uniformly dimensioned PDLspace 9 months after grafting with rhPDGF-BB +b-TCP (Fig. 13). The osseous reference notch is welldefined in the sagittally positioned ground section (Fig.13B). Note the partial occlusion of the osseous notch

by newly regenerated cementum.Residual b-TCP particles seen inboth ground sections appear tobe inhibiting the more robust boneregeneration seen in areas wherethe ceramic has already resorbed.

To better demonstrate and de-fine the absolute position of eachreference notch, multiple horizon-tal-CT sections were taken, with18 mm between cuts, beginningat the most apical extent of the im-aged en bloc resection. Micro-CTcuts 339 and 527 correspond tothe locations of the osseous andgingival reference notches, re-spectively, representing an inter-notch distance of 188 micro-CThorizontal sections or ;3.4 mm(Figs. 12, 14, and 15).

Under higher magnificationof the ground section in Figure13A, significant coronal bone

Figure 9.A) Bucco-lingual micro-CT image reveals no evidence of bone regeneration coronal to theosseous reference notch 9 months after correction of the recession defect with a CTG andCAF. B) Bucco-lingual micro-CT image through the central portion of tooth #21 demonstratesno bone regeneration coronal to the osseous notch (ON) at the tooth surface. A small amount ofpreexisting interproximal bone is seen above the osseous notch entry point (ONE) secondary toslight rotation of the image. OC = osseous crest; RC = root canal; GN = gingival notch.

Figure 10.A) Clinical photograph of gingival recession defect 2 months after its surgical creation. B) Osseous and gingival reference notches placed at thetime of surgical correction of the recession defect. In each case the osseous crest was placed ;3 mm apical to the precorrected gingival margin.C) One hundred percent root coverage is maintained 9 months after correction of the recession defect with rhPDGF-BB + b-TCP. D) Conservativeen bloc resection of tooth #12 along with adjacent hard and soft tissues.


559

regeneration is seen (Fig. 16). In addition, regenera-tion of cellular cementum as well as a uniform PDLspace between the newly formed bone and the adja-cent tooth surface are seen.

At still higher magnification, all three tissues of thenormal periodontium, i.e., cementum, PDL, and sup-porting alveolar bone, are seen in greater detail (Figs.17 and 18). Osteocytes and cementocytes are clearlyseen in newly formed bone and cementum, respec-tively. It is significant that connective tissue fibers ofthe newly regenerated PDL insert directly into newlyformed cementum and bone. Polarized light imagery

confirms the ground section findingthat connective tissue fibers insertdirectly into newly regenerated ad-jacent tissue (Fig. 19).

DISCUSSION

To examine therapeutic effective-ness when addressing gingival re-cession defects, researchers andclinicians must determine whetherany given therapy successfully ad-dresses the major components ofthese complex lesions. The follow-ing are common to all recessiondefects: loss of the protective func-tional morphology of the muco-gingival complex, an estheticimbalance among marginal tis-sues and the adjacent tooth root

and crown, and anatomic loss of portions of the at-tachment apparatus, i.e., cementum, PDL, and sup-porting alveolar bone.1 Comprehensive treatmentsuccessfully addresses each of these components.The current study, divided into two sections, exam-ined and compared the safety and efficacy of theCTG to a novel tissue-engineered approach on eachof the major components of recession defects.

Multiple studies,2-5 including two systematic re-views of randomized, controlled clinical trials, verifiedthe long-term efficacy of the CTG with a CAF in main-taining root coverage. However, unlike past studies,the current trial compared the CTG to a therapeuticapproach using rhPDGF-BB + b-TCP + a collagenwound-healing dressing, each covered with a CAF.In both therapies, clinically significant improvementsfrom baseline were observed at the 24-week postop-erative visit in recession depth reduction; CAL gain;PDRs at buccal, mesial, or distal surfaces; increasesin the height of keratinized tissue; root coverage;and recession width reduction. There were no statisti-cally significant differences observed between the twotreatments for all efficacy endpoints measured, ex-cept for the reductions in recession depth, recessionwidth, and root coverage, each favoring the subepi-thelial CTG treatment; PDR at the mid-buccal surfacestatistically favored the rhPDGF-BB + b-TCP treat-ment. No statistically significant differences wereseen between the two treatments in esthetically re-lated endpoints, i.e., color/texture of the tissues, asdetermined by the esthetic satisfaction questionnaire.In terms of overall satisfaction, no differences wereseen between the two treatment modalities. At theconclusion of the study, a number of patients requiredadditional surgery to correct recession defects adja-cent to other teeth. Each patient requiring addi-tional surgery was given a choice between CTG or

Figure 11.Nine months after grafting with rhPDGF-BB + b-TCP, sagittal (A) and coronal (C) micro-CT revealcoronal bone regeneration superior to the osseous notch (B). The regenerated tissue appears asdense cortical bone analogous to the lamina dura observed in dental periapical radiographs.ON = osseous notch; NB = new bone; GN = gingival notch; RC = root canal.

Figure 12.Osseous and gingival reference notches are clearly seen in thissagittal micro-CT image as well as bone regeneration coronal to theosseous notch. The distance between notches is ;3.4 mm.ON = osseous notch; GN = gingival notch.


560

rhPDGF-BB–mediated treatment; all chose therhPDGF-BB + b-TCP treatment. Two main reasonswere given: the subjective esthetic results were equiv-alent, and the treatment with the growth factor did notrequire an additional harvesting procedure.

Although the RCT portion of this study examinedefficacy parameters related to restoring the protective

functional morphology of the mu-cogingival complex as well as theesthetic balance between margi-nal tissues and the adjacent toothroot and crown, a histologic exam-ination was required to assess theability of each therapeutic modal-ity to effectively regenerate miss-ing portions of the attachmentapparatus. To compare the peri-odontal regenerative potential ofeach therapy, recession defectswith well-defined anatomic charac-teristics were surgically created buc-cal to six teeth planned forextraction secondary to orthodon-tic treatment.

Unlike traditional GTR proce-dures, barrier membranes werenot used in this study. Acceptedrecession-related treatment withCTGs does not call for barriermembranes; therefore, none wereused. In the test group, in lieu ofa barrier membrane, a porous col-lagen wound-healing dressing#

saturated with rhPDGF-BB was placed over theb-TCP + rhPDGF-BB grafts. The collagen dressinghelped to prevent b-TCP particle migration and pro-vided an additional source for the release of rhPDGF-BB into the surrounding environment. Unliketraditional GTR procedures, which exclude un-wanted cell migration into an underlying grafted site,growth factor–mediated treatment chemotacticallyseeks out mesenchymal stem cells, preosteoblasts,and fully committed osteoblasts that are found inlarge numbers within the overlying periosteum.21-23

Unlike barrier membranes, which would preventthe active migration of these latter cells into thegrafted site from the overlying periosteum, a porouswound-healing dressing likely permits such entranceinto the grafted area through directed growth factor–mediated cell migration.

Nine months after surgical correction of the defectswith a CTG or rhPDGF-BB + b-TCP, conservative enbloc resections were performed, and each specimenwas studied histologically and through micro-CT ex-amination. Clear and significant differences were seenin each treatment’s ability to regenerate new cemen-tum with insertingconnective tissuefibers and support-ing alveolar bone.

Two of six sites were treated with CTGs, and neitherexhibited signs of periodontal regeneration. In neithersite was there any evidence of cementum, PDL, or

Figure 13.A) At low power, a ground section demonstrates robust coronal bone regeneration 9 monthsafter grafting with rhPDGF-BB + b-TCP. Note the uniform PDL space between newly formedbone (NB) and the adjacent tooth surface as well as the formation of new cementum (C).Coronally, residual TCP particles appear to inhibit the amount of bone regeneration seen elsewherein the specimen. OB = old bone. B) Newly formed cementum (NC) is seen occluding the orificeof the osseous notch (ON) in this higher power sagittal ground section. New bone (NB) andcementum have formed coronal to the osseous notch. (Toluidine blue/Azur II; originalmagnification: A, ·7; B, ·20.)

Figure 14.Horizontal micro-CT image at level 339 clearly demonstrating thelocation of the osseous reference notch (ON; arrow).

# CollaTape, Integra LifeSciences.


561

supporting bone regeneration. In both instances, anLJE extended just coronal to the original osseous crest.In addition, although an abundant amount of connec-tive tissue formed, collagen fibers generally ran parallel

to theadjacent root surfaces,withnoevidenceof insert-ing Sharpey’s fibers.

In contrast, the representative site, as well as theother three sites grafted with 0.3 mg/ml rhPDGF-BB +b-TCP, exhibited definitive signs of periodontal regen-eration, i.e., formation of new cementum, PDL, andsupporting bone. In the present example, multiple his-tologic ground sections demonstrated significant newbone growth coronal to the osseous reference notch.Robust regeneration of cellular cementum occurred

Figure 15.Identification of the gingival reference notch (GN; arrow) at horizontalmicro-CT image level 527 represents a distance ;3.4 mm from theosseous notch.

Figure 16.Higher magnification reveals significant new bone (NB) formationfollowing grafting with rhPDGF-BB + b-TCP. Also note the well-developed PDL space and the regenerated area of newly formedcementum (C). (Toluidine blue/Azur II; original magnification: ·18.)

Figure 17.High-power ground section view of tooth #12 reveals regenerationof all tissues of the missing periodontium. New cementum is highlycellular. Cementocytes and osteocytes are seen in regeneratedcementum and bone, respectively. OC = old cementum; D = dentin;NCC = new cellular cementum; WB = woven bone. (Toluidineblue/Azur II; original magnification: ·45.)

Figure 18.At higher magnification, cementum, PDL, and alveolar bone areseen in greater detail. Osteocytes (OC) and cementocytes (CC) areclearly seen in newly formed bone and cementum, respectively.Connective tissue fibers of the newly regenerated PDL are seeninserting directly into newly formed cementum and bone. NC = newcementum; ICTF = inserting connective tissue fibers; WB = wovenbone. (Toluidine blue/Azur II; original magnification: ·75.)


562

coronal to the osseous notch in this and the other threesites grafted with rhPDGF-BB + b-TCP. High-powermagnification under polarized and non-polarized lightdemonstratedPDLcollagenfibers insertingdirectly intonewly regenerated cementum and bone.

In addition to light microscopy, micro-CT was usedto examine and infer the type of attachment obtainedfollowing CTG or growth factor–mediated treatmentof these surgically created recession defects. Recently,Nevins et al.24 used micro-CT technology to evaluateperiodontal regeneration in the treatment of intrabonydefects. Using micro-CT, the investigators were able toevaluate a number of parameters, including the two-and 3D anatomy of preexisting and newly regeneratedbone, the thickness and 3D shape of the PDL space,and the suggestion of regeneration as defined by themicro-CT findings of adjacent new bone connectedby a PDL space in relation to a reference root notch.Likewise in this study, micro-CT evaluation confirmedinformation gained from histologic ground sections

and yielded unambiguous information confirming thepresence or absence of bone regeneration coronal tothe osseous reference notch. Based on micro-CT find-ingsalone,graftingwith rhPDGF-BB + b-TCP led to sig-nificant coronal bone regeneration, whereas graftingwith CTGs led to none. In addition, micro-CT evalua-tion suggested that at 9 months postgrafting, thegrowth factor–mediated bone appeared as dense cor-tical bone, similar to that seen as the lamina dura in peri-apical dental radiographs. Finally, residual particles ofb-TCP inferior to the newly regenerated bone appearedto diminish the robust bone formation seen in areaswhere the ceramic had already resorbed (Fig. 13). Ac-cording to Ridgway et al.,17 residual particles of b-TCPmay delay PDGF-mediated bone formation.

CONCLUSIONS

The CTG and growth factor–mediated treatment mo-dalities led to clinically effective outcomes for thecorrection of gingival recession defects. Althoughchanges in recession depth, width, and percentageof root coverage statistically favored the CTG at 6months, the subjects’ esthetic satisfaction was thesame for both treatments. The tissue-engineeredapproach had the added advantage of requiring noharvesting procedure. Histologically, none of theCTG-treated sites yielded evidence of periodontalregeneration, whereas the combination of rhPDGF-BB + b-TCP + wound-healing dressing consistentlyled to the formation of new cementum with insertingconnective tissue fibers and supporting alveolar bonein all four treated defects. To our knowledge, this is thefirst published study to document human histologicevidence of regeneration in the treatment of recessiondefects using recombinant growth factor technology.

In this study, surgically created recession defectswere used to examine histologic and micro-CT re-sults. Exposed root surfaces were allowed time to be-come contaminated to simulate naturally occurringgingival recession defects. Although difficult becauseof the need to obtain en bloc resections, additional his-tologic studies using long-standing recession defectswould add to the body of knowledge and further definethe regenerative potential of growth factor–mediatedtreatment. Finally, additional studies examining car-riers for rhPDGF-BB, other than b-TCP, would provideadded insight into the potential effectiveness of tissue-engineered solutions for the treatment of gingival re-cession defects.

ACKNOWLEDGMENTS

The authors acknowledge the following individuals fortheir help in this project: Carol Waring, registered den-tal hygienist (RDH) and study coordinator, RebeccaGarcia, RDH and director of clinical research, CindyWainscott, certified dental assistant and office

Figure 19.Polarized light confirms that connective tissue fibers insert directly intonewly regenerated cementum (A) and bone (B). ICTF = insertingconnective tissue fibers; NC = new cementum; OC = old cementum;D = dentin; NB = new bone. (Toluidine blue/Azur II; originalmagnification: ·90.)


563

manager, and Dr. Jacquelyn Campbell, editor, PerioHealth Professionals, Houston, Texas, for assistancewith conducting the study and developing the manu-script; Jennifer Gramazio, marketing assistant, Osteo-health, Shirley, New York, for her assistance with theimages; Dr. Myron Nevins, private practice in peri-odontics, Swampscott, Massachusetts, for his assis-tance with the histology; Dr. Marc Nevins, privatepractice in periodontics, Boston, Massachusetts, forhis guidance with the block sections; Mr. Leigh Ren,statistician, Target Health, New York, New York, forassistance with the statistical analysis; Dr. PalleHolmstrup, Department of Periodontology, RoyalDental College, Copenhagen, Denmark, for his histo-logic review; and special thanks to Dr. Stuart Kay, sci-ence writer and consultant, Huntington, New York, forhis help with the organization and production of themanuscript. This study was supported by a grant fromOsteohealth, a division of Luitpold Pharmaceuticals,Shirley, New York. The rhPDGF-BB and b-TCP usedin this study are marketed and sold by Osteohealth.Drs. McGuire and Scheyer have received lecture feesfrom Osteohealth. Dr. Schupbach reports no financialrelationships related to any products or commercialfirms involved in this study.

REFERENCES1. Trombelli L. Periodontal regeneration in gingival re-

cession defects. Periodontol 2000 1999;19:138-150.2. Roccuzzo M, Bunino M, Needleman I, Sanz M. Peri-

odontal plastic surgery for treatment of localized gin-gival recessions: A systematic review. J Clin Periodontol2002;29(Suppl. 3):178-194.

3. Oates TW, Robinson M, Gunsolley JC. Surgical ther-apies for the treatment of gingival recession. A sys-tematic review. Ann Periodontol 2003;8:303-320.

4. Harris RJ. Root coverage with connective tissue grafts:An evaluation of short- and long-term results. J Peri-odontol 2002;73:1054-1059.

5. Rossberg M, Eickholz P, Raetzke P, Ratka-Kruger P.Long-term results of root coverage with connectivetissue in the envelope technique: A report of 20 cases.Int J Periodontics Restorative Dent 2008;28:19-27.

6. Harris RJ. Successful root coverage: A human histo-logic evaluation of a case. Int J Periodontics Restora-tive Dent 1999;19:439-447.

7. Goldstein M, Boyan BD, Cochran DL, Schwartz Z.Human histology of new attachment after root cover-age using subepithelial connective tissue graft. J ClinPeriodontol 2001;28:657-662.

8. Bruno JF, Bowers GM. Histology of a human biopsysection following the placement of a subepithelialconnective tissue graft. Int J Periodontics RestorativeDent 2000;20:225-231.

9. Harris RJ. Human histologic evaluation of root cover-age obtained with a connective tissue with partialthickness double pedicle graft. A case report. J Peri-odontol 1999;70:813-821.

10. Majzoub Z, Landi L, Grusovin MG, Cordioli G. Histol-ogy of connective tissue graft. A case report. J Peri-odontol 2001;72:1607-1615.

11. Cummings LC, Kaldahl WB, Allen EP. Histologicevaluation of autogenous connective tissue and acel-lular dermal matrix grafts in humans. J Periodontol2005;76:178-186.

12. Lynch SE, Williams RC, Polson AM, et al. A combina-tion of platelet-derived growth factor and insulin-likegrowth factor enhances periodontal regeneration. J ClinPeriodontol 1989;16:545-554.

13. Matsuda N, Lin WL, Kumar MI, Cho MI, Genco RJ.Mitogenic, chemotactic and synthetic responses of ratperiodontal ligament fibroblastic cells to polypeptidegrowth factors in vitro. J Periodontol 1992;63:515-525.

14. Nevins M, Giannobile WV, McGuire MK, et al. Platelet-derived growth factor stimulates bone fill and rate ofattachment level gain: Results of a large multicenterrandomized controlled trial. J Periodontol 2005;76:2205-2215.

15. Camelo M, Nevins ML, Schenk RK, Lynch SE, Nevins M.Periodontal regeneration in human Class II furcationsusing purified recombinant human platelet-derivedgrowth factor-BB (rhPDGF-BB) with bone allograft. IntJ Periodontics Restorative Dent 2003;23:213-225.

16. Nevins M, Camelo M, Nevins ML, Schenk RK, LynchSE. Periodontal regeneration in humans using re-combinant human platelet-derived growth factor-BB(rhPDGF-BB) and allogenic bone. J Periodontol 2003;74:1282-1292.

17. Ridgway HK, Mellonig JT, Cochran DL. Human histologicand clinical evaluation of recombinant human platelet-derived growth factor and beta-tricalcium phosphate forthe treatment of periodontal intraosseous defects. Int JPeriodontics Restorative Dent 2008;28:171-179.

18. McGuire MK, Scheyer ET. Comparison of recombinanthuman platelet-derived growth factor-BB plus betatricalcium phosphate and a collagen membrane tosubepithelial connective tissue grafting for the treat-ment of recession defects: A case series. Int J Peri-odontics Restorative Dent 2006;26:127-133.

19. Miller PD. A classification of marginal tissue recession.Int J Periodontics Restorative Dent 1985;5:8-13.

20. Lamster IB, Alfano MC, Seiger MC, Gordon JM. Theeffect of Listerine antiseptic on reduction of existingplaque and gingivitis. Clin Prev Dent 1983;5:12-16.

21. Malizos KN, Papatheodorou LK. The healing potentialof the periosteum: Molecular aspects. Injury 2005;36(Suppl. 3):S13-S19.

22. Emans PJ, Surtel DA, Frings EJ, Bulstra SK, Kuijer R.In vivo generation of cartilage from periosteum. TissueEng 2005;11:369-377.

23. Stevens MM, Marini RP, Schaefer D, Aronson J,Langer R, Shastri VP. In vivo engineering of organs:The bone bioreactor. Proc Natl Acad Sci USA 2005;102:11450-11455.

24. Nevins ML, Camelo M, Rebaudi A, Lynch SE, NevinsM. Three-dimensional micro-computed tomographicevaluation of periodontal regeneration: A humanreport of intrabony defects treated with Bio-Oss col-lagen. Int J Periodontics Restorative Dent 2005;25:365-373.

Correspondence: Dr. Michael K. McGuire, 3400 S. Gessner,Suite 102, Houston, TX 77063. Fax: 713/952-0614; e-mail:[email protected].

Submitted October 2, 2008; accepted for publicationDecember 5, 2008.


564

Documents

Growth Factor–Mediated Treatment of Recession Defects: A ... · recession-related treatment with CTGs does not call for barrier membranes; therefore, none were used. In the test